next section in this report >> |   previous section >> |   report contents >>


Sections 2.0 and 3.0 described various laws that potentially govern pre-1972 sound recordings. But how does one navigate these laws to determine whether to proceed with a particular activity in connection with a particular sound recording?

In any such determination, one begins by evaluating whether the proposed activity is legally objectionable. In other words, is there any possible claim, and by whom? A necessary corollary of that inquiry is the relevance of copyright exceptions, and more broadly, the special role of libraries. Then, assuming there is a possible claim, one must consider the likelihood that a right holder would assert it.

Section 4.1 explores the examples set forth in the Introduction to determine which laws are applicable to each type of sound recording, and who conceivably could assert a claim. Section 4.2 examines the relevance of library status and copyright exceptions in determining whether library Internet streaming activities would come within the scope of potential claims. Section 4.3 considers the relevance of library status and copyright exceptions in preservation activities. Section 4.4 provides questions to ask in evaluating permissible use of sound recordings, including questions that focus on whether a right holder is likely to assert a claim.

4.1 Applying the Law to Specific Examples

Prior to looking at each of the sound recording examples in the context of the laws described in sections 2.0 and 3.0, it is useful to determine whether copying or streaming those recordings could potentially be the basis of a legal claim. This section looks principally at the characteristics of the particular sound recording and the proposed use to see if it falls within the terms of a statute, or, in the case of common law, whether anyone has brought an analogous claim. The special considerations that relate to library use are discussed in 4.2 and 4.3.

In each of the examples, the proposed library activity is the same:
(1) Copying the recording in digital form to enable streaming
(2) Streaming the recordings, i.e., transmitting or performing them, without providing the user with a retention copy
(3) Copying the recording in digital form for preservation purposes

Again, it is assumed that the library will not charge a fee for access to the recordings, or derive any direct or indirect commercial advantage (which may include fundraising) for streaming.144 See the further discussion of this assumption in section 4.2.

4.1.1 Bootleg Recordings

Example 1: Bootleg recording made during rock concert (1971)
Example 2: Bootleg tape of a live performance of the opera
The Marriage of Figaro (1962)

Both of these examples are bootleg recordings, created without the consent of the performers.145 They are considered together since the analysis is similar in most respects.

a) Does federal antibootlegging law prohibit use of these sound recordings?
Both examples involve bootleg recordings made prior to 1972. Since they involve musical performances, they appear to come within the ambit of the federal antibootlegging laws. Even though the unauthorized recordings were originally made before December 8, 1994 (the date on which the federal antibootlegging law went into effect), further copying or distribution (or sale, rental, or trafficking) after that date appears to be actionable. The library’s copying would come within the terms of the law; streaming the recordings does not appear to violate the law, but as discussed above, there is some ambiguity in the way the law is drafted.
Are there exceptions that would permit the library’s activities? The statute does not explicitly incorporate the exceptions and limitations in the Copyright Act, such as “fair use” or the library privileges contained in section 108.146 Whether a court would be likely to incorporate these defenses, and whether a library’s preservation and streaming would come within the scope of section 108 or fair use, is discussed in sections 4.2 and 4.3.
Criminal bootlegging would occur only if the library engaged in any of these acts “without the consent of the performer or performers involved, knowingly and for purposes of commercial advantage or private financial gain.”147 Since it assumed that the library is not being compensated or deriving any financial benefit from its activities, it would not come within the criminal law.
The federal law does not preempt state law, so state protection against bootleg recordings must also be considered.

b) Does federal copyright law protect these recordings?
No, because the recordings were first fixed prior to February 15, 1972, and are not works of foreign origin. There may be copyright protection for the underlying works, however (see the discussion below).

c) Do state criminal record piracy or antibootlegging laws protect these sound recordings?
As discussed above, on the basis of the limited sample (and subject to the results of a full state survey), there is no significant risk that a library could come within the state criminal laws if it gets no compensation, direct or indirect, for streaming.

d) Does state civil law (common law) protect these sound recordings?
Concert performances have been recognized as a valuable asset and granted protection against the unauthorized recording and exploitation of recordings of performances such as these. For example, Metropolitan Opera involved an off-the-air recording of an opera radio broadcast.

e) Who are the right holders for these sound recordings?
Usually the right holders in a concert or opera performance are, in the first instance, the performers. This is true for federal law and under California’s civil statute. If the performer is a musician, the musician or the band or orchestra to which the musician belongs may have an exclusive recording contract with a record company. In that case, depending on the terms of the contract, the record company might own the performance or the performers might own it but be unable to authorize its exploitation without agreement from the record company.

f) What legal protection, if any, exists for the underlying work?
In the case of the rock concert, the underlying works are a series of musical compositions. These works are likely to be protected by federal copyright law. The names of the writers and the music publisher that can license the compositions can usually be determined through Copyright Office records or through ASCAP or BMI.148 Absent an exception, on-demand streaming would require license from the music publisher to make the server copies necessary to generate the streams, unless the use qualifies as a fair use (see the discussion in section 2.1.5). Those copies would not be covered by the compulsory license: first, because they are not being distributed in phonorecords; and second, because they are made from bootleg copies. The performances themselves (i.e., the streaming) could be licensed through ASCAP and BMI licenses.
The opera is in the public domain in the United States because it was first published before 1923.

4.1.2 Recordings of Radio Broadcasts

Example 3: Archival recording of a commercial network radio news broadcast (1943)
Example 4: Archival recording of a commercial radio variety program including music and comedy (1950)
Example 5: Airchecks from a defunct radio station (voice only, no music) (1946)

a) Does federal antibootlegging law prohibit use of these sound recordings?
No. All three of these examples involve legitimate copies of radio broadcasts. They are not bootlegs.

b) Does federal copyright law protect these recordings?
No, because the recordings were first fixed prior to February 15, 1972, and all are works of U.S. origin. There may be copyright protection for the underlying works, however, as discussed below.

c) Do state criminal record piracy or antibootlegging laws protect these sound recordings?
As discussed above, on the basis of the limited sample (and subject to the results of a full state survey) there is no significant risk that a library could come within the state criminal laws if it gets no compensation, direct or indirect, for streaming.

d) Does state civil law (common law) protect these sound recordings?
Example 3 involves a recording of a commercial network radio broadcast made in 1943. In CBS, Inc. v. Documentaries Unlimited, Inc.,149 a New York court upheld protection for a radio news broadcast. The defendant made an off-the-air recording of the broadcaster’s news announcement of President John F. Kennedy’s assassination, and incorporated it into a phonograph record for commercial distribution. When the broadcasting company and announcer brought suit, the defendant argued that a performance had to have artistic or literary value to warrant protection, and that old news does not qualify. The court held that “[a] broadcaster’s voice and style of talking is, to all intents and purposes, his personality, a form of art expression, and his distinctive and valuable property.”150 The court found it was “a clear case of appropriation for commercial profit of another’s property right.”151 The court did not clearly distinguish between the underlying work and the performance, as it also said that the announcer did not merely repeat news releases but “added to them matter of his own composition.”152
The sound recording in Example 4 (a commercial radio variety program with music and comedy from 1950) involves performances by musicians (and possibly vocalists), a comedian, and likely an announcer. As discussed above, radio broadcasts of musicians’ performances and announcers’ performances have been held protectable. A comedian’s performance should have at least as strong a claim to common law protection, although no cases were found recognizing common law rights in recordings of radio broadcasts of comedy performances.
Example 5 is a recording of airchecks from a defunct radio station (voice only, no music) made in 1946. An aircheck is a recording made during a broadcast. Sometimes an aircheck was made (or edited) to include only the announcer’s segments of the shows (which may have been used to demonstrate his talent to a potential employer or advertiser), as in this hypothetical. Sometimes airchecks were made of the entire program (including any music, advertisements, etc.) to provide to sponsors so they could confirm that their advertisements were delivered as agreed, and this would likely be deemed a limited publication.153 But in some cases airchecks were duplicated and distributed to the public, in which case the recording could be deemed published. The analysis for airchecks would be similar to that for commercial radio newscasts.

