The parol evidence rule

New York State, the fall of nineteen twenty three. The Laths own a farm they want to sell. Across the road, on land belonging to the lieutenant governor, they keep an ice house, which they are free to remove. Mrs. Mitchill looks the farm over with a view to buying it and finds the ice house objectionable. So the Laths orally promise, in consideration of her buying the farm, to remove the ice house in the spring.
Relying on that promise she signs a written contract to buy the farm for eight thousand four hundred dollars, cash and a mortgage, with all the usual provisions. It says nothing about the ice house. She takes her deed, moves in, and spends considerable sums making the place a summer residence. The Laths do not remove the ice house and do not intend to. She sues to make them.
So here is the question. Everyone agrees the promise was made. The trial court found it, the appellate court found it, and the Court of Appeals did not doubt it. The question is whether a promise that everyone agrees was made can be enforced when the written contract of sale, signed the same day, leaves it out. Think about it before I go on.
And now the question I keep asking. What exactly did these people promise each other? On paper, a farm for eight thousand four hundred dollars. In the air, an ice house gone by spring. The parol evidence rule is the law's answer to the question of which of those two conversations is the contract.
Here is the first line for the board. When the parties have adopted a writing as the final expression of their agreement, evidence of earlier or contemporaneous agreements cannot be used to contradict it. And if the writing was meant to be complete as well as final, such evidence cannot even add to it. The lawyers' word for a final writing is integration. A complete one is a total integration.
Second line. The rule is a rule of substantive law, not a rule about believing witnesses. It does not say the oral promise is unproved. It says the oral promise is not the contract, because the parties chose a writing to be the contract. That is why the Laths could lose the argument about whether they promised and still win the case.
Third line. The whole difficulty is deciding what the writing was meant to cover. The New York test asks whether the oral agreement is one the parties would not ordinarily be expected to put in the writing. The California test asks whether it is one that might naturally be made as a separate agreement by parties in that situation, looking at the circumstances and not the paper alone. Same question, different temperament.
Fourth line, for goods. Section two two oh two. A final writing may not be contradicted by evidence of any prior agreement or of a contemporaneous oral agreement, but may be explained or supplemented by course of dealing or usage of trade or by course of performance, and by evidence of consistent additional terms unless the court finds the writing to have been intended also as a complete and exclusive statement of the terms of the agreement.
Now the case. Mitchill against Lath, New York Court of Appeals, nineteen twenty eight, Judge Andrews. He began by saying what the rule is. This requires a discussion of the parol evidence rule, a rule of law which defines the limits of the contract to be construed.
It is more than a rule of evidence and oral testimony even if admitted will not control the written contract. It does not affect a parol collateral contract distinct from and independent of the written agreement. It is, at times, troublesome to draw the line.
Then the test. Under our decisions before such an oral agreement as the present is received to vary the written contract at least three conditions must exist. One, the agreement must in form be a collateral one. Two, it must not contradict express or implied provisions of the written contract. Three, it must be one that parties would not ordinarily be expected to embody in the writing.
The ice house promise passed the first test, and perhaps the second. It failed the third. An inspection of this contract shows a full and complete agreement, setting forth in detail the obligations of each party. The presence of the ice house, even the knowledge that Mrs. Mitchill thought it objectionable would not lead to the belief that a separate agreement existed with regard to it. Were such an agreement made it would seem most natural that the inquirer should find it in the contract.
Collateral in form it is found to be, but it is closely related to the subject dealt with in the written agreement, so closely that we hold it may not be proved. Complaint dismissed. The ice house stayed. Judge Lehman dissented, and his dissent is the case's other half.
He accepted the test and disagreed about its application. The written contract, he said, covered the conveyance of land completely, but a promise to do something later on other land was so loosely bound to the conveyance that nobody would expect to find it in a deed contract.
Forty years later, the same problem in California, and Justice Traynor writing. Masterson against Sine, nineteen sixty eight. Dallas Masterson and his wife owned a ranch. In February nineteen fifty eight they conveyed it by deed to Dallas's sister and her husband, the Sines, reserving unto the grantors an option to purchase the property back within ten years for the same consideration as being paid heretofore plus the depreciation value of any improvements.
Then Dallas went bankrupt. His trustee in bankruptcy, joined by his wife, sued to enforce the option, meaning to sell the ranch to pay Dallas's creditors. The Sines wanted to prove that the whole point of the option had been to keep the ranch in the Masterson family, so that it was personal to Dallas and could not pass to a trustee. The trial court would not hear that evidence. The deed said nothing about assignability, and an option is assignable unless it says otherwise.
The Supreme Court of California reversed. Traynor started where Andrews started. When the parties to a written contract have agreed to it as an integration, a complete and final embodiment of the terms of an agreement, parol evidence cannot be used to add to or vary its terms. The crucial issue in determining whether there has been an integration is whether the parties intended their writing to serve as the exclusive embodiment of their agreement.
