Fair Use in Music

Fair use is not a permission slip for using someone else’s music. In the United States, it is a case-specific copyright doctrine that can allow some unlicensed uses, including criticism, commentary, teaching, scholarship, research, and certain forms of parody. Whether a use is fair depends on the complete set of facts, not on a catchy rule about seconds, notes, percentages, attribution, or disclaimers. Only a court can conclusively decide whether a particular use qualifies.

This article explains the federal framework for music-related fair-use questions. It covers musical compositions, sound recordings, sampling, interpolation, parody, commercial releases, and the different approaches taken by the Sixth and Ninth Circuits. It is educational information about U.S. federal copyright law, not individualized legal advice. State-law claims, foreign law, platform policies, and the facts of a particular dispute are outside this article’s scope.

The basic question

The central question is not simply, “How much did I copy?” Courts ask what the use is doing and how it affects the interests protected by copyright. Section 107 of the Copyright Act identifies four factors: the purpose and character of the use; the nature of the copyrighted work; the amount and substantiality of the portion used; and the effect of the use on the potential market for, or value of, the copyrighted work. The statute describes these factors as part of the inquiry without assigning them fixed weights. Chapter 1, §107 — Limitations on exclusive rights: Fair use

The factors are balanced together. A favorable point on one factor does not automatically win the case, and a difficult point on one factor does not automatically defeat a claim. The U.S. Copyright Office explains that fair use depends on the circumstances and that creative works, including songs, generally receive stronger protection than factual works. More Information on Fair Use

A useful way to think about fair use is as a structured evaluation rather than a checklist with automatic answers:

1. What is the new use, and why is the existing work being used?
2. Is the source work highly creative or more factual?
3. How much was taken, and was the most important or recognizable part taken?
4. Does the new use substitute for the original or interfere with a plausible market for it?

These questions overlap. The same musical excerpt may look different depending on whether it appears in a review, a classroom example, a parody, a commercial song, a documentary, or background music for an unrelated scene.

Factor one: purpose and character

The first factor examines what the new work is doing and the context in which it is being used. Criticism, comment, teaching, scholarship, and research are examples identified by the statute, but listing a purpose does not make a use automatically fair. A commercial purpose matters, yet commerciality is not automatically decisive. Likewise, a nonprofit or educational purpose does not automatically make a use fair. The other facts and the use’s relationship to the original work still matter. Fair Use FAQ

Courts often discuss whether a use is “transformative.” In practical terms, that asks whether the new use serves a meaning, purpose, or function that differs from the source use. But “transformative” is not a magic word. A creator cannot end the analysis simply by saying that a new song, video, or performance has a different message. Courts still evaluate the remaining factors and the particular use at issue.

The Supreme Court’s parody decision in Campbell v. Acuff-Rose Music illustrates this point in a music setting. The Court held that a commercially released parody may be fair use in principle and explained that the four factors must be weighed together without bright-line rules. A commercial release therefore does not automatically lose, especially when the work comments on or targets the original. Campbell v. Acuff-Rose Music, 510 U.S. 569 (1994)

Parody is different from merely placing an old song in a new setting. A parody generally depends on the original work as the object of commentary or ridicule. If the copied material is used as background, decoration, atmosphere, or a convenient source of identity without commenting on the source, the justification may be weaker. The Supreme Court’s later decision in Andy Warhol Foundation for Visual Arts, Inc. v. Goldsmith emphasized that transformativeness is use-specific and that commercial context remains relevant. Commentary unrelated to the copied work may provide less support than commentary that bears a critical relationship to it. Andy Warhol Foundation for Visual Arts, Inc. v. Goldsmith, 598 U.S. ___ (2023)

The practical lesson is to describe the actual function of the excerpt honestly. “I changed the tempo” or “I put it in a different genre” may describe an artistic change, but those facts alone do not establish a fair purpose. Ask what the audience is meant to understand from the use and whether that purpose depends on engaging with the original work.

