ABA Guidance on Judicial Internet Research: Ethics, Due Process, and the Murky Law of Judicial Notice


In an age when smartphones are ubiquitous, and the ability to seek the answer to any question at any time has almost made “Google” a generic term, the paper-based information resources of years past have given way to voice search and virtual home assistants. Thus, despite the legal profession often lagging behind other industries on the tech front, its institutions are slowly being transformed by this ready access to limitless information, including within the judiciary.

It is perhaps for this reason that late last year, the American Bar Association (ABA) Standing Committee on Ethics and Professional Responsibility (the “Committee”) issued an advisory opinion on the subject of internet research by judges, or more specifically, when such research is permissible and when it is not. The opinion represents a renewed assertion by the ABA that judicial use of independent internet research presents challenges that are distinct from those associated with other extrinsic sources of information, which have long found their way into judicial reasoning.

The ABA’s recent opinion provides a clear explanation of the legal framework informing its conclusions, though its guidelines revolve around the somewhat nebulous law of judicial notice. Possibly recognizing the practical difficulties of complying with rules that are anchored to such an arguably messy body of law, the Committee provides a series of hypotheticals and questions that members of the judiciary may refer to in determining the propriety of running a case-related Google search at any given time. It also offers a more cohesive overview of the legal doctrines implicated by judicial use of internet research than much of the relevant legal authority has done to date, acknowledging the intersection of ethical, evidentiary, and constitutional questions at issue in this context. This sort of guidance is especially critical as the role of judicial internet research in legal proceedings continues to expand, bringing with it important consequences for core legal system values including due process, transparency, and an impartial judiciary.

A New Opinion Interpreting Existing Rules

Formal Opinion 478, issued in December 2017, warns that judges may risk violating the ABA’s Model Code of Judicial Conduct (“Model Code”) by going online and independently researching facts or parties pertinent to cases pending before them. The Committee explains that such information generally falls into the category of ex parte communications, which under Model Code Rule 2.9(A) are improper except in very limited circumstances lest they create the appearance of bias.

While Rule 2.9’s broad prohibition on ex parte communications has long been established, it wasn’t until 2007, with the internet’s use as a quick reference tool becoming commonplace within the halls of justice, that the Model Code was amended to add Rule 2.9(C), which provides 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.”

This amendment, which served to move the proscription on independent judicial research from a comment to the black letter text of the rule, highlighted the ABA’s growing concern with judges’ use of the internet to potentially (if inadvertently) conduct fact-finding outside the confines of the adversarial process. Indeed, Comment 6 to Rule 2.9 specifically states that this prohibition includes information procured from electronic, or internet, sources.

Independent Research and Judicial Notice

The addition of Rule 2.9(C) also placed the law of judicial notice squarely at the center of the debate regarding whether and to what extent judges may independently seek case-related information on the internet. Interestingly, judicial notice was not always part of Rule 2.9, but during the drafting process it was added in response to suggestions that judges be explicitly permitted to research general “background” information for a case online. See Elizabeth G. Thornburg, The Curious Appellate Judge: Ethical Limits on Independent Research, 28 REV. LITIG. 131 (2008). In incorporating a legal standard for judicial notice under Rule 2.9(C), Opinion 478 references Federal Rule of Evidence (FRE) 201, which permits judicial notice of facts that are “not subject to reasonable dispute” for one of two reasons: either they are “generally known within the trial court’s jurisdiction,” (e.g., the sky is blue) or they “can be accurately and readily determined from sources whose accuracy cannot reasonably be questioned” (historic examples of such sources include dictionaries and maps). In the opinion, the Committee states that FRE 201 only pertains to “adjudicative” facts, or those related to information in dispute in the case. These are distinct from “legislative,” or more general law and policy questions. The Committee also notes that although judicial notice standards can vary at the state level, many jurisdictions have evidence code provisions that track FRE 201.

Applying the Law to Potential Fact Patterns

The Committee goes on to explain that, in light of Rule 2.9, FRE 201, and due process concerns, while judges and those they supervise may independently engage in internet research related to legislative facts, they may not do so with regard to adjudicative facts without notifying the parties and providing them with an opportunity to be heard as to whether or not the information they seek to include is properly subject to judicial notice. The Committee then presents a series of hypotheticals to provide guidance in distinguishing permissible versus impermissible internet research. For example, independent internet research by a judge to ascertain a business’ hours of operation in an overtime pay lawsuit would be improper because that information would be central in a wage and hour dispute and should come from the parties. In contrast, a judge hearing a case on environmental contamination may use the internet without following judicial notice procedures to obtain a general understanding of environmental law via “online background information including articles,” as long as this information is not used in making adjudicative decisions of material fact.

In keeping with the hypotheticals above, the opinion also provides a series of questions members of the judiciary may ask themselves in deciding whether or not to undertake independent internet research. For example, any information that is critical to deciding the case should be presented by the parties or else is subject to judicial notice procedures. The opinion also states that it is improper for a judge to utilize such research to corroborate or disprove adjudicative facts. With regard to seeking general background information, which can be either legislative or adjudicative, the key inquiry is whether such information will be “of factual consequence in determining the case.”

