Notes and Questions on the Copyright Statute

1.  Copyright is a complex and highly conceptual field of law.  Fortunately, all of its basic principles are explicit or implicit in the copyright statute itself.  Indeed, the current statute as originally enacted—the Copyright Act of 1976 or "1976 Act"—is undoubtedly the best drafted federal statute in the field of intellectual property.  (Not so for many recent amendments!)  The statute's quality is not surprising, for the Copyright Office and Congress spent over twenty years preparing the text of the 1976 Act.  Rapid changes in technology and the marketplace, the greater importance of intellectual property generally, and changes in the nature of congressional deliberations make such leisurely and thoughtful consideration unlikely in the future.

For the copyright novice, the most important provisions of the 1976 Act are the definitions in Section 101, the statement of the scope of copyright protection in Section 102, and the list of the copyright owner's exclusive rights in Section 106.  The definitions section is hard to grasp in the abstract, but skim it quickly to see what kinds of definitions it contains.  Then read Sections 102 and 106 twice each, once for sense and a second time to identify key words, phrases and concepts.

2.  Section 102(a) has two parts, a first sentence containing general criteria for copyright protection, followed by a list of categories of copyrighted works.  The word "include" in the preamble makes clear that the list is nonexclusive, nonexhaustive and illustrative only.   Therefore the "meat" of Section 102(a) is in the first sentence, which lays out the two basic requirements for copyright protection.  What are they?

When and how does copyright protection arise?  Can you answer this question from the first sentence of Section 102(a) and the definition of "created" in Section 101?

3.  Subsection (b) of Section 102 is just as important as subsection (a).  Section 102(b) codifies a judge-made exclusion from copyright protection for ideas, processes, methods, etc.  The exclusion derives from a long series of decisions beginning with the seminal case of Baker v. Selden, 101 U.S. 99, 25 L.Ed. 841 (1880).  The exclusion goes by the general name of "idea/expression dichotomy," under which copyright does not protect abstract ideas or facts, but only the form in which they are expressed.

The Baker Court explained the rationale for this dichotomy as follows:

"A treatise on the composition and use of medicines, be they old or new; on the construction and use of ploughs, or watches, or churns; or on the mixture and application of colors for painting or dyeing; or on the mode of drawing lines to produce the effect of perspective,—would be the subject of copyright; but no one would contend that the copyright of the treatise would give the exclusive right to the art or manufacture described therein.  The copyright of the book, if not pirated from other works, would be valid without regard to the novelty, or want of novelty, of its subject-matter. The novelty of the art or thing described or explained has nothing to do with the validity of the copyright.  To give to the author of the book an exclusive property in the art described therein, when no examination of its novelty has ever been officially made, would be a surprise and a fraud upon the public.   That is the province of letters-patent, not of copyright.  The claim to an invention or discovery of an art or manufacture must be subjected to the examination of the Patent Office before an exclusive right therein can be obtained; and it can only be secured by a patent from the government.

"The difference between the two things, letters-patent and copyright, may be illustrated by reference to the subjects just enumerated.  Take the case of medicines.  Certain mixtures are found to be of great value in the healing art.  If the discoverer writes and publishes a book on the subject (as regular physicians generally do), he gains no exclusive right to the manufacture and sale of the medicine; he gives that to the [*103] public.  If he desires to acquire such exclusive right, he must obtain a patent for the mixture as a new art, manufacture, or composition of matter.  He may copyright his book, if he pleases; but that only secures to him the exclusive right of printing and publishing his book.  So of all other inventions or discoveries."

101 U.S. at 102-103.

What is the policy rationale behind this distinction between patents and copyrights?  Would a copyright on a book provide a "cheap patent" if it protected the ideas discussed in the book as well as the expression?  Why?  (This same principle applies to ideas, methods and processes embodied in computer programs.  See Computer Associates International, Inc. v. Altai, Inc., 982 F.2d 693, 703-712, 23 U.S.P.Q.2d (BNA) 1241 (2d Cir. 1992) (reviewing precedent, applying dichotomy to computer-program cases, and proposing abstraction-filtration-comparison test for applying it)).