e) Who are the right holders for these sound recordings?
Commercial sound recordings are usually owned by the record producer, but that is not necessarily the case with respect to the sound recordings in this study. Ideally, ownership of unpublished sound recordings would be established by looking at the agreements and employment arrangements, but written agreements may not be available (and in some cases may never have existed).
For radio broadcasts such as those in Examples 3 and 4, in many cases the broadcaster will have acquired all the rights in the recording, in which instance those rights may still be owned by the broadcaster or a successor in interest. Unless there is an agreement to the contrary, the broadcaster will likely own all rights to the contributions of its employees (e.g., the program producer, sound technicians, script writers) under the “work made for hire” doctrine or analogous common law principles.154 If, however, the announcer in Example 5 created his own aircheck not for the station’s purposes but for his personal use in seeking employment, it might be considered outside the scope of his employment and owned by him rather than by his employer.
A radio station may also have had employment agreements, particularly with respect to key individuals. For example, in CBS, Inc. v. Documentaries Unlimited, CBS had an exclusive employment contract with the announcer and consequently owned exclusive rights in the broadcast material. The extent to which the performers who were not technically employees could assert rights would depend on the working relationship and the terms of any agreements between the parties. Prominent guest entertainers like musicians, vocalists, and comedians would likely have had contracts or releases describing the grant of rights to the radio station, although whether those contracts could be located is another matter. If commercial distribution of copies was not envisioned, the station may have acquired rights from guest performers only for broadcast (and possibly for any incidental or archival copies). For example, in Ventura v. Titan Sports, the court held that Jesse Ventura’s contract for services as a wrestling commentator did not contemplate videotape sales.155
Even if some distribution was contemplated, those contracts may or may not have included rights to use the recording in later-developed media. In Ettore v. Philco Broadcasting Corp., discussed in section 3.2.1, the court held that Ettore had not consented to television broadcasts of his boxing match with Joe Louis because commercial television had not existed at the time. But in Silvester v. Time Warner, Inc.,156 plaintiffs—individual recording artists who signed recording contracts between 1956 and 1996—claimed that the record companies had no right to exploit their sound recordings in digital media. The court held that their contracts, which conveyed the right to reproduce and sell the sound recordings “by any method now known, or hereafter to become known,” conveyed full ownership rights to the record companies.157 It is possible, however, that in the context of a radio program whose distribution in copies was not contemplated, any release by performers may not have been so expansive as to include later-developed media.
In short, it cannot be assumed that the radio station or other producer of the recording is the sole right holder in the recording and could authorize reuse of the recordings without agreement from the performers. Absent evidence to the contrary, it must be assumed that performers also have rights. In addition, as suggested above, musicians under exclusive recording contracts may not be able to consent to further exploitation without the consent of their record company. If performers were members of unions such as AFM (American Federation of Musicians) or AFTRA (the American Federation of Television and Radio Artists), then terms on which performers can authorize further exploitation may be subject to union agreements.
The fact that a station is defunct, as in Example 5, does not necessarily say anything about the rights, which may have been transferred to a successor. If there is no successor, a court may find that the announcer holds the rights. Polygram Records, Inc. v. Legacy Entertainment Group, LLC158 involved a three-way dispute for the rights to exploit recordings of performances on WSN radio by singer Hank Williams and his band in the 1950s. The recordings were made to enable WSN to broadcast Williams’s regular show when he was out of town, and apparently were used for no other purpose. The claimants were Williams’s heirs, Polygram Records (the successor to the record company with which Williams had a recording contract during the relevant time period), and Legacy Entertainment, which acquired acetate recordings of the broadcasts from a former WSN employee (who apparently fished them out of a trash bin when WSN was moving offices) and purported thereby to be a successor to WSN’s rights.
The court dismissed Legacy’s claim because there was no evidence it acquired any intangible property rights. It also rejected Polygram’s claim of ownership based on an “exclusive services” contract with Williams, concluding that the contract covered only recordings for the purpose of producing records but not for making prerecorded radio broadcasts. Accordingly, the court awarded control of Williams’s interest to his heirs. It is not clear what the arrangement between Williams and WSN was, or how the court would have ruled had WSN been present.

f) What legal protection, if any, exists for the underlying work?
For radio programs in general, if the scripts had been separately copyrighted in another medium (e.g., in printed form) and if the copyright had been renewed, they would still be protected by federal copyright law.159 If the copyright in the underlying scripts had not been renewed, they would be in the public domain. It appears that scripts were more commonly copyrighted when the radio program involved a narrative storyline, such as the Amos ‘n’ Andy and Lone Ranger radio programs.160In Examples 3-5, if the scripts were not separately copyrighted and remained unpublished at the time the 1976 Copyright Act went into effect, there is still a possibility that protection for the underlying material could be asserted.161
Assuming there was originality in the description of the current news in Example 3 (radio news broadcast), it could be protected by federal copyright law, but it would enjoy only a very thin copyright.162 The same might be true of the announcer’s script in Example 5 (airchecks). Example 4 (variety program) also involves musical compositions and comedy routines. An underlying musical composition is likely protected by copyright, unless it is in the public domain (see section 2.1.5). There is relatively little law on copyright protection for jokes, but it is likely that jokes are protectable (though they would enjoy only a “thin” copyright against close duplication).163 In this case, an entire comedy routine, not just a single joke, is involved. A compilation of jokes could receive protection, but such protection would likewise be thin.
Nevertheless, the possibility of a right holder seeking protection with respect to the underlying work in these examples, other than in the case of musical works, seems very attenuated.

4.1.3 Interviews

Example 6: Press conference or radio interview with a well-known personality (1964)
Example 7: Oral history or man-on-the-street interview with an “average” person (1962)
Example 8: Taped interviews that contributed to a story (some of which are quoted in the story), done by a journalist who worked for a major weekly news magazine and donated to the library with a collection of the journalist’s papers (1968)

a) Does federal antibootlegging law prohibit use of these sound recordings?
It appears that all of these recordings were made with the knowledge and consent of the speakers, so federal antibootlegging laws would not apply. In addition, the federal antibootlegging laws apply only to musical performances.

b) Does federal copyright law protect these recordings?
No, because the recordings were first fixed prior to February 15, 1972, and all are works of U.S. origin. However, there may be copyright protection for the underlying works: see the discussion below.

c) Do state criminal record piracy or antibootlegging laws protect these sound recordings?
As discussed above, on the basis of the limited sample (and subject to the results of a full state survey) there is no significant risk that a library could come within the state criminal laws if it gets no compensation, direct or indirect, for streaming.

