Concept
Trustworthiness Clause
Business records
Trustworthiness Clause
Rule 803(6) admits “a record of an act, event, condition, opinion, or diagnosis” only when five conditions hold,1 and the fifth asks a different kind of question from the other four. Subsection (E) requires that “the opponent does not show that the source of information or the method or circumstances of preparation indicate a lack of trustworthiness.”2 Conditions (A) through (D) prove that the record came out of an enterprise’s routine, and they describe that routine in general terms. A regular system can still produce one unreliable entry, so the rule lets the judge look at the particular record offered. A record that satisfies (A) through (D) stays out when the opponent points to a concrete reason for distrust, drawn from where the information came from or from how and why the writing was prepared.
A business relies on its own records, and so typically do the third parties that deal with it, which gives the people who keep the figures an incentive to keep them accurate.3 The clause targets records in which that assurance fails: the information came from someone with no reason to be accurate, or the writing served a purpose other than running the enterprise. Courts apply it with that concern in view. Even for records that depart from routine, they ask less whether the record fits the literal words of the “regular practice” requirement than whether the basic concern of trustworthiness is met.4
Palmer v. Hoffman, covered above, is the source of the worry about self-serving, litigation-driven writings. A railroad’s accident report was routine, yet it was written with litigation in view, and the Court kept it out. Rule 803(6)(E) carries that worry forward as a clause the judge applies to each record, not as a categorical bar. A categorical bar would sweep in ordinary ledgers along with the suspect report, because nearly every record serves its maker’s interest to some degree. The clause lets the judge ask whether that interest actually distorted this record.
The subsections below follow the order a court uses. The first asks who must come forward with what once the proponent has laid the foundation. The other two take the settings where the clause does most of its work: records written with a lawsuit in view, and entries whose source or basis cannot be identified.
Burden on the Opponent
Subsection (E) divides the work between the parties. The proponent proves the conditions in (A) through (D), and the record then qualifies unless the opponent makes the separate showing that (E) describes. The 2014 amendment to Rule 803(6) made that allocation explicit. Its Advisory Committee note explains that once the proponent has established the stated requirements of the exception—“regular business with regularly kept record, source with personal knowledge, record made timely, and foundation testimony or certification”—the burden is on the opponent to show that the source of information or the method or circumstances of preparation indicate a lack of trustworthiness. The note gives its reason in one sentence: “the basic admissibility requirements are sufficient to establish a presumption that the record is reliable.”5 The same amendments added a similar clarification to Rules 803(7) and 803(8),6 and the absence-of-entry rules keep step with the exceptions they complement: Rule 803(10) relates to Rule 803(8) as Rule 803(7) relates to Rule 803(6).7
The foundation has already proved the features that make a routine record reliable: a regular business keeping the record regularly, a source with personal knowledge, a record made promptly, and a witness or certification vouching for the system. A record with those features is presumptively reliable, and that presumption lets the exception dispense with live testimony from every participant. If the proponent also had to prove that no hidden motive or defect existed, the proponent would be proving a negative with no natural stopping point, and the burden the exception was built to remove would return. The opponent is better placed to supply the missing piece, because a specific reason for distrust is a fact about one record, one source, or one occasion for writing, and the party seeking exclusion has the incentive to find it.
The presumption arises only after the proponent proves (A) through (D). An e-mail does not become a business record because it issued from a business or one of its employees. Many e-mails are one-time and non-routine, outside the author’s assigned duties, or personal, and they would fail the foundation before (E) ever mattered.8
The clause also limits what the opponent may point to. The distrust must come from the source of information or from the method or circumstances of preparation, and it must attach to a particular feature of this record. A remark that records may contain errors is true of every record ever kept, and an argument that the proponent has not proved accuracy asks the proponent to carry a burden the rule assigns to the opponent.
Students misapply the allocation in two directions. An answer that faults the proponent for failing to prove trustworthiness reverses the burden. An answer that makes the opponent show untrustworthiness against a record whose foundation was never laid skips a step, because the presumption exists only for records that satisfy (A) through (D).9 The next two subsections show what a real showing looks like.
Records Prepared for Litigation
The trustworthiness clause now decides whether a record written with a claim in view comes in. In Bean v. Montana Board of Labor Appeals, the Montana Supreme Court threw out a decision denying unemployment compensation to a nurse her employer had dismissed. The employer had prepared and offered an incident report saying the nurse committed misconduct and was fired for cause. The court held that the report was not part of the care center’s routine business activity and that it was made in anticipation of litigation, which made it untrustworthy.11 The Rules treat this as a question of circumstances and not as a flat bar on accident reports. Professor McCormick thought a doctor’s report or an accident report routine enough to justify admissibility.12
A litigation motive matters because of what makes routine records reliable in the first place. People who depend on recorded figures have reason to keep them accurate,13 and a report written when a claim is likely replaces that discipline with another motive. The writer may be the person who would be blamed, or may work for the party who would pay, and the account then serves the dispute more than the enterprise. The motive works through the parties: the decisions hinge on motivation and on which party is entitled to be concerned about it.14
The third question usually decides the matter, because it shows which way the litigation motive points. A report that is unfavorable to the business that prepared it, offered against that business by the opposite party, may be more trustworthy. So may the record of a business that has no part in the litigation, such as a bank’s record of a customer charged with an income tax violation.15 Physicians’ reports show the pattern. Yates v. Bair Transport, Inc. let the plaintiff introduce reports by physicians who examined him on the defendants’ behalf, but not reports by physicians he chose who examined him in anticipation of litigation.16 Korte v. New York, N.H. & H.R. Co. marks the other side of the same line, since a report came in there when the opposite party offered it.17
Preparers with no stake sit at the far end of the range. In Lewis v. Baker, the court found no similar motivation in the three people who prepared the reports, because none of them was involved in the accident or could possibly have been the target of the plaintiff’s lawsuit.18
A problem about a store’s incident reports rarely has one clean answer, and a strong answer says so. In the standard problem, the store prepares about fifty such reports a year, so they are a regular feature of its operations. Some courts have been reluctant to admit accident reports on the theory that they are prepared with litigation in mind, but modern businesses also try to acquire and analyze information about injuries, and a court would likely act within its discretion in admitting one. Legitimate grounds for rejecting it remain, and because the report in that problem depends on two exceptions, rejecting either one would exclude the entire report.19 A good answer establishes the routine, which satisfies the foundation, and then presses the variable: who wrote the report, when, and who offers it.