The law is equally clear that the idea/expression dichotomy applies to unpatentable ideas, such as historical facts, speculations and arguments, as well as patentable inventions.  See, e.g., Miller v. Universal City Studios, Inc., 650 F.2d 1365, 1367-1368, 1372 (5th Cir. 1981) (reporter's research on story of notorious kidnapping); Hoehling v. Universal City Studios, Inc., 618 F.2d 972, 978-79 (2d Cir. 1980) (facts and speculation regarding circumstances of destruction of "Hindenburg" dirigible).  See also, Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 340, 344-345, 111 S.Ct. 1282, 1287, 113 L.Ed.2d 358 (1991) ("That there can be no valid copyright in facts is universally understood.  The most fundamental axiom of copyright law is that [*345] ‘no author may copyright his ideas or the facts he narrates'"), quoting Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 556, 105 S.Ct. 2218, 85 L.Ed.2d 588 (1985).

What is the policy rationale for this rule?  What fundamental principle of our free society might be adversely affected if authors could control the publication by others of facts and ideas by writing books embodying them and copyrighting the books?

4.  Although facts themselves may not be copyrighted, original collections of facts (or other materials, whether copyrighted or in the public domain) may be copyrighted as "compilations."   Most databases are "compilations."

As the Supreme Court said in Feist : "This case concerns the interaction of two well-established propositions.  The first is that facts are not copyrightable; the other, that compilations of facts generally are.  Each of these propositions possesses an impeccable pedigree."  499 U.S. at 344.

Take a look at the definition of "compilation" in Section 101 and the rules for compilations and derivative works in Section 103.  What aspects of a compilation does copyright protect, if not the individual facts in it?  What standard do those aspects have to meet?  What words in the statute suggest the answers?

5.  Read again the first sentence of Section 102(a), together with the definitions of "copies," "fixed," and "phonorecords" in Section 101.  Do all these provisions have something in common?  Are they designed to make copyright protection depend upon the technology or medium used to record creative works, or are they designed to make it independent of technology and medium?  What words in each provision answer this question?  Why did Congress draft the statute this way?

6.  In the previous copyright statute, the Copyright Act of 1909, Congress had used the phrase "all the writings of an author," rather than the current language "original works of authorship" now in Section 102(a).   See Pub. L. No. 60-349, § 4, 35 Stat. 1075, 1076 (1909), codified at 17 U.S.C. § 4 (1970) ("The works for which copyright may be secured under this title include all the writings of an author.").  Congress intended the change in language to create a potential gap between the range of works covered by the Constitution, namely "Writings" of "Authors," and the scope of statutory copyright protection.  As the House Report explained, Congress wanted to avoid giving courts "the alternative of holding copyrightable something that Congress clearly did not intend to protect, or of holding constitutionally incapable of copyright something that Congress might one day want to protect."  H.R. Rep. No. 1476, 94th Cong., 2d Sess. 51 (Sept. 3, 1976), reprinted in 1976 U.S. Code Cong. & Admin. News 5659, 5664.  Can you think of any kind of work that might fall into this "gap?"  Should the answer depend upon the nature of the work or the technology used to record it?

7.  What do the definitions of "copies" and "phonorecords" have in common?  Are they similar in substance?  How do they differ?

This book and many on copyright often use the words "copy" or "copies" imprecisely and in a general sense to refer both to copies and phonorecords.  The alternative would be cumbersome: referring repeatedly to "copies and phonorecords" and using seven syllables when two would do.  The statutory regime for phonorecords and sound recordings differs from that for copies in certain respects, particularly with regard to exclusive rights, exceptions to those rights, and remedies.  The basic requirements for copyright protection, however, are the same for both copies and phonorecords.

8.  Read again the statutory definition of "fixed" and the first sentence of Section 102(a).  Now consider an ordinary paper book.  Suppose a machine-readable file contains the text of the book as a "text file,"(1) which appears on the screen as the text of the book but is stored in the computer as a binary file using so-called "ASCII code."(2)  Whenever the file is accessed by the computer user, the machine's software or operating system converts the binary file, consisting entirely of large blocks of ones and zeroes, into readable text on the screen.  Is the binary file still a book?  Is it protected by copyright to the same extent as the book in paper form?  What words in the definition and the first sentence of Section 102(a) provide an answer?  Is the answer clear?

9.  Computer systems have two general types of storage subsystems.(3)  First, there are so-called "permanent" or long-term storage devices, such as floppy disks, hard disks, tapes, CD-ROMS (compact disk read-only memories), DVDs (digital versatile disks), and writeable versions of the last two.  All of these are designed to retain storage when power is lost or the computer is shut down.  Second, there is temporary memory, including computers' working memory, often called "random-access memory" or "RAM,"(4) which is "volatile" in the sense that its contents are lost when power is lost or the computer is shut down.