d) Does state civil law (common law) protect these sound recordings?
Few cases have addressed rights in sound recordings of interviews, and none has included a comprehensive discussion of rights.
In fact, the state cases that deal with sound recordings of interviews are inconsistent. In Lennon v. Pulsebeat News, Inc.,164 the defendant sought to distribute records copied from earlier taped interviews with the Beatles. A New York trial court granted the Beatles a temporary restraining order. It rejected defendant’s argument that its sale of the recordings was justified because the interviews had originally been news, stating that while it is true there is no bar to reporting news, “there is no justification for utilizing for profit, without plaintiffs’ permission, their distinctive manner of speech and expression which for reasons not material herein have become valuable property.”165 The case provides no information about the source tapes.
Current Audio, Inc. v. RCA Corp.166 involved use of an audio recording of a press conference with Elvis Presley. The conference, held in anticipation of a series of concerts Presley was to give in New York, was attended by representatives of all communications media and recorded on audiotape and film for later television replay. Current Audio sought to publish a 2-1/2-minute excerpt in a record to be included with the debut issue of its multimedia news magazine. RCA, which had exclusive rights in Presley’s recordings and related publicity rights, sued to enjoin use of the recording, as well as photographs and other material from the press conference. RCA claimed that the recording would unfairly compete with recordings it sold of Elvis’s performance at Madison Square Garden and of a press conference held several years earlier when Presley entered the service.
The court denied the injunction. Recognizing that Presley was a “singer of note” the court said that he was not “performing” in the press conference as that word applies to his valuable property, but was engaged in “a public nonartistic use of his speaking voice,” in which RCA had no rights under the contract. The court stated:
The spontaneous “give and take” of an unrehearsed public press conference is of a wholly different character than the delivery of a formal speech or address, or the performance of a musical or artistic work. … [I]n many ways a press conference stands as the very symbol of a free and open press. … To hold, as defendant urges, that one who has freely and willingly participated in a public press conference has some property right which supersedes the right of its free dissemination and permits such party to control or limit its distribution would constitute an impermissible restraint upon the free dissemination of thoughts, ideas, newsworthy events and matters of public interest.167
Although the state law cases dealing with interviews in the form of sound recordings are sparse, other state courts have addressed the question of whether an interviewee has a protectable right in an interview in other contexts. In a 1926 case, Jenkins v. News Syndicate Co.,168 a New York trial court found a common law copyright interest in an interviewee’s contributions to a newspaper article detailing her ideas and opinions for a column she was in negotiations to write. But 50 years later, in Estate of Ernest Hemingway v. Random House,169 New York’s highest court rejected the Hemingway Estate’s common law copyright claims to his interviews. The court expressed reservations about giving “conversational remarks” the status of “literary creations,” although it did not categorically state that such copyright could never apply. It suggested that courts should apply a relatively high threshold when finding common law copyright, and require at least “that the speaker indicate that he intended to mark off the utterance in question from the ordinary stream of speech, that he meant to adopt it as a unique statement and that he wished to exercise control over its publication.”170
In Falwell v. Penthouse Int’l., Ltd.,171 the Reverend Jerry Falwell gave an interview to two journalists who sold the interview to Penthouse magazine. Falwell sued Penthouse for infringement of common law copyright, among other claims. The court dismissed his claim, stating that Falwell “cannot seriously contend that each of his responses in the published interview setting forth his ideas and opinions is a product of his intellectual labors which should be recognized as a literary or even intellectual creation.”172The court also observed that Falwell “willfully and freely participated in the interview,” which he was aware was not a private conversation but intended for dissemination to the public. Falwell was free to pursue contract claims against the journalists (who allegedly violated the conditions of the interview), but “he is trampling upon fundamental constitutional freedoms by seeking to convert what is essentially a private contractual dispute into a broad-based attack on these principles of freedom of speech and press which are essential to a free society.”173
A later case decided in federal court under the federal copyright law is also instructive with respect to interviewee rights. In Taggart v. WMAQ Channel 5 Chicago,174 WMAQ, a Chicago-based television station, videotaped a prison interview with a convicted sex offender in connection with a report on the lax regulation of summer camps. Taggart alleged that he had requested that the tape not be used in any manner, and when WMAQ broadcast an excerpt he sued for copyright infringement and other claims. The court held that he did not have a copyright interest in unprepared and spontaneous utterances during an interview and dismissed his copyright claim.
So what do the cases indicate about state law protection for the sound recordings in Examples 6-8? The state law cases focus primarily on whether or not the interviewee has an interest in an interview, and some of them appear to conflate the interests of the producer of the sound recording with those of the performer.175 Nevertheless, it appears as a general matter that such recordings are protectable, and that the producer of a sound recording of an interview, the interviewer, and the interviewee all might have an interest, although courts have not been particularly sympathetic to interviewees who seek to block dissemination of material from public press conferences or news interviews.
If Example 6 is a radio interview, the station likely has a proprietary interest in the recording, both as the producer of the recording and through the contribution of the announcer, who presumably was its employee.176 If the recording is of a press conference, the case for a protectable interest on the part of the producer of the recording may be weaker. There is no single interviewer to contribute protectable authorship, and although the circumstances of the recording in this example are unclear, it is possible that a court would not find sufficient authorship in its production to accord protection to the creator of the sound recording, particularly in light of the strong public policy concerns expressed in the Current Audio case involving Elvis Presley.
There were no cases found that were directly relevant to Example 7, the oral history or man-on-the-street interview. The source of the interviews and the circumstances under which they were done are unclear; therefore, any analysis of rights is speculative. With regard to the man-on-the-street interview, assuming the interviewer is employed by a radio station, as seems likely, the station or its successor would have a proprietary interest. In the case of the oral history interview, the interviewer was likely also the creator of the sound recording, so it is possible that the court would recognize a proprietary interest in that individual. The circumstances under which the recordings were created or acquired by the library, however, may indicate that the interviewer consented to their exploitation in a library context. Rights of the interviewees are discussed below.
Example 8 involves taped interviews that contributed to a story (some of which were quoted in the story) done by a journalist who worked for a major weekly news magazine and donated to the library with a collection of the journalist’s papers. No cases were uncovered that were directly relevant to this example. For the reasons discussed above concerning Example 7, a court may recognize a proprietary interest in the recordings on the part of the journalist and/or the employer magazine publisher as a work made for hire. Some publishers assert a strong proprietary interest in unpublished materials underlying news stories, although given the amount of time that has elapsed, this may not present a problem.

