The Row over Rowe: What to Do with Facts “Between” the Record

Judicial Realists

The advocates of independent judicial research, perhaps most vocally represented by Judge Posner, argue for what they have termed “judicial realism.” A judicial realist, Judge Posner contends, is generally defined as someone who understands the limits of the adversary system, strives to have decisions “make sense” to laypeople, puts function before form in interpreting texts (be they contracts, statutes, regulations, or constitutional provisions), and has a pragmatic approach to the law. Realists shun outcome-determinative analysis, seeking instead to understand the likely legal and real-world consequences of their decisions. Richard A. Posner, Judicial Opinions and Appellate Advocacy in Federal Courts—One Judge’s Views, 51 Duq. L. Rev. 3, 9 (2013). “The judicial realist is a consequentialist” who believes that “[w]e need evidence-based law, just as we need evidence-based medicine.” Id. at 16. And if the lawyers appearing before judges do not provide all the facts needed to fully understand a dispute—a problem that is only growing with the world’s increased complexity—judges must turn elsewhere to fill the gaps in their knowledge. Id. at 15.

This belief tends to lead judicial realists to look beyond the evidence offered by the parties in order to fully appreciate the nature of a dispute. It also often makes them critical of judges who confine their understanding of a case to the facts and evidence offered into the record by the parties, believing this to be either lazy or a way of evading difficult subjects and unpopular decisions. Id. at 18. Realists contend that judges should acknowledge that they do, in fact, make law; they do not passively interpret it. For that reason, judges should do all they can to understand the arguments and subject matter before them in order to shape the best possible solution. If gaining that knowledge requires independent judicial research, say by viewing a relevant location on Google Street View or learning about a relevant ailment by consulting medical texts, so be it.

Realists often emphasize that they limit their external research to “legislative facts”; that is, general rather than case-specific information. They say they do not independently explore “adjudicative facts”—those concerning the immediate parties as well as their actions and circumstances—setting them apart from judges in inquisitorial legal systems. But the line between these kinds of facts can quickly become “a bit blurry.” Layne S. Keele, When the Mountain Goes to Mohammed: The Internet and Judicial Decision-Making, 45 N.M. L. Rev. 125, 131–32 (2014). Nonetheless, realists claim they can draw the distinction and that their approach promotes accurate fact-finding and the public’s faith in a commonsense system.

Judicial Traditionalists

On the other side of this debate are those we might call “judicial traditionalists.” They defend the basic assumptions underlying our adversarial system, seeing it as the best method we have to discover truth. Allegations and claimed experts are properly vetted by cross-examination and competing evidence, rights to confront witnesses and have claims heard by a jury are protected, and due process is preserved because each side has an opportunity to be heard. Independent judicial research, they argue, undermines these essential safeguards by cloaking an otherwise transparent process in secrecy and depriving parties of the ability to test the accuracy of the judge’s findings and the reliability of the judge’s sources. Id. at 138, 154–59.

To traditionalists, the role of a judge is to “call balls and strikes, and not to pitch or bat.” Confirmation Hearing on the Nomination of John G. Roberts Jr. to be Chief Justice of the United States: Hearing before the S. Comm. on the Judiciary, 109th Cong. 56 (2005). This umpirical role, they assert, is a humble one that recognizes relevant competencies. Judges are not doctors, engineers, or scientists, among other things, and should take that to heart. A judge conducting his own research into a medical condition may very well misinterpret what he finds or allow unconscious bias to affect what is found. Further, judges must do all they can to stay within the boundaries of their limited role, to avoid not only substantive mistakes in unfamiliar territory but also prejudice and unfairness. Keele, supra, at 138–46.

Traditionalists maintain that the roles of judges and lawyers and the procedures they employ have evolved over centuries to their current state for good reason and should not be so easily or arrogantly discarded. They think independent judicial investigation, far from promoting fairness and the public’s faith in the system, undermines both and in doing so inevitably erodes the rule of law. See, e.g., Elizabeth G. Thornburg, The Lure of the Internet and the Limits on Judicial Fact Research, 38 Litig. (Summer/Fall 2012), at 41. Their view is reflected in Rule 2.9(c) of the ABA’s Model Code of Judicial Conduct (2007), which instructs that a “judge shall not investigate facts in a matter independently, and shall consider only the evidence presented and any facts that may properly be judicially noticed.”

These classifications are not absolute. Many judges fall somewhere in between. Some judges have few qualms about researching background facts (silently or otherwise) to understand the context of a dispute, but they draw the line at researching facts that might decide the outcome. As mentioned above, however, that line is not always as clear as we might imagine.