Consider again the book in Note 8, supra.  Are the images of its pages on the display screen "fixed"?  Is the ASCII text file "fixed" while in RAM?

It is now well established that data in RAM is "fixed" and constitutes a "copy" or "phonorecord."  See, e.g., Triad Systems Corp. v. Southeastern Express Co., 64 F.3d 1330, 1333, 1335, 36 U.S.P.Q.2d (BNA) 1028 (9th Cir. 1995), cert. denied 516 U.S. 1145 (1996) (citing MAI Systems, infra); NLFC, Inc. v. Devcom Mid-America, 45 F.3d 231, 234, 33 U.S.P.Q.2d (BNA) 1629 (7th Cir. 1995) (point not disputed); MAI Systems Corp. v. Peak Computer, Inc., 991 F.2d 511, 519 (9th Cir. 1993), cert. dismissed 510 U.S. 1033 (1994) (seminal case).  Looking at the definition of "fixed" and focusing on its key words, can you articulate a rationale for this result?  Does the same rationale apply to the display-screen image?

10.  Now take a closer look at Section 106.  It lists the "bundle of rights" that inhere in every copyright.  Notice how Section 501(a) refers to Section 106 (as well as the numerous statutory exceptions) to define infringement.

Is Section 106 generally exhaustive, or are there other rights not listed in Section 106.  Compare the introductory language of Section 106 to that of Section 102(a) and Section 107.  Does the comparison suggest an answer?  (Section 602 does provide certain additional rights against infringing imports, but those rights are explicit.  The question here is whether any rights should be deemed implicit in the statute, other than those listed in Section 106.)

Suppose someone gives you a pirated (unlawful) copy of a book?  Is your mere possession of that unlawful copy, without more, actionable under Section 106?  Is your mere receipt of the copy actionable without more?  Do the answers depend upon your state of mind, i.e., whether you know the copy was pirated?  Should they?  If you are not liable for direct or primary infringement, might you be liable for inducing, assisting, or contributing to infringement by the person who made the pirated copy? under what circumstances?

11.  As the edited version of the statute suggests, copyright law is relatively simple in general concept, but it is often devilishly hard to apply.  Most, if not all, of the statute's textual and conceptual difficulty lies in the numerous exceptions in Sections 107 through 122—all of which (except for "fair use" under Section 107) are omitted from the edited version in these materials.

Section 107 is perhaps the most general exception.  Another general exception is the so-called "first sale" doctrine under Section 109.  That exception allows the owner of an authorized copy to resell it, lend it (except for computer programs), or otherwise dispose of that single copy, or to display it publicly (for example, in a gallery) without transmitting its images to other locations.

Beyond these, the various exceptions are complex, detailed and specific.  Indeed, many are specific to particular industries, such as the music industry or the satellite/cable industry.  Those that deal with such modern technologies are models of prolixity and obscurity.  Even an introductory survey course in copyright does not normally cover them in any systematic or comprehensive way.  They best way to deal with them as a student is to look over their titles and be aware of their existence general subject matter, so that you can consult them as relevant problems arise.

12.  As mentioned earlier in out unit on software and the First Amendment, the computer software amendments of 1980 made clear that copyright protects computer programs.  The actual statutory changes wrought by those amendments, however, were minimal.  The 1980 amendments changed only two things: (1) they added the definition of "computer programs" (which appears out of alphabetical order at the end of Section 101) and (2) they amended Section 117.  (Before the 1980 amendments, Section 117 was simply a "standstill" provision, saying that copyright law as applied to computer programs just as it would have before the 1976 Act became law—whatever that may have meant.  Immediately after the 1980 amendments, Section 117 contained what is now Section 117(a) and (b) is slightly different format.)

What is now Section 117(a) and (b) were part of the 1980 enactment.  Sections 117(c) and (d) were added by amendment in 1998.  See 17 U.S.C. § 117(c), (d), as added by Title III, "Computer Maintenance Competition Assurance Act," of Digital Millennium Copyright Act, Pub. L. No. 105-304, § 302(3), 112 Stat. 2860, 2887 (Oct. 28, 1998).

