Concept
After the Bubble Bursts
Effect of civil presumptions (Rule 301)
After the Bubble Bursts
Rule 301 tells the opponent what to do and leaves the burden of persuasion where it started. In a civil case, unless a federal statute or the rules provide otherwise, the party against whom a presumption is directed must produce evidence to rebut it, and the rule does not shift the burden of persuasion from the party who had it originally.1 When the opponent produces evidence sufficient to support a finding that the presumed fact does not exist, the presumption bursts and stops operating. The rule says nothing about what happens next, and a presumption problem is often unfinished at that point.
Two questions remain: what evidence does the proponent still have, and what does the judge tell the jury? The first is a question about proof. The basic facts that triggered the presumption were evidence before the presumption took effect, and they stay in the record after it drops out. Whether those facts, by their own natural force, still allow a reasonable jury to find the presumed fact decides whether the proponent has a case to submit. The second is a question about words. Jurors who hear the word presumption may misjudge what it does, so the court's description of the issue becomes part of the doctrine, and that description varies with the state of the record. The court answers the questions in that order: it decides what the evidence permits, then decides how to put the issue to the jury.
The Residual Inference
A burst presumption does not erase the proponent's evidence. Thayer pictured a presumption as a legal relationship between two facts, the basic fact that triggers it and the presumed fact it supplies.2 Rebuttal evidence ends that legal relationship, but it leaves untouched whatever logical relationship the two facts have. The disappearance of the presumption does not deprive the offered evidence of its probative value, so the natural inference that flows from the basic fact, if there is one, remains for the jury, and in some circumstances that inference is enough by itself to support a finding.3
An inference differs from a presumption in force. Without a presumption, a jury might reasonably decide from a letter and a witness's testimony that the letter is authentic, but it would not have to, because an inference is always permissive; where a presumption exists, sufficient proof of the foundational facts requires the decision.4 The opponent escapes that command by producing contrary evidence, and need not persuade the jury of the opposite.5 Once that evidence is in, the court may not instruct in terms of a presumption, but it may, where the interests of justice demand, tell the jury that a natural inference exists.6
Keeping the inference alive matters because a presumption is a shortcut across a gap that evidence could bridge on its own. If rebuttal erased the force of the basic facts along with the presumption, a proponent who had proved a logically strong basic fact would lose automatically whenever the opponent offered any contradiction, even one the jury might reject.
The deed problem comes out the same way. When the defendant meets its burden of production, the presumption of authenticity has no further effect, but evidence that merely meets that burden is not so overwhelming as to require what amounts to a directed verdict, and the jury may still infer from the evidence proving the foundational facts that the deed is authentic.7 The foundational facts for an ancient writing show why that inference is reasonable: the writing is at least thirty years old, its condition raises no suspicion, it was kept where an authentic writing would likely be kept, and interested persons have generally acted on it as genuine.8 Each fact is a reason to believe the document is real, and each keeps pointing toward authenticity after the forgery evidence arrives. The inference stays permissive, though, and a jury that finds the forgery evidence more convincing may reject authenticity.
The residual inference is only as strong as the logic joining the basic fact to the presumed fact, and some presumptions have a thinner connection than the mailed letter does. The presumption of accidental death arises when death is shown to have been sudden and violent, but a sudden violent death is as consistent with suicide or with death during criminal activity as with accident, so the basic fact alone would not support an inference of accident.9 The presumption therefore does real work for a beneficiary who bears the burden of persuasion on accident: when the facts suggest suicide, it can decide whether the case reaches the jury at all. Once the insurer produces evidence of suicide, the presumption bursts and nothing in the basic fact remains to carry the point.
O'Brien v. Equitable Life Assurance Society shows the result. A man was shot by a woman's husband while in bed with her, and his widow invoked the presumption of accidental death, which could apply if he died as the victim of a criminal assault.10 The woman testified that the decedent had forced her onto the bed and that she could remember nothing more, and the Eighth Circuit treated the presumption as destroyed because her testimony suggested that he died while perpetrating an assault.11 With the presumption gone, the bare fact of a violent death did nothing to separate accident from the alternatives that the opposing evidence now supported. Put in terms of logic, a rebutted presumption whose basic facts do not support the presumed fact leaves the rebuttal evidence standing alone, while a rebutted presumption such as receipt of a mailed letter leaves the jury free to consider the inference naturally drawn from the basic facts.12
An exam stem that says the defendant's testimony destroyed the presumption and asks whether the plaintiff therefore loses is testing this distinction. The answer turns on what the basic facts prove by themselves, because the jury, not the presumption, now decides the presumed fact. Two errors are common. Students treat a burst presumption as an automatic win for the opponent, which ignores the natural force of the basic facts, or they treat the surviving inference as binding, which forgets that an inference permits a finding without compelling one. Throughout, the burden of persuasion stays with the party who had it before the presumption arose. What the court tells jurors about the inference is the subject of the next subsection.