But how do you find that intention? Not from the paper alone. The conception of a writing as wholly and intrinsically self-determinative of the parties' intent to make it a sole memorial of one or seven or twenty seven subjects of negotiation is an impossible one. Traynor was quoting Wigmore, and it is the line that separates his test from the older one.
Here is his test. Evidence of oral collateral agreements should be excluded only when the fact finder is likely to be misled. The rule must therefore be based on the credibility of the evidence. And the standard, from the Restatement. Proof of a collateral agreement is permitted if it is such an agreement as might naturally be made as a separate agreement by parties situated as were the parties to the written contract.
Applied to the deed. The option clause in the deed does not explicitly provide that it contains the complete agreement, and the deed is silent on the question of assignability.
Moreover, the difficulty of accommodating the formalized structure of a deed to the insertion of collateral agreements makes it less likely that all the terms of such an agreement were included. A family, without lawyers, putting an option in a deed to protect it against later buyers, might naturally leave the family understanding unwritten. The evidence should have come in.
Put Andrews and Traynor beside each other. Andrews looks at the writing, in the light of the circumstances, and asks whether a reasonable inquirer would expect the promise to be there. Traynor looks at the parties and their situation and asks whether people like these might naturally have made the side agreement separately.
Andrews trusts the paper. Traynor trusts the jury, once the judge is satisfied the story is not the kind that misleads. Most states sit somewhere between them, and the Restatement's second edition leans toward Traynor.
The Code, for goods, chose its own words, and Traynor quoted them. If the additional terms are such that, if agreed upon, they would certainly have been included in the document in the view of the court, then evidence of their alleged making must be kept from the trier of fact. Would certainly have been included. That is a narrower exclusion than either judge's test. Under Article Two, the writing keeps out only the terms that could not honestly have been left out.
And the Code adds something the common law resisted. Even a final writing may be explained or supplemented by course of dealing or usage of trade or by course of performance. What the parties did before, what the trade does, and what the parties did under this very contract all come in to explain the words, whether or not the words look complete. That is next lecture's subject.
Now let's change one fact. The Laths' promise about the ice house was not oral. It was in a signed letter from them to Mrs. Mitchill, dated the day before the contract. Does the parol evidence rule keep the letter out? Choose an answer before I go on.
Most people say the rule is about oral promises, so a letter is safe. Half right. The rule reaches prior agreements whether oral or written. The name is misleading. The question is still whether the contract of sale was the final and complete expression, and if it was, the earlier letter is displaced just as the conversation was.
What the letter changes is the evidence, not the rule. It makes it much easier to argue that the parties treated the ice house as a separate deal, because they wrote it separately.
Change one fact again. The Masterson deed had said, this option is assignable. Choose.
No evidence comes in, under any test. The Sines wanted to prove a term that would contradict an express term of the writing, and that is the one thing the rule forbids without exception. Traynor's test loosens the rule about adding terms. It does not touch the rule about contradicting them. Andrews said it as his second condition, and Traynor said it when he said parol evidence cannot be used to add to or vary the terms of an integration.
Change one fact a third time. The Mitchill contract had contained a merger clause, a sentence saying that the writing contains the entire agreement of the parties and there are no other understandings. Choose.
Most people say that settles it, and under Andrews it does. Under Traynor it very nearly does. He said the instrument may state that there are no previous understandings or agreements not contained in the writing, and thus express the parties' intention to nullify antecedent understandings.
He added that the collateral agreement must still be examined to see whether the parties meant the clause to reach it. But a merger clause is the strongest evidence of total integration there is, and a party who signs one has a hard road to a jury.
Here is what people get wrong here, and why it is tempting. The first mistake is thinking the rule keeps out all oral evidence. It keeps out evidence of prior and contemporaneous agreements offered to contradict or add to an integrated writing.
It has nothing to say about later agreements, about evidence that no contract was formed, about fraud, or about the meaning of the words that are there. The second mistake is calling it a rule of evidence. It is a rule about which agreement is the contract.
The third mistake is forgetting that the rule protects contradiction absolutely and addition only sometimes. Always ask first whether the offered term contradicts the writing. If it does, stop. If it merely adds, ask whether the writing was complete, under whichever test your state uses.
Here is the rule, in one breath. When the parties have made a writing the final expression of their agreement, no earlier or contemporaneous agreement may contradict it, and if the writing was also meant to be complete, none may add to it, the completeness being judged in New York by whether the parties would ordinarily have put the term in the writing and in California by whether such parties might naturally have made it separately.
And for goods, a final writing keeps out only terms that would certainly have been included, and always yields to course of dealing, usage of trade and course of performance. Now, five questions.