The Rowe Decision

Last year, Rowe crashed into the ongoing debate between realists and traditionalists. Jeffrey Rowe, an Indiana inmate, alleged that prison personnel denied him access to medication necessary to treat his gastroesophageal reflux disease by allowing him to take a drug only at times when it was least effective or by denying it altogether. The district court granted summary judgment to the defendants, dismissing Rowe’s civil rights claim. The court relied in part on the affidavit of the prison doctor, a named defendant, who opined that Rowe’s condition was not serious and that the restrictions on when he could take the medication did not interfere with its effectiveness. Rowe, 798 F.3d at 623–26. Rowe, representing himself, apparently did not produce any expert evidence countering the doctor’s opinion.

The Seventh Circuit reversed, and from the beginning of his opinion, Judge Posner cited to evidence outside the record. He referenced sources from the National Institutes of Health, the Mayo Clinic, and WebMD to explain the potential seriousness of Rowe’s condition. See id. at 623–28. He cited Wikipedia, Mayo, the Physicians’ Desk Reference, and the drug manufacturer’s website to describe the medication at issue and the manner in which it should generally be taken. See id. These sources recommended taking the drug in a way that contradicted the prison doctor’s conclusions on dosage and timing. Judge Posner referenced the American College of Preventive Medicine’s website, and another called “healthgrades.com,” to call into question the expertise of the prison doctor—at one point commenting that the doctor was “a frequent defendant in prisoner civil rights suits.” Id. at 625.

In calling on these sources, Judge Posner recognized that “we may be thought to be ‘going outside the record’ in an improper sense.” Id. at 628. He nonetheless rejected the view that “judges should confine their role to choosing between the evidentiary presentations of the opposing parties, much like referees of athletic events.” Id. He noted that it was important to distinguish between judicial web searches for mere background information that will help the judges and the readers of their opinions understand the case, web searches for facts or other information that judges can properly take judicial notice of . . . , and web searches for facts normally determined by the factfinder after an adversary procedure that produces a district court or administrative record.

Importantly, Judge Posner chose not to invoke Federal Rule of Evidence 201 to take judicial notice of the evidence he culled from the web, concluding instead that medical information gleaned from reputable sources “tends to fall somewhere between facts that require adversary procedure to determine and facts of which a court can take judicial notice.” Id. at 628–29. Then he asked several questions:

There is a high standard for taking judicial notice of a fact, and a low standard for allowing evidence to be presented in the conventional way, by testimony subject to cross-examination, but is there no room for anything in between? Must judges abjure visits to Internet web sites of premier hospitals and drug companies, not in order to take judicial notice but to assure the existence of a genuine issue of material fact that precludes summary judgment? Are we to forbear lest we be accused of having “entered unknown territory”? . . . Shall the unreliability of the unalloyed adversary process in a case of such dramatic inequity of resources and capabilities of the parties as this case be an unalterable bar to justice?

Id. at 629.

Not surprisingly, Judge Posner concluded that there is room for something “in between” facts introduced into the record through conventional means and facts judicially noticed. In doing so, however, he added one final twist, emphatically stating that “[t]his is not the case in which to fetishize adversary procedure in a pure eighteenth century form.” Id. “Pure adversary procedure works best when there is at least approximate parity between the adversaries,” a condition the court found lacking in this case. Id. at 631. Judge Posner thus implied the existence of an additional consideration to his new category of evidence: whether the parties are on a remotely equal footing.

To be sure, Judge Posner was careful to state that the court was not deeming its Internet evidence conclusive—merely that it is a useful tool to underscore the existence of a genuine factual dispute created by the conventional evidence introduced below; namely, Rowe’s reported pain. Id. at 629. Judge Rovner concurred but insisted that the case could be decided without outside research, which “merely illuminate[d] the factual dispute.” Id. at 635 (Rovner, J., concurring).

Writing in dissent, Judge Hamilton branded Judge Posner’s opinion as “an unprecedented departure” from the proper role of an appellate court that ran contrary to long-established law and raised a host of practical problems. Id. at 636 (Hamilton, J., concurring in part and dissenting in part). In the first place, he objected to the use of information found on websites to contradict the only expert evidence in the record. Id. Making clear he is not a strict traditionalist, Judge Hamilton stated that he does not oppose using “careful research” to provide context and background information. Id. at 638. What’s beyond the pale is using independent factual research to find a genuine issue of material, adjudicative fact. See id. Citing a number of cases in which appellate courts reversed decisions by district court judges, jurors, and administrative law judges who had conducted their own evidentiary research, Judge Hamilton said that appellate court judges are in no better a position to go outside the record and investigate decisive facts, and had no warrant to do so. See id.

