Concept
Integrated Agreements
Parol evidence
Integrated Agreements
The parol evidence rule protects a writing only when the writing is an integrated agreement, which section 209(1) of the Restatement (Second) of Contracts defines as a writing or writings constituting a final expression of one or more terms of an agreement.1 Integration matters because a binding integrated agreement discharges prior agreements to the extent that it is inconsistent with them.2 A writing earns that power by being the parties' chosen last word on a term. Where they made no such choice, the writing has no claim to exclude proof of what else they agreed.
Integration is really three questions, asked in a fixed order. Did the parties adopt the writing as a final expression of any term? If so, did they adopt it as a complete and exclusive statement of their terms, or only as final on the terms it contains? And who decides? The second answer controls what a party may add: a partially integrated writing leaves room for other agreed terms, so evidence of a consistent additional term may supplement it, while a completely integrated writing is closed to supplementation.3 The court decides the first two questions, as questions preliminary to interpreting the writing or applying the rule.4 The tests a court uses to decide completeness come in the next major section. The three subsections here supply the vocabulary and the division of labor those tests presuppose.
Final Expression
A writing is a final expression when the parties meant it to settle their agreement on a term, so that what they said or exchanged along the way gives way to it. Parties typically arrive at terms through preliminary negotiations and then produce a writing containing the terms they have mutually adopted. That writing is the best evidence of the contract, and it displaces earlier agreements and proposals, whether oral or written.5 A document can look like a contract and still fail that description. A proposal sent while the parties are still bargaining records a stage of the process, and nothing in it shows that either side treated the matter as closed.
The label matters because the rule prefers a writing to testimony only when the writing fixes what the parties decided at the moment they decided it. Without a decision to treat the writing as final, that rationale disappears. A proposal one party sent has no better claim to represent the bargain than the conversations around it, and excluding those conversations would shut out proof of the real deal in favor of a document that never expressed it. A writing that is not an agreed final expression of any term therefore falls outside the rule altogether, and the rule has nothing to say about what else the parties may prove.6
Parties seldom announce that a document is their final word, so courts infer finality from the document itself. Section 209(3) supplies the inference: where the parties reduce an agreement to a writing that, in view of its completeness and specificity, reasonably appears to be a complete agreement, the writing is taken to be an integrated agreement unless other evidence establishes that it was not a final expression.7 A signed contract of many pages that fixes the price, the delivery date, and the remedies for breach looks like the end of a negotiation, and the presumption lets the judge treat it that way without first holding a trial on what each side privately intended. The presumption puts the burden on the party who wants to escape the writing, who must show by other evidence that the signed document was only a step toward something else, such as a placeholder pending a formal contract. Evidence of earlier agreements and negotiations is admissible for exactly that showing, so the inquiry into finality is one place where material the rule later excludes may come in.8 The word “completeness” in section 209(3) is a clue to finality. It is not a finding that the writing covers the whole deal, which is the separate question taken up next.
The two findings produce different results because the first finds no integration at all, while the second finds an integration limited to one term.9 A document prepared for some purpose other than recording the bargain sits on the first side of that line. A shipping document that one party prepares to evidence a shipment of goods may never have been intended to represent the contract, and if it was not, it is no integration, so the actual agreement may be proved from outside it.10 The reasoning is the same as for the proposal: a paper that one side generated for a logistical purpose was never adopted by both as the record of their deal, so there is no joint decision for the rule to protect.
The phrase “one or more terms” in the definition matters because finality is assessed term by term. A written agreement may fully and finally express some terms and leave others unaddressed, or it may finally express none.11 A purchase agreement that both parties understand as their definitive statement of price and delivery is an integrated agreement as to those terms, even if it says nothing final about warranties. Students make two errors here. One treats any formal-looking paper as triggering the rule, when a proposal or a one-sided shipping document triggers nothing. The other treats a signed, detailed contract as protected without asking whether it was a final expression, when its protection rests on the presumption in section 209(3) and lasts only until other evidence rebuts it. How much of the deal a final writing covers is the next question.