e) Who are the right holders for these sound recordings?
The possible rights of the sound recording producer and the interviewer are discussed above. But what about the rights of the individuals being interviewed?
Concerning Example 6 (press conference or radio interview with a well-known personality), the cases suggest that under state law an interviewee in a press conference would not have a protectable interest in the substance of the interview (e.g., one that would prevent use of a transcription), but the cases addressing interviewees’ rights in their performances in sound recordings of press conferences and interviews (where performance is also involved) provide no clear answer. The court in Current Audio concluded that Presley had no ownership in the recording of his remarks, while the Lennon case held otherwise in connection with a recording of the Beatles interview, with little explanation. To the extent federal law might be instructive, the Taggart case rejected an interviewee’s attempt to prohibit use of his recorded remarks.
Example 7 (oral history or man-on-the-street interview) presents similar considerations concerning rights of interviewees. If the interviewer obtained a release, it would govern the scope of permissible use, at least as between interviewer and interviewee (i.e., it would determine whether or not the interviewer could convey rights to stream the interview recording or not). On the other hand, a 1962 release may not have addressed later-developed means of dissemination. If the interviewer did not get a release (or the release cannot be located), a court might be willing to find an implied right to use the recording on the ground that the interviewee knew the purpose of the interview, spoke freely, and should be presumed to have consented to use of the recording. Whether it would be willing to allow use for on-demand streaming is another matter.
Example 8 (journalist’s taped interviews) presents similar considerations concerning the rights of interviewees. The nature of the recordings and the resulting story are relevant to whether the interviewees could assert any other rights. It is possible that the interviewees’ willingness to speak to a reporter for the express purpose of assisting in a news article might lead a court to conclude that the tapes could be used (similar to the court’s rationale in the Current Audio case), though whether that use would extend to on-demand streaming is another matter, as discussed above. However, people often speak to reporters for background and allow taping merely to assist the reporter in writing the article, so the rationale in Current Audio may be less persuasive here. The fact that some portions of an interview were published in a news article does not necessarily make the unpublished portions of the sound recording fair game. If a particular interviewee served as an undisclosed source pursuant to an agreement with the interviewer, and the information was used by the journalist accordingly, that interviewee might be able to assert a common law copyright claim or a privacy claim if the tapes were made publicly available. A third party such as a library would not be bound by an agreement between the interviewer and his sources, but such an agreement could distinguish this example from the Falwell case, where Falwell was well aware his remarks would be published.
In sum, the law on proprietary interests in sound recordings of interviews is far from clear. A cautious assumption (absent evidence otherwise) is that the sound recording producer, the interviewer (if different from the sound recording producer), and the interviewee all have potential interests, though courts have shown themselves unsympathetic to interviewees who seek to block dissemination of material from public press conferences or news interviews done in public or with knowledge that they would be made public.

f) What legal protection, if any, exists for the underlying work?
State courts often conflate the interest in the sound recording with rights in the underlying work unless the underlying work has been separately published and copyrighted. This is likely a consequence of the way in which cases are presented to the courts. However, theoretically there may be a different set of rights in the underlying work, even in the case of an interview. Courts have held that interviews fixed in a tangible medium of expression are eligible for federal copyright protection.177 Unpublished interviews (even if their tangible medium is a pre-1972 sound recording) received federal copyright protection effective January 1, 1978.
Who would own the rights to such an interview? Both the interviewer and the interviewee would likely have an interest. Professor Paul Goldstein in his copyright treatise states that courts will often infer nonexclusive licenses from interviewee to interviewer to exploit answers given in an interview. For this and other reasons he believes they should be considered joint works of the interviewee and the interviewer. (In the case of a joint work, any of the authors may grant a licensee to use the work.) Nevertheless, he reports that the “scant case law on the question treats interviews as consisting of two individual works rather than as a single joint work.”178 While it may theoretically be possible, the likelihood of a federal copyright claim based on an interview recorded in a pre-1972 sound recording seems to be remote.

4.1.4 Recording of Live Performance (Foreign Origin)

Example 9: University student violin recital, performing Beethoven violin sonatas, recorded in England, with permission (1971)

a) Does federal antibootlegging law prohibit use of this sound recording?
No. Since the recording was made with permission, federal antibootlegging law is not applicable.

b) Does federal copyright law protect this recording?
To answer this question, it will be assumed that the violinist was an American, but her accompanist was an English national, as was the party who fixed the recording, who did so at the performers’ request. On these facts, the recording would likely be protected by federal copyright law under the copyright restoration provisions, which are discussed in section 2.1.6. It would meet the eligibility requirements for restoration because (1) England is an “eligible country” for purposes of copyright restoration because it was a member of the Berne Convention on the date that the United States passed the URAA; (2) the recording was created prior to February 15, 1972 (the date on which sound recordings were first protected under U.S. copyright law) and after January 1, 1946, so it was likely not in the public domain in England on the effective date of the URAA, January 1, 1996; and (3) there is at least one right holder who is not a U.S. national.179 Under the URAA, the owner of the restored work is the author or initial right holder of the work (presumably the performers, the sound recording producer, or both) as determined by the law of the source country. Since the party who fixed the recording and one of the performers are English nationals, the sound recording appears to be eligible for restoration.

c) Do state criminal record piracy or antibootlegging laws protect this sound recording?
As discussed above, on the basis of the limited sample (and subject to the results of a full state survey) there is no significant risk that a library could come within the state criminal laws if it gets no compensation, direct or indirect, for streaming.

d) Does state civil law (common law) protect this sound recording?
Sound recordings of musical performances have been protected by state civil law, as discussed above. No court has yet ruled on the permissibility of concurrent state and federal protection, and as a practical matter, a right holder would be more likely to bring suit under federal copyright law, if such protection is available.

e) Who are the right holders for these sound recordings?
Ownership of the restored copyright would vest initially in the initial right holder of the sound recording (presumably the performers, the sound recording producer, or both) as determined by the law of the source country, which in this case would be England.180 If the initial right holder(s) later transferred the rights to another party, it appears that the transferee would be the right holder of the restored copyright, but the law in this area is not well settled.181

f) What legal protection, if any, exists for the underlying work?
The underlying works—Beethoven violin sonatas—are in the public domain. A particular arrangement of a public domain work may have sufficient originality to qualify for copyright protection as a musical composition.182 Protection for musical compositions is discussed in section 2.1.5, and special considerations concerning streaming to other countries are discussed in section 4.2.5.

4.2 Streaming of Pre-1972 Recordings: Are Libraries Privileged?

Section 4.1 discussed possible claims by right holders of pre-1972 sound recordings. Case law is sparse, and in many situations the ability of a right holder to bring a claim is speculative and uncertain. Most of the cases that have arisen involve unauthorized copies made and distributed for profit by commercial competitors. How relevant are they to library use for research and scholarship? Can one justifiably conclude that if no one has yet sued a library under state law for use of pre-1972 sound recordings, the risk of a claim is negligible?

Experience suggests that the risk of suit is low, but the track record to date is not dispositive. There have been very few federal copyright cases against libraries, yet one should not conclude for that reason alone that libraries therefore cannot be liable for copyright infringement. One of the reasons for the dearth of cases may be that libraries have been conservative in their approach to copying and disseminating pre-1972 sound recordings, and no sensible right holder would sue on the basis of the dissemination of a few copies. But digital technology provides libraries with technological capabilities they have not had in the past: with relatively few copies, a library can theoretically enable wide dissemination that would have an impact much greater than previously possible.

Unlimited on-demand streaming by libraries could impair the commercial value of some types of pre-1972 sound recordings, by competing with an existing market or discouraging market entry. Many of these sound recordings (for example, the airchecks, oral history, and man-on-the-street interviews) have little commercial potential, but some do. A bootleg recording might compete with an authorized recording by the same band or of the same opera. Digital technology has provided new markets for works thought to have exhausted their commercial value.183 These observations are not meant to suggest that a library will be liable in any particular instance, but merely that the possibility cannot be dismissed based on the historical record alone.

Libraries have a valuable role in society, and the important public policy objectives they serve will almost certainly be a significant factor in any determination of liability. This section considers more closely the status and role of libraries and archives and explores the likelihood that any of the potential claims already discussed would be successful in the face of possible justifications put forth by libraries.

