Contract Interpretation: Plain Meaning, Ambiguity, and Parol Evidence

A court asked to settle the meaning of a contract term starts with the words the parties wrote. If those words have one reasonable meaning when the document is read as a whole, that is the meaning the court applies, and evidence about what someone meant to say or thought had been agreed is kept out. Outside evidence gets in through a single door: ambiguity. If the term is genuinely open to more than one reasonable reading, the court may look beyond the page. Contract interpretation is state law, and states disagree about how easily that door opens.

1.Where a court starts

The first step is not a search for intent in the abstract. It is a reading of the document. Words get their ordinary meaning unless the contract defines them or a trade gives them a special sense. Defined terms control over dictionary meanings, specific provisions control over general ones, and a reading that leaves every clause with work to do beats one that makes a paragraph meaningless.

Courts also read the agreement as a single instrument. A clause is interpreted against the recitals, the definitions, the exhibits, and anything incorporated by reference. A poorly drafted schedule can undo a carefully drafted body.

That framework is shared by almost every state, but it is state common law, not a national code. The general contract overview at Cornell LII describes the shared vocabulary; the details vary. Where a dispute could be filed in more than one place, the answer may turn on the drafting covered in choice of law and forum selection clauses.

2.What makes a term ambiguous

Ambiguity is not the same as disagreement. Two sides always read a disputed clause differently; that is why they are in court. A term is ambiguous only when it is reasonably open to more than one meaning. Whether that threshold is met is usually decided by the judge as a question of law, which is why so many contract cases end on summary judgment rather than at trial.

Courts often separate two kinds. A patent ambiguity shows on the face of the document — the contract says the price is in Exhibit B, and Exhibit B lists two prices. A latent ambiguity is invisible until facts are supplied, as when delivery is due at "the Fairview warehouse" and the buyer runs two of them. The latent kind requires outside evidence just to be shown, and most states allow that limited showing.

• The language must be capable of more than one reasonable reading, not merely inconvenient for one side.
• The reading offered must survive when the whole document is considered, not the sentence alone.
• An undefined term is not automatically ambiguous if ordinary usage supplies a clear sense.
• Silence is a gap to be filled by default rules, not an ambiguity to be construed.

When a term stays ambiguous after the usual tools are exhausted, some courts apply contra proferentem and construe the language against the drafter. It is generally a tiebreaker of last resort, weighing most where one side had no realistic chance to negotiate the wording.

3.The parol evidence rule

The parol evidence rule is not really a rule about evidence. It is a rule about which document counts. Where the parties adopted a writing as the final statement of their agreement, prior agreements and contemporaneous oral agreements that contradict it are not enforceable, however well they can be proved. The overview at Cornell LII on the parol evidence rule sets out the basic shape.

Three points do most of the work. First, the rule reaches backward and sideways, not forward: it says nothing about agreements made after signing, so a later modification is a separate question. Second, it bars contradiction, and depending on how complete the writing is, may also bar consistent additional terms. Third, it applies only if the writing is integrated.

Integration is where merger clauses matter. A clause stating that the document is the entire agreement and supersedes all prior discussions is strong evidence of a complete writing. Most states treat it as strong but not conclusive, and some still ask whether the alleged side agreement would naturally have been left out.

Caution: The parol evidence rule does not bar a claim that the contract was procured by fraud, that a party lacked capacity, that a condition precedent never occurred, or that the writing should be reformed because it does not reflect the actual bargain. Those claims attack the validity of the writing rather than ask a court to read it differently, and states differ on how far a merger clause can limit a fraud claim.

4.The state split on context

Here the map divides. One group of states follows a four corners approach: if the judge finds the language clear within the document itself, extrinsic evidence is not consulted at all and the inquiry ends. Texas courts, for example, have long framed the question as whether the contract can be given a certain and definite meaning as written.

California took a different path. In Pacific Gas & Electric Co. v. G.W. Thomas Drayage & Rigging Co. (1968), the California Supreme Court held that a court may provisionally consider extrinsic evidence of the surrounding circumstances to decide whether the language is reasonably susceptible to the meaning a party urges, rather than deciding clarity from the page alone. A California court may therefore hear the context before ruling that a term is unambiguous.

Neither approach is the national rule. Most states sit between them, and a state's position can differ across contract types — an oil and gas lease, an insurance policy, and a software license may not be read with equal strictness even within one state. Drafting therefore matters as much as litigating, a theme that also runs through liability caps and consequential damages waivers.

Four corners approach

The judge decides clarity from the document alone. If it is clear, context is never heard. Predictable, and unforgiving of sloppy drafting.

Contextual approach

The judge may look at surrounding circumstances first to see whether the urged meaning is reasonably possible. More accurate in hard cases, less predictable, and more expensive to litigate.

5.Trade context and the parties' conduct

Some outside evidence is not really outside. Under the Uniform Commercial Code, adopted with variations by each state for sales of goods, the agreement includes course of performance, course of dealing, and usage of trade. Those sources can explain or supplement terms even where the writing is not ambiguous, though they cannot flatly contradict express terms. The Uniform Commercial Code text at Cornell LII collects the articles, and the Uniform Law Commission publishes the model text states enact.

Course of performance is how the parties behaved under this contract; course of dealing is how they behaved under earlier contracts with each other; usage of trade is what the practice means in the industry. Express terms generally control, then course of performance, then course of dealing, then usage of trade, though states vary in how mechanically they apply that order.

In a goods dispute the fight is often not what a term means but whose form supplied it, a problem covered in the battle of the forms that can decide a case before interpretation begins. Where the remedy sought is performance rather than money, definiteness becomes critical, as explained in specific performance and injunctions.

6.Questions this raises

Does a merger clause make prior emails irrelevant?

Usually it makes them unenforceable as separate promises, which is not the same as irrelevant. Emails may still be offered to show a term is ambiguous, to establish trade usage, or to support a fraud or reformation claim. In a four corners state, if the judge finds the language clear, the emails never reach the fact-finder. In a contextual state, they may be considered first.

Who decides whether a term is ambiguous, the judge or the jury?

In most states the judge decides ambiguity as a question of law. If the judge finds no ambiguity, the judge also supplies the meaning and the case can end on summary judgment. If ambiguity is found and the extrinsic evidence conflicts, the fact-finder resolves the meaning at trial. The line between those two steps is drawn differently across states, so the governing state controls.

Can the parties agree in advance how a term should be read?

Partly. Defined terms, interpretive provisions, and a clause setting an order of precedence among documents are all enforceable and are read like any other term. What the parties cannot reliably do is direct a court to ignore ambiguity or to skip a fraud inquiry. A clause instructing the court not to apply contra proferentem is generally respected between represented commercial parties.

7.Working through a disputed term

Read the whole document first, including exhibits, schedules, and anything incorporated by reference, and mark every clause touching the disputed subject. Identify the governing state, then confirm whether it leans toward the four corners approach or the contextual approach, because the answer determines what evidence will ever be seen.

Next, state both readings in one sentence each and test whether the losing clause is left with any work to do under either. Locate the integration clause, note whether it lists documents that survive, and check for a side letter signed the same day. Then collect the conduct evidence — invoices, acceptance records, prior orders between the same parties — which may come in where negotiation history does not.

Finally, separate interpretation from validity. If the real complaint is that a party lied during negotiations, or that the writing does not match what was agreed, that is a fraud or reformation theory and it follows different rules. Court structure and filing procedure are described at the United States Courts website. No summary substitutes for the rule in the state that governs the agreement.