The purposes of subsections (a) and (b) are self-evident.  Section (a) allows owners of (legitimate) copies of computer programs to make backup and archival copies and to make, modify and adapt copies for use in machines.  Subsection (b) allows owners of hardware on which software is legitimately stored to transfer or dispose of the hardware along with the backup and archival copies (including any stored inside the machine, for example, on a hard disk) but not to transfer adaptations.  Even these provisions, however, are not without difficulty. Compare Vault Corp. v. Quaid Software Ltd., 847 F.2d 255, 260, 266 (5th Cir. 1988) (interpreting Section 117(a)(1) as allowing making of archival copies "for any reason so long as the owner uses the copy for archival purposes only and not for an unauthorized transfer") (footnote omitted), with Micro-Sparc, Inc. v. Amtype Corp., 592 F. Supp. 33, 35 (D.Mass.1984) (owner could make archival copies only to protect against mechanical or electrical failure, to which paper copies of source code were not subject), and Atari, Inc. v. JS & A Group, Inc., 597 F. Supp. 5, 9-10 (N.D.Ill.1983) (to similar effect).

The purpose of subsections (c) and (d) is a bit more obscure.  They derive from the case of MAI Systems Corp. v. Peak Computer, Inc., 991 F.2d 511, 518 (9th Cir. 1993), cert. dismissed 510 U.S. 1033 (1994).  There a third-party servicer, at the request of the owner of a computer and licensee of its operating system, turned on the computer in order to service it, causing a copy of the operating system to be made automatically in RAM during the "boot up" process.  The Ninth Circuit held this was copyright infringement because the servicer, unlike the computer owner, had not been licensed to use the operating system.  Congress adopted subsections (c) and (d) in order to overturn this result without affecting the more general principle that copies in RAM are "fixed" and therefore are "copies" within the meaning of the 1976 Act.

1.   Such a file has many different but loosely equivalent names.  These include: "text file," "ASCII file," and "DOS text file."

2.   ASCII code is a conventional, standardized numerical code used to store written characters in computers, including letters, numbers and punctuation.  "ASCII" is an:

"[a]cronym for the American Standard Code for Information Interchange.  Pronounced ask-ee, [it] is a code for representing English characters as numbers, with each letter assigned a number from 0 to 127.  For example, the ASCII code for uppercase M is 77.  Most computers use ASCII codes to represent text, which makes it possible to transfer data from one computer to another.

"Text files stored in ASCII format are sometimes called ASCII files."

3.   In the early days of the computer era, input and output devices such as printers and magnetic storage units (disks or tapes) were called "peripheral devices" or "peripherals."  The reason was that they were usually made, sold and installed separately from the computer itself (including the "CPU," or central processing unit, and main memory).  Often so-called "peripheral" devices were made by different manufacturers than those that made the CPU and main memory.  Indeed, early computers did not even have some of the "peripherals" that we take for granted today, such as displays and keyboards.  Instead, an early computer might have had two rows of lights for immediate output, displaying in binary form (off for "zero" and on for "one") the machine-language instruction currently in the machine's instruction register and the last instruction executed.  A similar row of switches was often used to place instructions manually into the computer's central processing unit in binary form, for example, to begin the "boot up" process.

Today the term "peripheral devices" is inappropriate for two reasons.  First, many if not all of these devices can be and often are packaged, installed, and sold as part of the computer and by the same manufacturer.  Second, even when they are added latter and/or bought from a different manufacturer, both the manufacturer and consumer usually have a choice: to install the device "internally" as a circuit board or "externally" in a separate box with cables connecting to the computer.  Under these circumstances it makes no sense to describe these devices as "peripheral," and the term has fallen into disuse.

4.   The term "random" here is not used in the lay sense of "arbitrary" or "uncontrolled."  Rather, it is used to distinguish "random-access" storage from a different type of storage using "sequential access" methods.

Magnetic tape, for example, is said to have "sequential access" because a person wanting to read something in the middle of the tape must play or spool the tape sequentially until she comes to the part with the desired data.   In contrast, a computer's main memory (usually embodied in semiconductor chips) allows every memory cell or location to be accessed separately and in any order.  Thus, the access is "random" in the sense that one can access any stored datum "at random" without spooling or paging through unwanted data.

Disk-based magnetic storage units (both floppy and hard) have a little bit of both random and sequential access.  Physical subunits of the disks called "sectors" contain the data.  The computer can address each sector separately, just like a cell of main memory. Yet within each sector the data are stored sequentially, and the computer can read or write to the sector only as a whole.  This is the reason why very short files consisting of only a few characters or bytes often appear longer when stored on disk.  An entire sector—usually consisting of a round number of bytes in binary terms, such as 512 or 1024—must be used to store even the shortest file.  Unused portions of the sector are simply "wasted" insofar as availability for further storage is concerned.