Throughout this study, it has been assumed that the library will not receive any payment for streamed recordings. This assumption should not be interpreted to mean that it can never be legal for a library to receive compensation of any kind. But as a general matter, receipt of compensation can create a heightened risk of liability. Nonprofit entities can engage in activities that are interpreted as “for profit” or “for commercial advantage.”184 Moreover, some of the state criminal statutes simply refer to receiving “compensation” for a recording, or “selling” it, without specifically conditioning that activity on obtaining a commercial advantage. It is simply not possible in a study of this nature to explore the circumstances and extent to which payments might be permissible without violating the various laws that have been discussed here; that issue is worthy of a study in itself. This study is designed to provide general guidance; accordingly, it assumes that libraries will seek to minimize the risk of liability by streaming without compensation. In the specific circumstances of any particular case the risk of liability from some form of compensation may be slim, but that is a determination better made on a case-by-case basis.

4.2.1 Claims under Federal Copyright Law

Section 108. The important role of libraries and archives is recognized in section 108 of the Copyright Act, which provides exceptions for various library activities. Section 108(b) allows libraries to copy unpublished works in digital form (a necessary step to enable streaming), but limits the number of copies to three. Even if libraries were allowed to make enough server copies for streaming, section 108 limits use of the digitized works to the library premises. So an unpublished pre-1972 foreign sound recording whose copyright was restored would be limited to on-premises use under section 108. The same would be true of any unpublished underlying work. If the underlying work were published, a library could digitize it only if it were lost, damaged, stolen, deteriorating, or obsolete, as discussed in section 2.1.4. The digital version would likewise be limited to on-premises use. Legislation to update library privileges may be introduced in the 111th Congress, but how that legislation might affect a library’s ability to stream pre-1972 sound recordings is speculative.

If the underlying work meets the requirements of section 108(h), i.e., the work is in the last 20 years of copyright and the library has determined after a “reasonable investigation” that the work is subject to normal commercial exploitation or available at a “reasonable price,” then creation of server copies and off-premises streaming would be permitted. Section 108(h) provides little help for streaming pre-1972 copyrighted foreign sound recordings, since none of these recordings will be in the last 20 years of their copyright term for many years.185

Section 115. Where the work underlying a sound recording is a copyright-protected musical composition, the streaming of that composition could be authorized under a public performance license from ASCAP or BMI, but that license would not authorize the making of server copies. Legislation or regulations under section 115 to reform music licensing could address this issue, but have not done so yet.

Fair Use. Fair use is also available to libraries. Could streaming of sound recordings qualify as fair use? In some instances it could, but fair use is a fact-based determination, and no conclusion can be drawn that applies across the board to all sound recordings protected under federal copyright law, or even to all unpublished sound recordings. The first fair use factor, the purpose and character of the use, would in many cases tend to favor library copying for preservation and limited dissemination by a library focused on scholarship and research uses. Streaming does not appear to be transformative, but that is not dispositive of factor one, particularly in the context of library use.186

The second fair use factor, the nature of the copyrighted work, would likely favor copyright owners of unpublished works, particularly if the work was neither published nor publicly disseminated. One of the considerations under the second factor is whether the works are predominantly creative or factual, and in this respect there are differences among the examples. It is doubtful that this consideration would have much weight in the context of use for scholarship or research. The other consideration under this factor is whether or not the works are published. Fair use is narrower with respect to unpublished works; the law has been very protective of the copyright owner’s right of first publication. The recordings in this study are unpublished. However, courts are generally more receptive to fair use claims concerning a work that has been publicly disseminated, even though not technically published. Some of the recordings (e.g., the recordings of radio broadcasts) appear to fall into this category, but others (e.g., the journalist’s interview tapes) appear not to. Any underlying works, such as musical compositions, have to be considered separately, and those may well be published.

The third factor would also favor copyright owners, if the entire works were used. Using only excerpts would enhance the possibility that the use would be considered a fair use, but scholars and researchers may find excerpts inadequate for their purposes.187 In any event, the third factor has been given little weight in some recent fair use cases, particularly in the digital environment.188

The fourth factor is the effect of the use on the actual or potential market for or value of the work. Here, courts will look not only at the likely effect of a single defendant’s use but also at the effect on the market if the use should become widespread. If many libraries provided on-demand streaming of the same recording, there could potentially be a market effect that would not exist if only a single library made the work available.189 Unpublished works often exist in only a single copy or few copies, so there may be only one or a handful of libraries making the work available. (In the case of bootleg recordings, there may be many copies.) Even one library’s activities could have an economic impact, though, if it streamed simultaneously to multiple users.

Courts recognize potential market harm only with respect to “traditional, reasonable or likely to be developed markets.”190 The market potential of these sound recordings (and the underlying works) varies significantly. In some cases, streaming could harm the value or potential market for an unpublished work. In others, it could create a market for such a work. Some sound recordings may never have a commercial market, but only a very limited scholarly interest.

Finally, it should be noted that ease of licensing can be a factor in determining fair use; if there is no “ready market or means” to pay for the use, it is more likely to be considered a fair use.191 In some of the examples, licensing certain rights could be a challenge.

Much of the fair use discussion has focused on the sound recordings themselves, but of course only a small fraction of pre-1972 sound recordings are protected by copyright. But the discussion is also relevant to fair use of the underlying works, which may or may not be published.

ATRA. Under ATRA (discussed in section 2.1.4), the Library of Congress is exempt from federal copyright liability for certain activities with respect to news programs, which it may under certain conditions reproduce and lend to researchers or deposit in other libraries. A few of the examples are new programs. But even assuming that the exemption were applicable to the LC’s activities with respect to pre-1972 sound recordings protected by state law, ATRA does not authorize the Library to stream the recordings.

Orphan Works. Orphan works legislation, if passed, might provide a means by which libraries could achieve greater certainty with respect to their use of some older copyrighted works. Some of these old recordings are likely to be orphan works or to embody orphan works. Without knowing the conditions that must be met before a work may be used or the extent to which libraries might receive special treatment for educational and scholarly uses, any observations concerning the likely effect of orphan works legislation are necessarily speculative. But it is virtually certain that any limitation of liability will be conditioned on first undertaking a reasonably diligent search for the right holder, and such searches can be time-consuming. There may be groups of works with similar ownership that can be “cleared for use” in the same search, however, making the process more efficient. Orphan works legislation that limits liability for federal copyright infringements would not be binding with respect to state law claims. Nevertheless, such legislation could be influential. In any event, good faith efforts to find right holders would reduce the likelihood that a state law claimant would later emerge.

Federal Criminal Copyright Law. What is the effect of library status on federal criminal copyright charges? Since criminal copyright is grounded largely on the civil law, the exceptions discussed above are relevant. In addition, to constitute a criminal offense the activity must be done willfully “for purposes of commercial advantage or private financial gain.” The assumption throughout has been that library use of pre-1972 recordings would not be for compensation or commercial advantage (and in any event this is something within a library’s control). Alternatively, a criminal charge can be brought when the infringing activity is done willfully “by the reproduction and distribution, including by electronic means, during any 180-day period, of 1 or more copies or phonorecords of 1 or more copyrighted works, which have a total retail value of more than $1,000.” For several reasons (e.g., the probability that library exceptions and fair use would be broadly construed, or the difficulty of demonstrating that the retail value of the copyrighted works exceeded this limit), it is highly unlikely that library activities could lead to criminal charges. But perhaps most significant, criminal charges are brought by public prosecutors, not right holders, and it is hard to imagine criminal charges being brought against a library on the basis of the activities under discussion.