A further limit keeps the business-records route from becoming a way around the public-records rule. Rule 803(8) reaches “a matter observed while under a legal duty to report, but not including, in a criminal case, a matter observed by law-enforcement personnel.”20 Rule 803(6) has no such exclusion, so a prosecutor might hope to offer a barred police report as a business record. Courts generally do not allow that route around the restriction. The report the public-records rule bars is the kind prepared under adversarial circumstances and favorable to the government, and a report with those features is a record prepared in anticipation of litigation that favors its preparer, which the trustworthiness clause screens out in the same way. In United States v. Bohrer, for example, an IRS contact card prepared under adversarial circumstances was excluded under Rule 803(6) for lack of trustworthiness. Consider a prosecutor who offers a police report stating that the accused, arrested about twenty minutes after a robbery, had a Beretta 9mm pistol in a shoulder holster under his jacket.21 The officer’s observation is barred in a criminal case by the public-records rule, and on the prevailing view the business-records route is closed as well.
In a criminal case, litigation-made evidence also raises a constitutional problem that the hearsay analysis does not resolve. Melendez-Diaz v. Massachusetts declined to treat forensic analysts’ certificates as business or public records, and the Court stressed that forensic analyses are not uniformly neutral: the disciplines vary widely in technique and reliability, and common tests carry risks of subjectivity, bias, and error.22 The Confrontation Clause is developed elsewhere in the course. For this section, it is enough to see that the instinct behind Palmer, that a record made to prove a case differs from a record made to run an enterprise, reappears in the constitutional rule.
Cryptic Entries and Unclear Sources
A record that meets the foundation conditions still falls outside the exception when the opponent shows that “the source of information or the method or circumstances of preparation indicate a lack of trustworthiness.”24 The preceding subsections applied the clause to the burden of proof and to litigation motive, both questions about circumstances. This subsection applies the other half of the clause, the source of information, to an entry that states a consequential fact and never says who knew it or how. The source must have personal knowledge, but need not be the person who made the entry, so the exception contemplates multiple hearsay. A dock worker who receives widgets can tell the inventory clerk the count, the clerk can pass it to the finance staff, and each entry made on the dock worker’s information can fit the exception.25 The chain works only when the record lets a reader see it. When the record hides the chain, the judge cannot tell whether the routine touched the entry at all.
Petrocelli v. Gallison shows the problem. The plaintiffs offered references in a hospital record to a severed nerve, and the record did not say who supplied that information. The foundation was probably not the difficulty, since Dr. Swartz, whose record it was, likely recorded relevant data as a regular practice.26 The First Circuit observed that if Petrocelli or his spouse was the source, their recorded statements were arguably admissible as patient history under Rule 803(4), but it did not think that theory warranted reversal or a new trial. It added, without deciding, that the patent ambiguity about the true source of information in the hospital record could have created a lack of trustworthiness that authorized the trial judge to exclude the references on that separate ground.27
The nerve sentence could rest on several sources, and the entry does not say which. An inference from the writer’s own examination would be a diagnosis, a category the rule names.28 A statement from the patient or a spouse would add an outsider’s layer that needs its own exception. A statement from an unknown source gives the judge nothing to test, and the danger is greatest when the unexplained sentence concerns the very point in dispute. The First Circuit reached a similar result in Ricciardi v. Children’s Hospital Medical Center, where it rejected a medical record because the person who wrote a note two days after the surgery could not identify whom he got the information from.29 In both cases the missing piece is the same: no one can say who knew the fact.
This chart and Dr. Swartz’s record share the features a careless answer would seize on: a physician wrote each in a hospital’s routine, and each contains a conclusion. Those features cannot explain the different results. The emergency physician’s entry reveals its basis in the examination, while the nerve sentence conceals whatever it rested on. A student who reasons that a doctor wrote it and the rule covers diagnoses has answered the wrong question, because the rule’s coverage of opinions says nothing about whether this opinion’s basis can be identified.
The nerve sentence could have been analyzed differently if the plaintiffs had shown that the patient was the source. The hospital’s routine would then support the record layer, and the patient’s own statement would need its own basis, such as the exception for statements made for medical diagnosis or treatment.30 In Merrow v. Bofferding, a suit against a landlord, the Michigan Supreme Court admitted a doctor’s record that the plaintiff was hurt when his arm “went through the window,” an entry that presumably reflected what the plaintiff told the doctor. The case is a reminder that the proponent must show each level of hearsay fits an exception.31 A second exception needs a declarant to attach to, so a proponent who cannot say who supplied the information has nothing to match against any exception.
The fourth step is the fallback when the first three yield no answer, and it rests on the judge’s discretion instead of a categorical rule. Because the clause allows exclusion without requiring it, the same entry might come in if the proponent produced a witness who could say how the writer knew the fact.