Adjudicative or Legislative Fact?

Through the questions and hypotheticals it provides in Opinion 478, the Committee acknowledges the often challenging issue of distinguishing adjudicative facts from legislative facts, as well as some of the broader difficulties that plague the judicial notice doctrine to which Rule 2.9(C) is anchored.

In the article cited above, Professor Thornburg provides an in-depth discussion of these topics and highlights the fact that in practice, it can be very difficult to draw a clear separation between adjudicative and legislative facts. As Opinion 478 also notes, she explains that one reason for this challenge is that that the same facts can be categorized as either legislative or adjudicative, depending on how they are used. For example, general “background” information could be used to understand a how a machine operates, but could also provide a foundation for deciding a factual issue in a dispute regarding whether that machine malfunctioned in injuring a plaintiff. The same can be said of research related to general principles of science or social science that may inform the factual issues in a dispute; this problem can be even more perplexing in the context of evaluating expert testimony. Moreover, judges setting out in search of legislative facts that would permissibly be part of their analysis may accidentally but easily encounter adjudicative facts in the process, potentially running afoul of judicial notice rules, thwarting due process procedures, and compromising their impartiality by acquiring personal knowledge of the facts at issue.

Judicial Notice and “Reliable” Internet Sources

Another practical barrier to implementing the guidelines set forth under Opinion 478 and Rule 2.9(C), and one that is not addressed as explicitly in the opinion, relates to the reliability of sources found online. Again, only facts that are generally known in the jurisdiction or that come from sources whose reliability can’t reasonably be questioned may appropriately come into evidence though judicial notice. Certain internet sources, such as government websites or online maps, may arguably be considered sufficiently reliable in some cases to meet the requirements of FRE 201 and its state law analogs. But others, such as Wikipedia, which can be updated by virtually anyone at any time, or sources of questionable news such as those that seem to have influenced the 2016 presidential election, arguably cannot provide a credible basis for evaluating a legal dispute, though examples of such sites abound across the internet.

Case Law Provides Scant Guidance

Despite the seemingly obvious disparity in the reliability of the kinds of internet sources that may relate to adjudicative facts, there is a troubling lack of consistency in how trial courts apply judicial notice rules to information found online. Further, and as discussed above, the lines between adjudicative and legislative facts are often blurry at best. In providing a survey of case law outcomes involving judicial incorporation of extrinsic sources, Thornburg states that while improper internet research can lead to reversal in some cases, apparently more common are cases in which judges engage in such research regarding adjudicative facts without any repercussions. Moreover, she discusses the fact that even in cases where potentially improper judicial research or ex parte communications are addressed, courts tend to look at only one piece of the equation (e.g., ethical, evidentiary, or constitutional), rather than all the relevant doctrines together. Another scholarly article, discussing judicial notice in the internet era, similarly characterizes the manner in which courts are applying the law of judicial notice to information found online as “haphazard.” See Jeffrey Bellin & Andrew Guthrie Ferguson, Trial by Google: Judicial Notice in the Information Age, 108 NW. U. L. REV. 1137 (2014).

A Step Toward More Cohesive Guidelines for Trial Court Judges

In light of this lack of clear guidance from the case law on judicial notice, it may seem curious that judicial notice was the vehicle chosen for controlling judicial internet research of case-specific facts when Rule 2.9(C) was added to the Model Code in 2007. Indeed, in the 2008 article discussed above, Thornburg calls on states to adopt Rule 2.9(C) in an amended form that would remove judicial notice as the legal standard for determining the propriety of judicial internet research.

Rightly or wrongly, in the intervening years, over 30 states have chosen to adopt Rule 2.9(C) in substantially its original form, making it one of the more prevalent points of reference on the question of internet research by judges across the country. Nonetheless, and despite the confusion that may be endemic to any effort to comply with a legal rule tethered to judicial notice, through the questions and fact patterns it offers, Opinion 478 may illuminate a more clearly demarcated way forward for trial court judges looking to utilize internet research in an appropriate manner. Moreover, and perhaps in response to scholarly observations regarding the lack of cohesion among legal authorities regarding the separate but related issues of ethical, evidentiary, and constitutional import that can come up in this context, Opinion 478 seems to reflect an effort to synthesize the relevant rules from these different areas. So, even if Rule 2.9(C)’s implementation in the internet research context will always be bogged down to some extent by the arguably intractable nature of some of the judicial notice issues discussed above (e.g., separating adjudicative and legislative facts, unintentionally encountering information relevant to adjudicative facts, finding “reliable” sources online), Opinion 478 may represent a step in the right direction by offering a more unified analysis of the relevant doctrines than has been seen in much of the case law.