4.2.2 Claims under Federal Antibootlegging Law

What justifications could a library raise in the face of possible claims under the federal antibootlegging law? What relevance do copyright exceptions have? First, it is important to note that the federal antibootlegging law applies only to musical performances, so it is irrelevant to many pre-1972 sound recordings. Second, the law is narrower in scope than federal copyright law, and apparently would not prohibit public performance, such as streaming, of a bootlegged recording (subject to the points made in section 2.3 concerning the lack of clarity in the drafting). Accordingly, the only activities that seem to be actionable are the making of copies by the library for its server to enable streaming and any copies made by a library for preservation (addressed separately in section 4.3).

The exceptions available in the Copyright Act have not been explicitly incorporated into the federal antibootlegging law, as discussed above. A court might read such exceptions into the law, however. Fair use developed as a common law doctrine, and became part of the federal copyright statute only in the 1976 Copyright Act.192 It is one of the principal means by which copyright accommodates First Amendment values. The role of fair use in copyright suggests that a court would allow a fair use defense to any claim under the federal antibootlegging law.193 While public policy considerations suggest that the exceptions in section 108 should be available to libraries for anti-bootlegging law claims as well, the legislative history is scant, and no cases have yet shed light on this issue.194

The criminal antibootlegging provision is narrower than the criminal copyright provision, and requires that the infringement have been done “knowingly and for purposes of commercial advantage or private financial gain.”195 This would exclude the library activities under consideration in this report, which, it is assumed, would not be done for commercial advantage or for compensation.

4.2.3 Claims under State Criminal Laws

As discussed in section 3.1, some of the criminal statutes in the sample have special exceptions for libraries, while others do not. Nevertheless, the laws in the sample states had requirements that the activity be undertaken for commercial advantage, for profit, for monetary consideration or the like. This would exclude the library activities under consideration in this report, provided they are not done for commercial advantage or for compensation. Finally, it should be emphasized again that criminal charges are brought by public prosecutors, not right holders, and it is hard to imagine criminal charges being brought against a library based on the activities under discussion.

4.2.4 Claims under State Common Law

What is the significance of libraries’ status, and what is the relevance of copyright exceptions in claims brought in state court under state law? Would state courts allow libraries the same privileges as they have under copyright law? State courts have recognized a protectable interest in many of the types of recordings in the examples, but none of the cases involved libraries or not-for-profit uses. Few of the cases involved unpublished sound recordings, and none involved either preservation copying or Internet streaming. For all of these reasons, the analysis is necessarily speculative.

Some states might not recognize a claim at all based on the library activities described above. Claims grounded in unfair competition/misappropriation or on right of publicity require a showing of commercial benefit and/or commercial harm. If state law will not recognize a claim in the absence of commercial benefit to the user, then a suit against a library for use of a sound recording on the conditions described in this study will be unsuccessful. Were a state court to recognize a claim based on commercial harm, there is possible exposure. But a court could conclude that the market impact of streaming in a particular instance does not warrant recognizing a claim, particularly if the streaming activities were limited (e.g., restricted to scholars and researchers streaming a particular recording to only one user at a time). In short, if a common law court does not perceive a commercial wrong, it may not recognize a claim.

In New York, sound recordings are protected under common law copyright, regardless of whether or not they have been published. That protection consists of two elements: (1) the existence of a valid copyright; and (2) unauthorized reproduction of the work protected by the copyright.196Apparently it is not necessary to show either commercial benefit or commercial harm to establish a prima facie case of infringement. Other states might also use this approach, at least with respect to unpublished sound recordings.

If a claim against a library were recognized by a state court, would the court apply exceptions akin to those in federal copyright law? Paul Goldstein observes in his copyright treatise:

The fact that common law copyright is primarily a judge-made doctrine means that it will change over time, and the fact that it is a state law doctrine means that its content will vary from state to state. Further, courts have had little opportunity to flesh out common law copyright’s bare bones on such important points as standards for protection, proof of infringement and remedies for infringement. As a consequence, courts in common law copyright cases frequently consult counterpart provisions in the Copyright Act to fill doctrinal interstices.197

The court in Capitol Records v. Naxos referred to fair use considerations in analyzing Naxos’s claims,198 bearing out Professor Goldstein’s observation that state law courts consult federal copyright to “fill doctrinal interstices.” As discussed above, fair use developed as a common law doctrine, and is one of the principal means by which copyright accommodates First Amendment values.199 The willingness of state courts to look to federal law exceptions in this area is also demonstrated by Comedy III Productions, Inc. v Saderup,200 a case that involved a claim under California’s right of publicity statute against an artist who sold lithographs and T-shirts with a likeness of “The Three Stooges.” In ruling on the artist’s First Amendment defense, the Supreme Court of California adopted wholesale the first fair use factor (the purpose and character of the use), and in particular, the “transformative use” analysis, from the fair use doctrine under federal copyright law.201 The refusal of the court in the Current Audio case, discussed in section 4.1.3, to protect the recording of Presley’s remarks is further evidence of the weight that public policy considerations can have in common law cases.202 In short, it is reasonable to assume that a common law copyright court would recognize a fair use-type exception to accommodate First Amendment interests. At the same time, where a work is unpublished and undisseminated, it is likely that fair use would be narrowly construed.

4.2.5 Claims under Foreign Law

The focus throughout this study has been on the potential for liability in the United States for certain uses of pre-1972 sound recordings. If a U.S. library were to stream to other countries, however, the laws of those countries might govern.203 The term of protection for sound recordings in most countries is 50 years from fixation,204 so this is a concern primarily for sound recordings first fixed between 1958 and 1971. But there is also potential exposure in other countries for use of the underlying works. In the United States, musical compositions and literary and dramatic works first copyrighted in 1922 and earlier are in the public domain, but this cutoff date does not apply in countries that for many years had a term of protection longer than that in the United States.205 There may also be differences in the scope of protection: for example, most other countries do not have a fair use doctrine. In short, a library cannot assume that streaming to users outside the United States would be governed by the same legal rules as streaming to users within the United States. Streaming to foreign countries, particularly of works by foreign nationals, requires consideration of potential exposure under foreign laws.

4.3 Preservation of Pre-1972 Recordings: Are Libraries Privileged?

Creating digital copies solely for preservation purposes is extremely unlikely to be actionable under state law, civil or criminal. It is assumed that libraries are not undertaking these activities for their commercial benefit, and it is difficult to envision any commercial harm to the right holders. Preservation copying is also unlikely to be the basis of a claim for infringement based on common law copyright. Although state courts are not required to recognize federal copyright exceptions, the public policy basis for long-term preservation, particularly as the need for digital preservation becomes increasingly apparent in this country and around the world, would surely persuade a court to reject any claim based on preservation activities in the very unlikely event that such a claim was asserted.

Preservation of unpublished, copyrighted sound recordings is governed by section 108(b) of the Copyright Act and fair use. As discussed above, section 108 is currently under consideration for possible legislative amendment because, inter alia, the three-copy limit in sections 108(b) and (c) are not meaningful in the context of digital preservation.