Factor two: nature of the copyrighted work

The second factor considers the source material itself. Copyright generally gives stronger protection to creative expression than to factual material. Songs combine melody, harmony, rhythm, lyrics, arrangement, performance, and production choices, so the nature factor can matter significantly in music disputes. The Copyright Office describes songs and other creative works as generally receiving stronger protection than factual works. More Information on Fair Use

This does not mean that fair use of music is impossible. It means that a creator should not assume that the second factor is neutral simply because the excerpt is short. A short portion of a highly creative work can still be important, recognizable, or expressive. The character of the source is one part of the overall balance, not a standalone prohibition.

Factor three: amount and substantiality

The third factor asks both how much was copied and how significant the copied material was. Quantity matters, but quality matters too. A small excerpt may carry the most recognizable hook, lyric, melody, beat, or sonic identity of the source. Conversely, a longer use might be justified in a criticism or teaching context if the amount is reasonably related to the purpose. There is no universal formula that resolves this factor.

The U.S. Copyright Office states that the law does not prescribe a number of notes, seconds, or percentage that may always be used safely. Its guidance also advises considering permission in doubtful cases. Fair Use FAQ

That means the familiar “seven-second rule” is not a legal rule. The Copyright Office’s educational material on sampling and interpolations specifically rejects the idea that seven seconds automatically makes a sample permissible and explains that disclaimers do not create protection. Sampling, Interpolations, Beat Stores and More: An Introduction for Musicians Using Preexisting Music

A better question is: what amount was reasonably necessary for the claimed purpose? If a review needs to let listeners hear the feature being criticized, that purpose may call for a different excerpt than an unrelated commercial track that uses the same hook because it sounds effective. Even then, the amount factor is only one part of the analysis.

Factor four: market effect

The fourth factor considers whether the new use harms the market for the original or for derivative uses that copyright owners would ordinarily develop or license. A new recording that competes with the original may present a different market question from a short excerpt used to criticize the original. Courts may also consider whether the type of use is one for which licensing markets exist or are reasonably expected to exist.

Market effect is not limited to whether the new work sounds identical to the source. Consider whether listeners might use the new work instead of the original, whether the copied portion is a commercially important feature, and whether the use could displace a licensing opportunity. These are factual questions. The purpose of the use, the amount taken, and the nature of the source can affect how the market analysis is understood.

The four factors should be read together. For example, a commercial parody may still have a strong fair-use argument when it comments on the original, even though it is sold. By contrast, a commercial song that copies a distinctive element merely to make the new track more appealing may have less support if it is not commenting on the source and could substitute for a licensed use.

Two copyrights may be involved in one song

Music creators often say they used “a song” when the legally relevant question is which part of that song was used. In the United States, a musical composition and a sound recording are separate copyrighted works. The composition is the underlying musical and lyrical work. The sound recording is a particular recorded performance and production. They may have different owners. Musical Compositions and Sound Recordings

This distinction matters for sampling and interpolation.

Sampling copies sounds from a preexisting recording. The copied material may therefore implicate the copyright in that sound recording. Because the recording embodies a composition, the use may also raise questions about the underlying composition, depending on what expression was copied and what rights apply.

Interpolation generally means replaying or re-recording musical material rather than lifting the original recorded sound. That may avoid copying the particular sound recording, but it can still implicate the underlying composition if melody, lyrics, or other protected musical expression is reproduced. The exact rights required for a particular interpolation depend on what was reproduced and the applicable licensing arrangements.

A practical rights map looks like this:

• Original composition: melody, lyrics, and other underlying musical expression.
• Original sound recording: a particular recorded performance and its fixed sounds.
• Sample: use of fixed sounds from the existing recording; potentially recording rights and composition rights.
• Interpolation: newly recorded performance of musical material; potentially composition rights, even if the original recording is not copied.

This is why “we replayed it ourselves” does not answer every rights question, and “we only used a tiny sample” does not answer every fair-use question.

Sampling: the law is not uniform nationwide

U.S. appellate precedent is not uniform on de minimis copying of sound recordings. “De minimis” generally refers to copying that is so minimal that the law does not treat it as actionable in the relevant context. The Sixth Circuit adopted a stricter approach to unauthorized digital sampling of sound recordings. The Ninth Circuit later held that a de minimis exception applies to sound-recording copying in its circuit. Bridgeport Music, Inc. v. Dimension Films, 410 F.3d 792 (6th Cir. 2005)

The Ninth Circuit’s VMG Salsoul decision involved an alleged 0.23-second horn sample. The court held that the alleged appropriation was de minimis because an average audience would not recognize it. That is a case-specific holding, not a seconds-based safe harbor. It does not mean that 0.23 seconds is always permissible, or that a recognizable short sample is always protected. VMG Salsoul, LLC v. Ciccone, 824 F.3d 871 (9th Cir. 2016)

Creators should not generalize either circuit’s rule nationwide. Circuit coverage, later precedent, the precise copyright claim, and the facts of the use matter. If a dispute concerns a sound recording, identify the governing jurisdiction before relying on a de minimis argument.