Jury Instructions
Rule 301 allocates burdens; it does not write jury instructions, so the court must translate the rule's effect into words that lay jurors can use.14 One idea organizes the translation. Under the general provisions of Rule 301, a jury should never be instructed in terms of a presumption, though where the interests of justice demand it, the jury may be told about the natural inference that flows from the basic facts.15
The word itself causes the trouble. Under the predominant Thayer view and the minority Morgan view alike, the court need not use the term presumption, and avoiding it is desirable because jurors may misunderstand a presumption's function and effect.16 In ordinary speech, to presume is to take something as true, which suggests that the law has already settled the point. A presumption's actual effect depends on the record: it compels a finding when nothing answers it, and it drops out once the opponent's rebuttal evidence is in. A juror who carries the first meaning into a case where only the second applies may treat the presumption as one more piece of proof to weigh against the addressee's testimony. The judge therefore describes the finding the jury must or may make. When the presumption controls, the judge simply says: you must find that the letter in question, written by A and addressed to B, was received.17
The Thayer approach yields a short ladder of instructions, one rung for each state of the record, and the mailed letter fits every rung. Which rung applies follows from what the court has already decided about the evidence.
The third rung is the only one that gives the court a choice, and the choice reflects a trade-off. Mentioning the presumption would give it a force it no longer has. Saying nothing risks that jurors who hear an addressee swear the letter never came will not realize that proof of proper mailing still counts for something. Whether to give an inference instruction rests in the court's discretion.21 When the court gives one, it phrases the instruction as permission: if the jury finds that the letter was properly addressed, stamped, mailed, and not returned, it may but is not required to infer receipt, and if it draws that inference it weighs the inference with all the other evidence in deciding whether the plaintiff has shown receipt to be more probably true than not.22 The instruction matters most where the basic facts strongly support the presumed fact, as mailing supports receipt. It adds little where they do not, as with accident inferred from a sudden violent death, because the natural inference there is too thin to name.
Had the owner stayed silent and the mailing gone undisputed, the same judge would have been right to tell the jurors that they must find the notice was received, without using the word presumption. The retailer's burden of production, and whether it was met, decides which instruction is proper; the label does no work. An exam stem that reports a sworn denial of receipt and then a presumption instruction is usually testing both errors at once.
Morgan jurisdictions cannot take this quiet route. A Morgan presumption survives contrary evidence and shifts the burden of persuasion to the opponent, so the court must tell the jury how that burden operates even after the opponent has produced evidence. When the mailing is undisputed and the opponent attacks only receipt, the jury is told that it will find the letter was received unless it believes non-receipt is more probable than receipt.23 When the opponent attacks both mailing and receipt, the instruction becomes conditional: if the jury believes the letter was accurately addressed and properly mailed, it must also find receipt unless it believes non-receipt is more probable.24 Even in a Morgan jurisdiction the court need not name the presumption; it may instruct in terms of the presumption's procedural effects.25 The Rule 301 bursting bubble approach is simpler for court and jury alike, because once the presumption bursts the jury faces an ordinary preponderance question.26
Some states prescribe the wording by rule, and they split on the label. Alaska requires the court, once the burden of producing evidence to meet a presumption is satisfied, to instruct that the jury may but need not infer the presumed fact from the proved fact, and it forbids any mention of the word presumption. Maryland takes the opposite view: once the opponent offers some evidence against the presumed fact, the matter goes to the jury, and the judge should mention the presumption so that jurors appreciate the legal recognition of a slant of policy or probability.27 Both rules answer the same dilemma, whether the word oversells the underlying policy, undersells it, or only confuses. A federal-law civil claim follows the Rule 301 ladder, but for a claim or defense on which state law supplies the rule of decision, state law governs the effect of the presumption, and with it the state's choices about instructions.28