4.4 Assessing the Risk: Questions to Ask

How does one decide whether and how to proceed with a particular activity in the face of legal uncertainty? As discussed in section 4.0, the first step is to consider the existence of possible claims, as was done above with the examples. Another important consideration in assessing risk is whether someone who has a valid legal claim is likely to assert it. Where there is considerable ambiguity about the law, the factors that go into determining whether there is a possible claim can also be relevant to whether someone is likely to assert that claim. Significant uncertainty about whether a claim exists at all, or about who holds the rights, affects the probability that someone will come forward to assert it. Similarly, the scope of library use (a factor within the library’s control) can affect both the existence of a claim under state law and the likelihood that someone will bring a claim.

Below is a list of questions to consider in assessing the risks inherent in library streaming of a pre-1972 sound recording. The questions relate to the nature of possible claims, the likelihood of suit, and the strategies that libraries may use to reduce the risks. It may not be possible to answer all these questions in every case. Decisions have to be made on the best available information, but with awareness of material gaps.

*    *    *

1. What were the circumstances of the acquisition of the sound recording?
Is there any donor agreement that governs the terms of the library’s use of sound recordings? Acts for which a library has obtained permission will not violate the law. It is important, however, to determine whether the donor is the right holder and whether there are any additional right holders, as a donor or licensor cannot grant any more rights in a work than he or she owns.
Even in the absence of a donor agreement, are there circumstances in connection with sale or acquisition that would support a claim that rights were transferred?

2. What is the provenance of the recording? Who created it, where, and under what circumstances?

3. Is the sound recording protected by federal copyright law? Specifically, when was it first fixed?
(a) If it was first fixed on or after February 15, 1972, it is protected by federal copyright law.
(b) If it was first fixed before February 15, 1972, two additional questions must be answered to determine if the sound recording has federal copyright protection as a “restored work”: (1) Was at least one of the authors or right holders, at the time of creation, a national of an “eligible country”? Eligible countries include all members of the principal international copyright treaties, which encompass most nations around the world and (2) Was the work first fixed on or after January 1, 1946? A sound recording was eligible for copyright restoration only if it was still protected in its source country on January 1, 1996, and the term of protection in most countries is 50 years. Both conditions must be met for a pre-1972 sound recording to have federal copyright protection.
If the sound recording does not meet (a) or (b), then it is protected only under state law. The underlying works may be protected by copyright, however, regardless of whether there is protection for the recording.

4. What are the contents of the recording? Who are the performers, and what is the nature of the performance? Is there an underlying musical composition or other work protected by federal copyright law? (On the latter question, while Copyright Office records may not be dispositive, they may be helpful.)

5. Has the sound recording previously been publicly disseminated? If so, under what circumstances? Was it broadcast? Were copies distributed? If so, how widely and under what conditions?

6. Does the nature of the recording suggest possible privacy interests on the part of the speaker, privilege issues, or the like?

7. Who are the potential right holders? Can anything be inferred about rights from past practices? (For example, there may be evidence that a particular producer generally had consistent practices with regard to releases.)

Other questions may be relevant to assessing whether a right holder is likely to assert a claim. They include the following:

8. Is the right holder still in business? If not, can anything be learned about the disposition of its assets?

9. Is there any evidence that the right holder has complained in the past about unauthorized use of other recordings that he produced or in which he performed? Conversely, is there any evidence that other similar recordings owned by the same right holder have been freely used without apparent complaint? Although it is no defense to copyright infringement that others are also infringing, wide availability of similar works owned by the same right holder may suggest that the risk of a claim is low.

10. What is the potential commercial impact of the use? Is there any evidence that the right holder is preparing to market this work or similar works? (This is likely to be most relevant where there are a number of similarly situated works, such as the tapes of an old radio show.) Is there any evidence that the right holders of similar works are preparing to market them? What evidence is there that the proposed library use of the work could affect an existing market? A realistic potential market? Although these recordings are unpublished, a bootleg recording could compete with an authorized performance of the same work.

11. Are there particular circumstances about this recording that might cause the right holder to object to its use? (This relates to question 6 above.) Sometimes people bring claims for personal or political reasons rather than for commercial ones.206

Finally, in assessing the risk, it is important also to consider factors that to some extent are within the library’s control:

12. Would the library receive any compensation, direct or indirect, from the use? This study has assumed that the library is not being compensated, but it is important to confirm this in any particular factual situation.

13. Are there limitations that a library could put into place to avoid potential commercial impact of library’s use? This is a very important consideration. A library may be able to reduce its risk by limiting the number of simultaneous users, restricting use to registered researchers and scholars, etc.

14. Is there any likelihood that the use could become widespread (i.e., are other libraries streaming the same work)? If the work is unique, this is unlikely, but it is possible that a library has supplied or will supply preservation/deposit copies to other libraries.

None of these questions is itself dispositive. Libraries differ in the extent of their potential legal exposure, as is apparent from the discussion of remedies in section 2.1.8. They also differ in their tolerance for risk.


FOOTNOTES

144 The assumption is that the library will not impose any charge for streaming to avoid any possible argument that it is selling or renting copies under the various statutes.

145 It is worth noting, particularly in connection with the rock concert, that not all recordings taken at a rock concert are necessarily bootlegs. In some instances bands like the Grateful Dead permitted fans to tape and trade copies of performances as long as they did not use the tapes for commercial purposes. See Cason A. Moore, Tapers in a Jam: “Trouble Ahead” or “Trouble Behind”?, 30 Colum. J. L. & Arts 625, 626-30 (2007) (re Grateful Dead taping policy). A band may not consent to recording the underlying musical compositions, however, unless it owns the rights.

146 Nimmer on Copyright, supra note 9, § 8E.03[B][2].

147 18 U.S.C. § 2319A(a) (2000). The ambiguity in the statute regarding liability for streaming is another reason to discount any risk of criminal liability on that ground.

148See ASCAP, ACE Title Search, http://www.ascap.com/ace/; BMI, Search BMI Repertoire at http://www.bmi.com.

149 248 N.Y.S.2d 809 (N.Y. Sup. Ct. 1964)

150Id. at 811.

151Id. at 812.

152Id. at 811.

153See supra note 9.

154See, e.g., Storer Broadcasting Co. v. Jack the Bellboy, Inc., 107 F. Supp. 988 (E.D. Mich. 1952) (radio program name, material, and scripts written and developed by employee in the course of his employment became property of radio station employer).

155Ventura, 65 F.3d at 731.

156 763 N.Y.S.2d 912 (Sup. Ct. N.Y. Cty. 2003).

157Id. at 916-17.

158 205 S.W. 3d 439 (Tenn. Ct. App. 2006).

159 In that case, the copyright owner of the script might be able to limit use of the sound recording. See, e.g., King v. Mister Maestro, Inc., 224 F. Supp. 101 (S.D.N.Y. 1963) (court enjoined distribution of phonograph records of plaintiff’s “I Have a Dream” speech; King had registered the copyright in the text of his speech as an unpublished work, and the court held that he had not published the speech by delivering it). See supra notes 9 & 15.

160E.g., Silverman v. CBS Inc., 870 F.2d 40 (2d Cir. 1989) (scripts of Amos ‘n’ Andy radio programs); Lone Ranger Television, Inc v. Program Radio Corp., 740 F.2d 718 (9th Cir. 1984) (scripts of Lone Ranger episodes).

161 In Silverman, the court raised but did not decide the issue whether CBS could claim protection not for the actual sounds of the radio program tapes, but for expressive content contained in them beyond what was included in the scripts (e.g., ad-libbed dialog), which had entered the public domain for failure to renew the copyright. 870 F.2d at 43 n. 2.