Internet Research by Appellate Judges

It is worth noting that there are additional due process concerns when appellate judges utilize internet research to bring in factual information from outside the record, as the parties do not have the same opportunity to address or contest the consideration of that information that they might at the trial level. A notable example of such research can be found in Rowe v. Gibson, 798 F.3d 622 (7th Cir. 2015), a case featuring a majority opinion and dissent in which Judges Posner and Hamilton sparred over the propriety of Posner’s use of internet research in reversing a summary judgment ruling. Posner’s and Hamilton’s debate also highlights some of the main arguments surrounding the question of whether such research may properly take place at the appellate level.

In Rowe, the plaintiff was an Indiana prison inmate who sued prison administrators and staff because, among other things, they refused to provide him with Zantac for his reflux esophogitis at the prescribed times of day. Specifically, Rowe was supposed to take Zantac with meals, but the prison would only provide it to him hours after the fact, causing him severe pain and discomfort. Rowe alleged deliberate indifference to a serious medical need in violation of the Eighth Amendment, but the district court granted summary judgment to the defendants on this claim.

In a majority opinion authored by Judge Posner, the Seventh Circuit panel reversed the summary judgment ruling on the claim related to the defendants’ refusal to administer Rowe’s medication with his meals. Though the majority stated that the record alone provided a sufficient basis for the reversal, Judge Posner went on to cite multiple websites, including Wikipedia, the Mayo Clinic website, and the Zantac manufacturer’s website, as a means of illustrating that there were genuine issues of material fact unresolved as to Rowe’s claim related to the medication dosage schedule. While acknowledging that his research fell neither into the category of judicially noticeable evidence, nor facts presented and entered into evidence through the adversary process, Posner proposed that there may be a need for something in between. He suggested that particularly in a case reflecting such a disparity in resources between the parties (Rowe had been unable to secure expert medical testimony in support of his case), it could amount to a miscarriage of justice for the matter to be decided on such an underdeveloped record. He also repeatedly stressed that his internet research did not influence his ruling, but served only to “underscore the existence of a genuine issue of material fact created in the district court proceedings by entirely conventional evidence.”

In his dissent, Judge Hamilton took issue with what he perceived to be Posner’s reliance on these sources in making his ruling. Hamilton disputed not only the accuracy of Posner’s cited internet sources, but also argued that their use constituted a violation of FRE 201. He further stated that judicial use of extrinsic internet research at the appellate level posed several practical problems, not the least of which being that the record in any given case would effectively never be closed if judges were permitted to continue supplementing it in this manner.

Whether Judge Posner’s references to internet research in the majority opinion were proper or not, the debate between him and Judge Hamilton provides an overview of some of the competing values at stake in this context, and some of the difficult questions members of the judiciary face when deciding whether to open a web browser to look for case-related answers. In suggesting a need for something in between adversary procedure and judicial notice for utilizing internet-based information in relation to case-specific facts, whether that is a feasible solution or not, Posner seemed to be putting forward the proposition that with relevant information so easily accessible, particularly in cases with woefully incomplete records and underresourced parties, procedural innovation of some sort is necessary to find a way to properly incorporate such information. On the other hand, and though the need for innovation articulated by Judge Posner may be very real, due process concerns are significant if the parties have no way of foreseeing or contesting new information added during appellate proceedings via independent judicial internet research.

While Opinion 478 does not directly address internet research by appellate judges, it is perhaps notable that it came out following the very public debate on that topic between the majority and dissent in Rowe, which shone a somewhat high-profile spotlight on the concerns the ABA sought to address in enacting Rule 2.9(C). While Opinion 478 may not have been a direct response to Rowe, the guidance it provides arguably fills some of the void highlighted by that case, or at the very least, explicit questions for judges to consider before embarking on independent internet research projects.

To Google or Not to Google?

As referenced above, practical difficulties abound in applying judicial notice rules to internet research given the lack of consistency in the case law. These hurdles are further compounded by the twin quagmires of distinguishing adjudicative from legislative facts, and of accidentally finding information pertinent to adjudicative facts when researching legislative information. Though some of these challenges are arguably not unique to internet research, the ease with which courts can now access online resources, even in the middle of courtroom proceedings, has shed new light on these obstacles. The temptation to quickly resolve a seemingly minute factual question can be strong, particularly in an age when most courts are faced with colossally high caseloads and severely limited resources. Moreover, the question of reliability is arguably a trickier one relative to internet research (as opposed to paper-based research) given the vast and ever-changing nature of internet sources, many of which are anonymously authored. Further, when such information is introduced at the appellate level, judges risk effectively inserting new information into the record that the parties had no chance to address or contest in the fact-finding proceedings below.

The debate over judicial use of internet research at both the trial and appellate level is unlikely to be resolved anytime soon, particularly when judicial discretion plays a central role in this context, and arguably does not lend itself to greater consistency among the cases. However, Opinion 478 is a step toward providing a clearer framework among many confusing sources of authority on the question of independent judicial internet research, and one that members of the judiciary may find helpful as the internet’s reach becomes ever more infinite.

Comments are closed.