An early case often associated with sampling history is Grand Upright Music, Ltd. v. Warner Bros. Records, Inc. It involved Biz Markie’s unauthorized use of material associated with Gilbert O’Sullivan’s “Alone Again (Naturally).” It is useful historical context, but it should not be described as a universal rule that every tiny sample infringes. The reported case was decided in 1991, not 1989, and it does not replace the later and differing appellate precedent. Grand Upright Music, Ltd. v. Warner Bros. Records, Inc., 780 F. Supp. 182 (S.D.N.Y. 1991)

Common shortcuts that fail

Several popular claims are legally unreliable:

• “It is under seven seconds.” No fixed duration creates fair use.
• “I used less than one percent.” No fixed percentage creates fair use.
• “I changed the pitch, tempo, key, or instrument.” Artistic alteration does not end the four-factor inquiry.
• “I gave credit.” Attribution may be responsible practice, but it does not create fair use.
• “I added a disclaimer.” A disclaimer does not transform an otherwise infringing use.
• “It is for education.” Educational purpose is relevant context, not an automatic exemption.
• “It is commercial, so it cannot be fair.” Commerciality matters, but it is not automatically decisive.
• “I replayed it, so no permission is needed.” Replaying may avoid copying the original recording while still implicating the composition.
• “A court once allowed a tiny sample.” A case-specific result is not a universal license.

These shortcuts are especially risky because they focus on one fact while ignoring purpose, source creativity, substantiality, and market effect.

A practical decision route

Before releasing music that uses preexisting material, document the decision process. First, identify exactly what was used: lyrics, melody, a recording, a performance, a drum sound, a vocal phrase, or a combination. Second, identify the likely copyright interests and owners. Third, write down the purpose of the use in plain language. Is the work criticizing the original, teaching about it, parodying it, or simply borrowing an appealing element?

Fourth, assess the amount and importance of the material taken. Note whether the excerpt is recognizable or forms a signature part of the source. Fifth, consider market effects, including whether the use competes with the original or with a licensing market. Sixth, check the governing jurisdiction, especially for sound-recording sampling and de minimis arguments. Seventh, if the analysis is doubtful, consider permission or a license rather than relying on a slogan about fair use.

A worked example shows how the route works. Suppose a producer wants to include a recognizable vocal phrase from a prior recording in a commercially released track. The producer should ask whether the phrase was copied from the recording, the composition, or both. If the original recording was lifted, sound-recording rights may be implicated. If the phrase reproduces lyrics or melody, composition rights may also be involved. The commercial release is relevant, but not automatically fatal. The producer must then ask whether the new track comments on or targets the original, how much was taken, how important the phrase is, and whether the new track could replace or compete with a licensed use. The analysis cannot be resolved by saying the phrase lasts only a few seconds or by adding a credit in the liner notes.

What fair use can and cannot tell you

Fair use can provide a legal defense or limitation in appropriate circumstances, but it does not function like advance clearance. A creator may believe a use is fair and still face a claim, a takedown, a demand for payment, or litigation. Conversely, the presence of risk does not prove that a use is infringing. The final answer depends on the facts, the claim, the jurisdiction, and the court’s application of the law.

The safest general principle is simple: treat fair use as a fact-intensive legal analysis, not as a production shortcut. Separate the composition from the sound recording. Distinguish sampling from interpolation. Be precise about the use’s purpose. Do not rely on fixed time limits, percentages, attribution, or disclaimers. For a planned release with meaningful commercial or legal consequences, obtain advice from a qualified U.S. copyright attorney who can evaluate the specific material and jurisdiction.

Open Music Business is educational content only. This article does not provide individualized legal, financial, tax, contract, or royalty advice.