162 Whether an underlying script still retains copyright protection is a complicated question, but assuming neither the script nor the sound recording was published, the script would become eligible for federal copyright protection as an unpublished work on January 1, 1978.

163 See Foxworthy v. Custom Tees, Inc., 879 F. Supp. 1200 (N.D. Ga. 1995) (holding “redneck” jokes copyrightable); Nimmer on Copyright, supra note 9, § 2.13.

164 143 U.S.P.Q. 309 (N.Y. Sup. Ct. 1964).

165Id. at 309.

166 37 N.Y.S.2d 949 (N.Y. Sup. Ct. 1972).

167Id. at 953 (citations omitted).

168 219 N.Y.S. 196 (N.Y. Sup. Ct. 1926).

169 244 N.E.2d 250 (N.Y. 1968).

170Id. at 256.

171 521 F.Supp. 1204 (W.D. Va. 1981).

172Id. at 1208.

173Id. The court also dismissed claims based on privacy and publicity rights.

174 57 U.S.P.Q.2d 1083 (S.D. Ill. 2000).

175 In the case of a radio station, the production of the recording likely entails some judgment and skill, so there would be sufficient authorship on that score alone. But presumably because the radio station often owns performers’ rights through releases or under the work for hire doctrine, the courts have had little occasion to determine what constitutes a copyrightable contribution on the part of the producer.

176 In the Taggart case discussed above (based on federal law), WMAQ had a copyright interest in the taped interview. State courts have not addressed whether an interviewer has a protectable interest, but see the discussion of the participants’ respective interests under federal law in note 180, infra, and accompanying text.

177 Obviously, sound recordings made of interviews prior to 1972 would not enjoy such statutory protection. However, interviews in that time period recorded in copyrightable media do appear to have enjoyed statutory protection.

178See Goldstein, supra note 11, § 4.2.1.3, at 4:21; Suid v. Newsweek Magazine, 503 F. Supp. 146, 148 (D.D.C. 1980) (interviewer had no copyright in the text of interview he conducted absent an assignment from interviewee). See also Falacci v. New Gazette Lit. Corp., 568 F. Supp. 1172 , 1173 (S.D.N.Y. 1983) (assuming, without discussion, that the author/interviewer in a newspaper interview was the copyright holder); Quinto v. Legal Times of Washington, Inc., 506 F. Supp. 554, 559 (D.D.C. 1981) (stating that interviews contain sufficient creativity in the selection and arrangement of questions to qualify for copyright protection).

179See 17 U.S.C. § 104A(a), (h).

180 § 104A(b).

181See Nimmer on Copyright, supra note 9, § 9A.04[B][2][b].

182Id. § 2.05[C].

183See, e.g., Daniel J. Wakin, The Multiplex as Opera House: Will They Serve Popcorn?, N. Y. Times Sept. 7, 2006 (Metropolitan Opera to make its archive of historic radio broadcast performances available through Rhapsody Internet service); see generally Chris Anderson, The Long Tail: Why the Future of Business is Selling Less of More (2006).

184See, e.g., Hustler Magazine, Inc. v. Moral Majority, Inc., 796 F.2d 1148, 1152 (9th Cir. 1986); Worldwide Church of God v. Philadelphia Church of God, 227 F.3d 1110, 1118 (9th Cir. 1999).

185 Under § 104A(a)(1), a restored work is protected by copyright for the remainder of the term it would have received in the United States had it not entered the public domain. For works that were unpublished on January 1, 1978, the term would be the life of the author plus 70 years. Published works would have a 95-year term. For the reasons discussed in section 2.1.6, only foreign recordings first fixed on or after January 1, 1946, were eligible for restoration. A published sound recording fixed on that date would be protected by copyright through 2040.

186 Moreover, courts are increasingly willing to find transformative use where use of works has a transformative purpose. See, e.g., Perfect 10, Inc. v. Google, Inc., 487 F.3d 701, 724 (9th Cir. 2007).

187 For some users whose goal is simply to identify a particular work and determine its general style or whether it is the same as or different from another work, a short excerpt may be enough. Others, however, may need to study—and possibly to transcribe—the entire work.

188E.g., Perfect 10, Inc. v. Google; Kelly v. Arriba Soft Corp., 336 F.3d 811, 820-21 (9th Cir. 2003).

189 A library may have some control over how widespread streaming of an unpublished work becomes, by limiting its own streaming, conditioning use of copies of unpublished works made under § 108(b) for deposit in other libraries, etc.

190 American Geophysical Union v. Texaco, Inc., 60 F.3d 913, 930 (2d Cir. 1994).

191Id. at 930-31.

192See, e.g., Folsom v. Marsh, 9 F. Cas. 342 (C.C.D. Mass. 1841).

193 Jane C. Ginsburg, The Pros and Cons of Strengthening Intellectual Property Protection: Technological Protection Measures and Section 1201 of the United States Copyright Act, 16 Info. & Comm. Tech. L. 191, 207-09 (2007).

194See Nimmer on Copyright, supra note 9, § 8E.03[B][2][b] and § 8E.03[C][5].

195 18 U.S.C. § 2319A(a).

196 830 N.E.2d 250, 266. Commercial harm would of course be relevant to damages.

197 Goldstein, supra note 11, § 17.5 at 17:44.

198 830 N.E.2d at 267.

199See, e.g., Folsom v. Marsh, 9 F. Cas. 342 (C.C.D. Mass. 1841).

200 21 P.3d 797 (Cal. 2001).

201Id. at 807-10. The court declined to incorporate the entire fair use doctrine into right of publicity law, however, because it found the second and third factors to be not “especially useful” in that context. Id. at 807-808. The court ultimately denied the artist’s First Amendment defense because it found his depiction not sufficiently transformative. Id. at 811.

202 While neither ATRA nor the audiovisual news exception in the Copyright Act embraces streaming, they do suggest a strong public policy in favor of access to news programs for scholars and researchers.

203 Some countries may apply the law of the country where the host server is located, while others may apply the law of the country where the recipients are located, if there is a “real and substantial connection” with that country. See generally Susanna H.S. Leong & Cheng Lim Saw, Copyright Infringement in a Borderless World—Does Territoriality Matter?, 15 Int. J. L. & Inf. Tech. 38 (2007). The “choice of law” rules with respect to Internet transmissions are still developing. Under choice of law principles adopted by the American Law Institute in 2007, the law of the country of receipt of an Internet transmission would govern. Intellectual Property Principles Governing Jurisdiction, Choice of Law, and Judgments in Transnational Disputes § 301 (American Law Institute 2008) (Rochelle C. Dreyfuss, Jane C. Ginsburg & François Dessemontet, reporters).

204 The European Union is considering extending the term to 70 years from fixation.

205 Under the Berne Convention’s “rule of the shorter term,” the term of protection is governed by the laws of the country where protection is claimed, but is limited to the term of protection in the work’s country of origin unless the laws in the country where protection is claimed provide otherwise. The Berne Convention for the Protection of Literary and Artistic Works, art. 7(8) (Sept. 9, 1886, 25 U.S.T. 1341, 828 U.N.T.S. 221, Paris Act 1971).

206 See Donald Liebenson, “Should ‘Dated’ Films See the Light of Today?,” Los Angeles Times. May 7, 2003 (discussing the unavailability of the Disney film Song of the South, presumably because of “racially insensitive” material). The reason that the right holder may be reluctant to have the work disseminated may also bear on a fair use claim, however.