Complete and Partial Integration
Section 210 answers the second question by sorting integrated agreements into two kinds. A completely integrated agreement is one the parties adopted as a complete and exclusive statement of the terms of their agreement.14 A partially integrated agreement is any other integrated agreement.15 Because the second definition is residual, a final writing that the parties did not adopt as complete and exclusive falls into it automatically. The question here starts from a yes on finality and asks how much of the deal the final word covers.
Final and complete are different words, and the section depends on keeping them apart. A writing is final when the parties adopted it as their last word on the terms it contains, which is what makes earlier understandings on those terms drop away. It is complete when it covers every term the parties agreed to, and exclusive when the parties meant it to be the only place those terms are found. Finality concerns the writing's authority over what it says, and completeness concerns whether it speaks to everything. A signed contract can be final as to price and delivery and still leave warranties to some separate arrangement, and that contract is partially integrated. The reverse cannot happen, because a completely integrated agreement is by definition integrated, so a writing cannot be complete without first being final. Every writing therefore lands in one of three states, and each state answers differently the question of which earlier agreements survive:
- A writing that is not integrated does not trigger the rule at all, so the parties may prove what they agreed without its limits.16
- A partially integrated writing may not be contradicted by evidence of prior agreements or negotiations, but evidence of a consistent additional term may supplement it.17
- A completely integrated writing may be neither contradicted nor supplemented, and it discharges prior agreements to the extent they are within its scope.18
How these effects operate when an offered term is inconsistent, consistent, or doubtful is the subject of the effects section later in the chapter, and the tests courts use to choose between the two labels follow in the next major section. The task here is to see what each label does and why the law keeps two of them.
A single all-or-nothing rule would misfire somewhere. If every integrated writing excluded everything outside it, parties who wrote down some terms and left others to a separate understanding would lose the unwritten ones, though they never chose to drop them. If every integrated writing excluded only contradictory evidence, a writing the parties meant as their entire deal could be enlarged by later claims of side promises, which is just what the rule exists to screen out. Two labels let the rule follow what the parties actually did. A writing that is final only on some terms says nothing about the subjects it never addresses, so a consistent term on one of those subjects takes nothing away from what the writing settled. A writing adopted as complete and exclusive does speak to those subjects, because it says in effect that the entire deal is in the writing and no other agreements exist. Evidence of an additional term then asks the factfinder to believe in a term the writing, by its own account, does not contain.19
The Hotelco dispute shows all three states on one set of facts. If the May document was not a final expression of anything, the rule does not apply. If the court finds it the parties' final agreement on scope of work and on nothing else, it is partially integrated: it bars proof of an April agreement that conflicts with the scope-of-work section and leaves room for consistent additions.20 Whether a promise to repair plasterwork conflicts with a section describing painting is a separate argument, since Hotelco can say the section covered only painting, and that question belongs to the effects of integration. If the court finds instead that Remod and Hotelco assented to the May document as the final agreement on their entire deal, it is completely integrated, and the April plaster promise is excluded even if it contradicts nothing, because it is a consistent additional term on a subject the writing never addresses.21
The second writing's recital says in words what complete integration means: the whole deal is here, and no other agreement exists.22 How much weight a court gives such a recital belongs to the discussion of merger clauses. For now, the recital changes the classification, the classification changes the outcome, and the bench promise stays identical in both sales.
Article 2 of the Uniform Commercial Code draws the same line in different words. Section 2-202 provides that terms set forth in a writing intended by the parties as a final expression of their agreement on the terms it includes may not be contradicted by evidence of any prior agreement or of a contemporaneous oral agreement, though they may be explained or supplemented.23 A writing that is merely final thus corresponds to a partial integration, and a party may offer consistent additional terms to supplement it.24 A writing the court finds was intended as a complete and exclusive statement is the whole contract, and anything contradictory to its terms has no force.25 Students sometimes read “final” in section 2-202 to mean that the whole deal is written down. It does not. As at common law, a final writing says nothing about completeness, and it is the partial kind unless the court also finds that the parties meant it to be complete and exclusive.