To Judge Hamilton, the foundations of our legal system rest on our belief in adversarial procedures—tools that are not available when a court conducts its own factual inquiry and bases its decision on what it finds. The only exception to this hard and fast rule is the concept of judicial notice, embodied in Federal Rule of Evidence 201, which substitutes the acceptance of a “universal truth” for the conventional method of introducing evidence. Id. at 639. There is simply no room for anything in between, including the court’s “entirely new, third category of evidence.” Id. “Before this decision,” Judge Hamilton added with a touch of snark, “American law has not recognized this category of evidence, which might be described as ‘non-adversarial evidence that the court believes is probably correct.’” Id. at 640. Without providing meaningful guidance as to how judges should conduct this research and what restrictions apply, the court’s approach invites endless expansion of the record, turns the role of the judge from neutral decision maker to advocate and quasi-expert in matters beyond the law, burdens parties with having to anticipate what might turn up in a judge’s research and somehow prepare to meet it, and creates a host of other problems. Id. at 640–42.

Other Relevant Cases

While Judge Hamilton is probably correct that Rowe “will become Exhibit A in the debate” on the use of independent judicial research, id. at 638, the decision is not entirely novel, either within or outside the Seventh Circuit. For example, in Mitchell v. JCG Industries, Inc., 745 F.3d 837 (7th Cir. 2014), another Posner opinion that garnered widespread attention, one judge had his staff don and doff protective clothing to see how long it took them to do so, a question at the heart of the plaintiffs’ claim that they were entitled to compensation for the time they spent at work changing. The court insisted that this “experiment,” which perhaps would have been better performed by the parties below, did not affect the outcome of the case. Id. at 842. But most observers were skeptical.

In Bourgeois v. Peters, 387 F.3d 1303 (11th Cir. 2004), the Eleventh Circuit cited Wikipedia to determine the Department of Homeland Security’s “threat level” at a given time, using that information to reject an argument that law enforcement officials could force protestors to submit to a metal detector search before entering a protest site.

And in United States v. Bari, 599 F.3d 176, 181 (2d Cir. 2010), the Second Circuit refused to find reversible error when a judge, during the course of a probation revocation hearing, independently researched factual questions on the Internet to confirm his intuition on a “matter[] of common knowledge.” Specifically, the judge found that a yellow hat discovered in the garage of the defendant’s landlord was the same type worn by the perpetrator of a recent bank robbery, while a Google search confirmed that this style of hat was rare. Id. at 178. The district court judge found that this was the “strongest piece of evidence” linking the defendant to the crime. Id.

On the other hand, the Supreme Court recently entered the fray in Maryland v. Kulbicki, 136 S. Ct. 2 (2015). The Court reversed a decision of the Maryland Court of Appeals, which had conducted its own Internet research into the continued scientific validity of comparative bullet lead analysis used in a murder case. The Maryland court had set aside the defendant’s conviction and ordered a new trial after finding defense counsel ineffective because they failed to find a report available on the Internet coauthored by the same FBI agent who testified against the defendant highlighting methodological flaws in the bullet analysis. The Supreme Court reversed in a per curiam decision, finding it unreasonable to expect defense counsel to provide “perfect advocacy” by discovering the report. Id. at 4. Still, the Court did not directly criticize the practice of Internet research.

If Rowe Stands

If the decision in Rowe stands, it and future similar cases will present a host of issues that require frank examination. This is not because the use of independent judicial research in Rowe was inherently right or wrong, but because of the uncertainty in expectations and outcomes it creates. Judge Posner has squarely acknowledged the criticism he and other “judicial realists” have drawn for engaging in independent judicial research. His response: “lawyers should do the Web research and spare me the bother.” Judicial Opinions, supra, at 11. While oddly endearing, this is obviously not a sufficient answer. Whether one is a realist who agrees with that sentiment or a traditionalist who does not, predicting what topics might pique a trial or appellate judge’s curiosity is easier said than done.

At the trial court level, attorneys are primarily concerned with controlling what information is heard and considered by the fact finder. Lawyers are focused on getting their facts into the record and, in some circumstances, keeping the other side’s facts out. If judges are free to conduct their own research into factual issues, attorneys must try to predict what those facts might be and find a way to influence how the judge views them. Each side will want to put its own spin on a judge’s research, but this will be impossible if they do not know what topics will pique a judge’s curiosity, much less what the court’s independent research will uncover. Moreover, allowing judges to conduct their own research when juries are strictly admonished not to raises obvious questions about double standards. Then there are important questions about addressing authentication requirements, expert requirements, and hearsay prohibitions, among other issues.

The expectation that judges will supplement the fact-finding with their own journeys through cyberspace will likely add considerable time and expense to an already costly process. To a certain extent, trial lawyers already try to anticipate the fact finder’s points of curiosity, understanding the importance of crafting a narrative and tying up loose strings. But summary judgment briefs have page limitations, trials cannot go on forever, most clients have only so much money to spend on litigation, and attention spans are only so long. Good trial lawyers therefore also understand the importance of economy in introducing the facts they need to prove their case. That balance may well be more difficult to strike if lawyers have to predict what outside facts a judge might seize on.

