Copyright · 17 U.S.C. § 107
Not a word count. Not a 10% rule. Not "I credited the author." Fair use is an affirmative defense — a four-factor balancing test, applied case by case, and recently recalibrated by the Supreme Court.
→ arrow keys, swipe, or the dots below to advance · deep-link any slide via # in the URL
Why it exists
Copyright hands creators a temporary monopoly so they'll keep creating. But culture is cumulative — and a monopoly with no give would choke the very progress it was built to promote. Fair use is the breathing room.
The author alone may copy, adapt, distribute, and perform the work. That control is the reward that makes writing, filming, and recording worth the risk.
Critics quote. Scholars analyze. Comedians parody. Journalists report. New technology indexes and reuses. None of it works if every borrowing needs permission.
Copyright exists to "promote the Progress of Science and useful Arts" (U.S. Const. art. I, § 8). Fair use is one of copyright's "built-in First Amendment accommodations" (Eldred v. Ashcroft, 2003) — which is why courts rarely entertain a standalone free-speech defense to infringement: the speech interest runs through Factor 1, not a separate constitutional test.
The statute
"…the fair use of a copyrighted work … for purposes such as criticism, comment, news reporting, teaching …, scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include—"
Two things the text quietly settles: the preamble examples (criticism, teaching, news) are illustrative, not a safe harbor, and the factor list is non-exhaustive — "shall include."
The mental model
You don't tally the four factors and declare a winner. Courts weigh them together, and underneath all four sits one question: does this use substitute for the original in a market the author can claim — or does it add something new, for a different purpose? Set the factors below and watch the balance tip.
Click each factor to cycle favors → neutral → opposes. Factors 1 and 4 carry the most weight — so the scale won't track a simple count.
Factor one
The center of gravity. Two questions dominate: is the use transformative — does it serve a genuinely different purpose, adding new expression or meaning — and is it commercial? Campbell made transformativeness the fulcrum: the more transformative the work, the less the other factors weigh against it.
A transformative purpose — criticism, parody, commentary, reporting, or a new function. After Warhol, new meaning alone isn't enough; the purpose must differ.
Commercial use sharing the original's purpose. Commerciality isn't fatal (Campbell), but when the purposes overlap it pulls hard against fair use.
Campbell v. Acuff-Rose
510 U.S. 569 (1994)
The facts
2 Live Crew recorded a raunchy rap parody of Roy Orbison's "Oh, Pretty Woman" and sold it — after being refused a license.
The holding
"Parody has an obvious claim to transformative value." A commercial purpose is not presumptively fatal — but the Court reversed and remanded, leaving the Factor 4 market harm (the rap-derivatives market) to be decided below.
Why it matters
It put transformative use at the heart of Factor 1 and demoted the old presumption against commercial use.
Factor two
The least decisive factor, but real. Copyright protects expression, not facts or ideas — so the kind of work being borrowed sets how much breathing room a user gets. Two axes do the work:
Google v. Oracle
593 U.S. 1 (2021)
The facts
Google copied ~11,500 lines of Java API "declaring code" so developers could reuse familiar calls in Android.
The holding
Fair use. The declaring code is inherently functional — bound up with uncopyrightable ideas and methods of operation — so it sits far from copyright's core, leaving wide room to reuse.
Why it matters
The Court assumed, without deciding, the code was even copyrightable, then resolved everything on fair use — folding a copyrightability question into the test. It really turned on Factors 1, 2 and 4 together (transformative "reimplementation," the functional nature, no market substitution) — but it's the nature analysis that did the distinctive work.
Factor three
Both quantitative and qualitative — and there is no magic percentage. Taking a little can still lose if you take "the heart" of the work; taking a lot can survive if the purpose genuinely requires it — a thumbnail, a search index, the lines a parody must evoke.
No more than needed. Copying tailored to a legitimate transformative purpose supports fair use — even verbatim, whole-work copying can survive.
Taking "the heart." The most central, expressive core of a work — regardless of how small a fraction it is.
Sony Corp. v. Universal
464 U.S. 417 (1984)
The facts
Home viewers used Betamax VCRs to record entire TV broadcasts and watch them later — "time-shifting."
The holding
Copying the whole program didn't defeat fair use: for private, non-commercial time-shifting, taking all of it was reasonable for the purpose.
Why it matters
Proof there's no magic percentage — even 100% can be fair use when the purpose genuinely needs the whole work. The mirror image of a tiny excerpt that takes the "heart."
Factor four
The one that most often decides cases — and, after Warhol, tightly linked to Factor 1. The question is market substitution: does the use stand in for the original, or for a market the owner could reasonably license?
It substitutes. If the use fills demand for the original — or a licensing market, including derivatives — Factor 4 cuts hard against.
Harm from criticism doesn't count. A parody that kills demand by mocking the work isn't cognizable harm (Campbell) — only substitution is.
Harper & Row v. Nation
471 U.S. 539 (1985)
The facts
The Nation obtained a stolen manuscript and published ~300 words from President Ford's unpublished memoir about the Nixon pardon — scooping Time's licensed serialization.
The holding
Time backed out, refusing the remaining $12,500 of a $25,000 deal. The use usurped the market for first publication — market effect, the Court said, being "the single most important element." (Campbell later softened that into a holistic weighing of all four.)
Why it matters
Real, provable market substitution — a lost licensing deal — is the core of Factor 4. The scoop didn't borrow mere facts; it took the memoir's first-publication value, the one thing the author hadn't yet sold.
Weighed, not scored
No factor is automatically decisive and none is ignored. But in practice the analysis pivots on Factors 1 and 4 — and Warhol welded them together: a use that shares the original's purpose and is commercial is, almost by definition, a market substitute, which is why a "transformative-looking" work can still lose.
Clear these out
2023 · the modern shift
In Andy Warhol Foundation v. Goldsmith (7–2), the Court looked at Warhol's "Orange Prince" — based on Lynn Goldsmith's photograph — as licensed to a magazine. The Foundation's use and Goldsmith's original shared "substantially the same purpose": illustrating stories about Prince. And the use was commercial.
So Factor 1 favored Goldsmith, even though Warhol added his own expression. New meaning, by itself, does not make a use transformative.
It is the cleanest illustration of the whole deck: Factors 1 and 4, pulling the same direction, can override striking artistic transformation.
Warhol Foundation v. Goldsmith
598 U.S. 508 (2023)
The holding to remember
"If an original work and a secondary use share the same or highly similar purposes, and the secondary use is of a commercial nature, the first factor is likely to weigh against fair use."
The narrow point
The Court ruled only on the specific commercial licensing use — not on Warhol creating the art, or on museum display.
Your turn · learn by doing
Rate the four factors as you read each use, then reveal how courts actually weigh it. The Scenario buttons move between cases — arrow keys still drive the deck.
Illustrative model only — these reflect the doctrine and leading cases, not legal advice. Real disputes turn on detail no widget can capture.
Test yourself
Trust surface
Every load-bearing claim here traces to primary law or a Supreme Court opinion. This deck explains a doctrine; it is not legal advice, and fair use is decided one case at a time. For a real question, consult counsel.