Court Decides Integration
The court, not the jury, decides whether there is an integrated agreement. Section 209(2) of the Restatement (Second) of Contracts treats the question as preliminary to interpretation or to application of the parol evidence rule,29 and section 210(3) gives the court the same job in deciding whether an integrated agreement is complete or partial.30 The assignment follows from what the rule is for. The rule screens evidence before the factfinder hears it, and a screen cannot work if the factfinder must hear the evidence to learn whether the screen applies. The judge therefore acts as a filter, deciding as a matter of law whether the parties intended the writing to be a final expression and, if so, a total or only a partial one. Evidence the rule bars never reaches the jury.31
In a jury trial the judge holds an in camera hearing, outside the jury's presence. The party offering the evidence outlines what it would show, both sides argue the rule, and the judge decides admissibility.32 The question can arise long before trial. Because admissibility is a legal question, a party facing a claim or defense built on an alleged prior term may seek summary adjudication, and a ruling that the writing is completely integrated can end the claim without a trial.33
The judge's work is a sequence of gates, and the jury's work begins only when the sequence ends. At the first stage the judge decides admissibility as a matter of law, and at the second the jury evaluates whether the alleged term was in fact agreed.34
Courts do not always follow this order. Unless the intent to integrate is beyond doubt, how well the alleged term fits the writing can shape the judge's view of whether the parties meant the writing to be complete, so the first and third gates often blur.39 The sequence is still the best way to organize an answer, because it keeps the judge's questions apart from the single question that belongs to the jury.
In the first scene the judge's ruling ended the dispute, and in the second it only sent the dispute to the jury. Admission settles that the jury may hear the evidence, and whether the parties agreed to the term stays with the jury throughout. The usual error is to read an admissibility ruling as a finding for the party who offered the evidence.
Article 2 divides the work the same way. Section 2-202 is framed in terms of the court's finding: if the court finds that the writing was intended as a complete and exclusive statement of the terms of the agreement, the writing alone constitutes the contract, and to the extent it incorporates agreed terms, anything contradictory is of no force or effect.40
Deciding integration creates a puzzle. The judge must decide whether the parties meant the writing to be their final and complete word, and the best evidence of what they meant may lie in the negotiations the rule would otherwise exclude. Section 214 resolves the puzzle by listing, among the purposes for which agreements and negotiations before or at the adoption of a writing are admissible, two that serve the judge directly: establishing that the writing is or is not an integrated agreement, and establishing that an integrated agreement is completely or partially integrated.41 The other purposes in that section, which go to meaning, invalidating causes, and remedies, belong to the later treatment of evidence outside the rule. For the integration questions, the evidence comes in for a limited purpose: the judge uses it to decide whether the writing was final and complete, not to decide that the parties actually agreed to the term.
MCC-Marble shows the distinction at work. The buyer contended that the seller's standard form was not the parties' final agreement and that the operative agreement was oral, and it offered an affidavit from its president stating his intent not to be bound by the form he had signed. The affidavit was not offered to add a term to an integrated writing but to show that the form was not integrated at all, so the parol evidence rule did not bar it.42
The circle is real, and courts acknowledge it. The rule restricts the extrinsic evidence available for interpretation, yet interpretation of the writing, often in light of that very evidence, decides whether the writing should exclude it.43 Courts manage the circle by giving great weight to what the writing itself says. Even a court receptive to extrinsic evidence may refuse to consider it where the writing strongly expresses an intent to integrate, particularly through a merger clause. If the judge does consider such evidence, the evaluation is preliminary, and a jury remains the final arbiter of the term's factual truth.44 How far a judge may look beyond the writing is the dividing line between the competing tests taught in the next major section.