District judges will face a similar problem. In Rowe, the appellate court implicitly criticized the district court for crediting the prison doctor’s supposedly expert opinion over the allegations of the inmate’s complaint because independent research would have revealed that summary judgment was inappropriate. This happened in an earlier case, Jackson v. Pollion, 733 F.3d 786 (7th Cir. 2013). In Jackson, the Seventh Circuit found that the district court correctly entered summary judgment for the defendants due to the plaintiff’s failure to produce any evidence to support his claim, but Judge Posner again criticized the magistrate and district court judges for not doing their own research into the medical condition at issue where the lawyers hadn’t done so. This followed his criticism of lawyers in general for being wary of confronting scientific and technical issues. Id. at 787–90.

In a concurring opinion, Judge William Bauer pithily responded that

many lawyers decided against medical school because of a lack of interest in the clinical aspects of medicine or a deeper interest in the less scientific aspects of the law. I was one of those who chose law as opposed to medicine. I think that the opinion made the necessary legal point when it said that the record shows that summary judgment was clearly the right decision. That’s where I would stop.

Id. at 790–91 (Bauer, J., concurring in the result).

But if the expectations expressed in Rowe and Jackson hold, trial judges will not have the luxury of stopping; they will be obliged to go further until they obtain the information they believe necessary to satisfy their colleagues on review. Moreover, as at least one commentator has noted, the medical research conducted by the appellate court in Jackson was flawed, partly focusing on a different disorder than the plaintiff’s, illustrating the dangers of judges straying too far from their legal expertise. Keele, supra, at 151–52.

Appellate practitioners face somewhat different challenges. Appellate lawyers must live with the record they have. There is generally one exception, and only one exception, for facts subject to judicial notice. Federal Rule of Evidence 201 sets a high bar for the introduction of such facts, allowing reviewing courts to take judicial notice only of those facts “generally known within the trial court’s territorial jurisdiction” and those facts that “can be accurately and readily determined from sources whose accuracy cannot reasonably be questioned.” Fed. R. Evid. 201(b). While some states have comparatively relaxed standards, particularly for legislative facts, they do not allow for the introduction of adjudicative facts by way of judicial notice. On review, lawyers will not only have to predict what extra-record facts judges may find relevant. They might also face a Hobson’s choice: cite authority that does not meet the applicable standard for judicial notice or be accused of failing to do all they could to represent the client.

Furthermore, independent judicial research presents considerable due process concerns, especially involving the opportunity to respond and be heard. Rule 201 requires courts to give parties such an opportunity, but what chance do they have to respond to independently researched “facts” that fall short of the gold standard set for judicial notice? Parties should have the opportunity to contest the truth of an ostensible fact as well as the reliability of its source. Doing so in post-judgment motions, as some have suggested, is of little practical utility because convincing a court to change its mind can be extremely difficult. This is why motions for reconsideration and rehearing have notoriously low success rates. Ideally, the opportunity to respond should therefore occur before the court considers a “fact” discovered in the course of its own research.

Supplemental briefing after the court discovers a point with at least an arguable chance of affecting the outcome of the case may provide a partial answer. After all, courts sometimes ask parties to submit such briefing when events have occurred since a case or issue was originally argued; for example, on a jurisdictional issue like mootness. The same procedure could be employed where a judge uncovers new and important facts in the course of her own investigation, either at the trial or appellate level. It would be preferable, however, for reviewing courts to remand the case for further development of the unresolved factual issue, as trial courts are better equipped to act as fact finders.

Admittedly, reliance on supplemental briefing is neither a perfect nor an elegant solution. Judges inclined to do their own factual research will be understandably reluctant to say that what they found could affect their decision. More briefing will also likely lead to greater expense. But it will at least allow courts to fairly consider the kind of situation that existed in Rowe, where a pro se party with a potentially legitimate claim had little chance of making his case due to the disparity of relative resources.

In a system of justice where so many go without needed representation, this leveling provides some relief. In such circumstances, pre-decision supplemental briefing would give the defendant an opportunity to respond to any authentication and reliability issues that may arise. It would also allow everyone, represented or not, to have their fair say before a judge makes a determination about what the evidence means. Supplemental briefing would, in other words, strike a reasonably fair and potentially workable balance, addressing many of the expectation problems such research creates. This is one possible solution to controlling the evolution of this phenomenon in a way that maintains the public’s trust in the rule of law.

Judges are driven to conduct independent research out of the best of motives, as are the judges who resist it. Regardless of whether one is a judicial realist, a judicial traditionalist, or something in between, it is difficult to deny that such investigation is almost certainly here to stay in one form or another. It requires so little effort. Rather than staking out fixed positions about the propriety of the practice itself, perhaps we should shift to developing procedural safeguards that address this new category of “evidence” and thereby work to ensure the continuing integrity of our adversarial system of justice.