Milkovich v. Lorain Journal Co.: The Provably False Test

In Milkovich v. Lorain Journal Co., 497 U.S. 1 (1990), the Supreme Court held 7–2 that the First Amendment does not create a separate privilege shielding statements labeled “opinion” from defamation liability when those statements imply a false factual claim that can be proven true or false. Chief Justice Rehnquist wrote the majority opinion, rejecting the notion that prefacing an accusation with “in my opinion” strips it of factual meaning.¹

The Column That Started It

The dispute began with a 1974 brawl at an Ohio high school wrestling match involving Maple Heights High School. The Ohio High School Athletic Association held a hearing and placed the wrestling team on probation. Coach Michael Milkovich testified at that hearing, as did the school superintendent. A court later overturned the probation.

Sports columnist Theodore Diadiun then wrote in the News-Herald that anyone who attended the hearing “knows in his heart” that Milkovich “lied at the hearing,” and framed the episode as a lesson in how someone can get away with “the big lie.”² That accusation was not a small one. Perjury under federal law carries up to five years in prison.³

Milkovich sued for defamation. Lower courts sided with the newspaper on the theory that Diadiun’s statements were constitutionally protected opinion, drawing on a widely shared assumption at the time that “opinion” enjoyed its own First Amendment privilege insulating it from defamation claims.

Why the Court Refused to Create an Opinion Privilege

The Supreme Court rejected that assumption. Rehnquist’s majority opinion reasoned that existing First Amendment defamation doctrine already gave speakers enough breathing room without adding a new categorical shield for anything called “opinion.” The Court pointed to four protections already in place:

Public officials and public figures must show actual malice under New York Times Co. v. Sullivan, meaning the speaker knew the statement was false or acted with reckless disregard for its truth.⁴
In cases involving matters of public concern, the plaintiff bears the burden of proving the statement false under Philadelphia Newspapers, Inc. v. Hepps.¹
Loose, figurative language that no reasonable listener would take as a literal factual claim is protected as rhetorical hyperbole.
Appellate courts independently review the record in defamation cases rather than deferring to the trial court.

Given those safeguards, the Court saw no need for an additional privilege. Creating one would let anyone dodge accountability by prefacing a specific accusation with “I think.” As the opinion put it, couching “Jones is a liar” as opinion does not strip the statement of its factual implications.²

Justice Brennan, joined by Justice Marshall, dissented. Brennan accepted the majority’s framework but argued that Diadiun’s column, read as a whole, would strike readers as the columnist’s personal conclusion rather than an assertion of undisclosed facts. The disagreement was about how the test applied here, not whether the test was correct.⁵

The Provably False Test

The rule that came out of Milkovich asks two questions. Would a reasonable reader interpret the statement as implying an assertion of objective fact? And can that implied factual assertion be proven true or false? If both answers are yes, the statement can support a defamation claim even if it is framed as commentary.²

Applied to Diadiun’s column, the reasoning worked like this. Saying Milkovich “lied at the hearing” implied that his sworn testimony contradicted what actually happened at the wrestling match. That implied claim could be tested against a factual record: hearing testimony, later court testimony, witness accounts. Because the perjury accusation was “sufficiently factual to be susceptible of being proved true or false,” it fell outside the zone of protected opinion.²

The practical difference is easier to see with an example. “The mayor did a terrible job” is a value judgment no court can prove or disprove. “The mayor lied about crime statistics during last night’s press conference” implies specific verifiable facts: that particular statistics were presented and were knowingly false. The second statement carries the factual connotation that Milkovich puts within reach of defamation law.

Who Has to Prove What

Milkovich operates on top of an earlier burden-shifting rule. Under Philadelphia Newspapers, Inc. v. Hepps, when a defamation claim involves a matter of public concern, the plaintiff must prove the statement false. The old common-law presumption that defamatory statements are false no longer applies. If the evidence cannot resolve truth or falsity, the defendant wins.¹

So a plaintiff cannot just point at a column they dislike. They have to identify the specific factual assertion inside the commentary and prove it was false. Public officials have the added burden of showing actual malice. This layered structure is exactly what convinced the majority a separate opinion privilege was unnecessary.

Context Shapes the Analysis

Courts do not read the challenged words in isolation. They look at the type of publication, the conventions of the medium, the full text surrounding the statement, and whether a reasonable audience would understand the language as factual or rhetorical. A political cartoon calling a senator a “puppet” reads differently from a news report calling the same senator a “puppet” of a specific lobbyist. The first is metaphor. The second implies factual knowledge of an undisclosed relationship.

What Milkovich Left Protected

The decision did not eliminate the opinion defense. Genuine opinion, rhetorical hyperbole, parody, and loose figurative language remain fully protected. The condition is that the speech cannot reasonably be understood as stating actual facts about a real person.

Rehnquist’s opinion drew on Greenbelt Cooperative Publishing Assn., Inc. v. Bresler, where a newspaper reported that people at a public meeting had called a developer’s negotiating tactics “blackmail.” The Court found no defamation because the audience understood the word as a vigorous epithet, not a literal accusation of the crime.⁶ The same logic protected the crude parody at issue in Hustler Magazine, Inc. v. Falwell, since no reasonable reader would take it as describing real events.

The dividing line under Milkovich is whether the language, in context, implies a false and verifiable factual claim. Calling a political opponent “the worst governor in history” is subjective and unverifiable, so it is protected regardless of how harsh it is. Writing that a governor “falsified budget reports to hide a deficit” is specific, factual, and provable, so it is actionable if false.

Diadiun’s column fell on the wrong side of that line. The Court found the language was not “loose, figurative or hyperbolic” in a way that would signal readers the columnist was just venting. The tone and specificity communicated a serious charge of perjury, the kind of claim a jury can evaluate against a factual record.²

How the Test Applies Today

Milkovich predates the internet as a mass medium, and its test has only become more relevant. Online reviews, social media posts, and comment sections constantly produce statements that blur the line between opinion and factual assertion. The provably false factual connotation test applies to a Yelp review or a tweet the same way it applied to Diadiun’s column.

The test also intersects with anti-SLAPP laws, which allow defendants sued over speech on matters of public concern to file a special motion to dismiss early in the case. If the challenged statement carries no provably false factual connotation, Milkovich itself becomes the basis for tossing the suit out, and many anti-SLAPP statutes require the losing plaintiff to pay the defendant’s attorney fees. A plaintiff who sues over pure opinion or hyperbole risks not just losing but paying the other side’s legal bills.

Why the Case Still Controls

More than three decades on, Milkovich remains the framework courts use to separate protected opinion from actionable defamation. Its refusal to create a categorical opinion privilege means writers and speakers cannot rely on disclaimers or “I think” prefaces to escape accountability for false factual claims that damage someone’s reputation.¹ At the same time, sharp criticism, satire, hyperbole, and genuine expressions of personal belief keep their full First Amendment protection. The principle is simple even when the line is hard: say what you think, but do not dress up a provable lie as your opinion.

1
Justia. Milkovich v. Lorain Journal Co., 497 U.S. 1 (1990)
2
Cornell Law Institute. Milkovich v. Lorain Journal Co., 497 U.S. 1 (1990)
3
Office of the Law Revision Counsel. United States Code Title 18 – 1621 Perjury Generally
4
Justia. New York Times Co. v. Sullivan, 376 U.S. 254 (1964)
5
Cornell Law Institute. Milkovich v. Lorain Journal Co. – Dissent
6
Justia. Greenbelt Co-Op. Publ. Assn., Inc. v. Bresler, 398 U.S. 6 (1970)