Appellate Courts and the Question of Social Facts

The following is a condensed introduction to Professor Caitlin Borgmann’s article “Appellate Review of Social Facts in Constitutional Rights Cases,” which is forthcoming in the California Law Review. The full article is available here.

Borgmann joined the faculty of CUNY Law in 2004. She is currently spending a sabbatical year as a senior fellow at the Columbia Center for Gender & Sexuality Law. Her scholarship focuses on the respective roles and authority of the courts and the legislatures in protecting constitutional rights, and on the role and judicial treatment of fact finding in constitutional rights cases. She has also written extensively about reproductive rights.

Professor Caitlin Borgmann

Professor Caitlin Borgmann

When legislation or government policies are challenged in federal court as violating individual constitutional rights, federal district court judges may issue written findings of fact after preliminary hearings or bench trials, which often occur before the law or policy goes into effect. The facts that are relevant to these cases are social facts, also commonly called “legislative” facts. In contrast to case-specific or “adjudicative” facts (the “who, what, when, where, how, and why” facts specific to the case), social facts are general facts, often predictive in nature, that have significance for society more broadly and that often form the basis of judicial rulemaking. Examples of such questions of fact are whether violent video games cause aggression in children, or whether certain abortion procedures carry safety advantages for the pregnant woman.

A major unresolved question is how appellate courts should treat these findings of social fact. Normally, facts found by a federal district judge are protected on appeal by Federal Rule of Civil Procedure 52(a)(6), unless those findings are “clearly erroneous.” Although nothing in the rule exempts social facts from its scope, it is widely believed that such facts are not subject to the rule and that appellate courts should review them independently. Remarkably, considering the importance of social facts to constitutional rights litigation, the Supreme Court has never addressed the question in any detail or answered it definitively.

Despite the lack of Supreme Court precedent, some federal courts of appeals have held that social facts are subject to independent appellate review.1 I refer to this exclusion of social facts from the scope of Rule 52(a)(6) as the “social facts exception.” Legal scholars often repeat as dogma the proposition that social facts are not subject to Rule 52(a)(6).2 But judges and legal experts disagree about this.3 After Perry v. Schwarzenegger,4 in which a federal district judge struck down California’s Proposition 8 banning same-sex marriages, legal experts made starkly contrasting predictions about how the trial court’s findings of fact would be treated on appeal. Some said that Judge Vaughn Walker’s findings would insulate the decision on appeal because the “clearly erroneous” standard would apply,5 while others maintained that the nature of the facts found made it unlikely that the Ninth Circuit would defer.6 On appeal, the Ninth Circuit panel seemed uncertain about how the facts should be treated, although it ultimately sidestepped the question.7

I propose a clear-cut solution to all the confusion. Appellate courts should straightforwardly apply Rule 52(a)(6) to constitutional social facts. This approach is consistent with the language and intent of the rule. It also reinforces the judiciary’s role as a protector of constitutional rights against oppression by the majority. When legislatures “find facts” to support legislation that encroaches on constitutional rights, especially when the right implicates a controversial or hot-button social issue, this fact finding is often biased and unreliable. Constitutional rights claimants look to the federal courts for dispassionate, independent review of the relevant social facts. Trial courts are well-positioned to perform this function.

For example, trial judges are able to observe the verbal and nonverbal behavior of trial participants, including expert witnesses. This can be particularly helpful when a trial court confronts the contentious factual issues often underlying laws that restrict controversial or minority rights. For example, in a legislative hearing on a “partial-birth abortion” ban in Alaska, witnesses compared abortion procedures to the “Nazi Holocaust” and testified that the medical profession had “cut its own throat” by allowing abortion providers to “perform unjustified abortions” and “falsify the patient’s records” in order to secure a “prearranged convenience for the mother and a financial benefit for the doctor.”8 In contrast, when “partial-birth abortion” bans were tested in federal court, the district judges had the opportunity to see and hear abortion doctors in person, which helped counter these caricatures prevalent in public discourse and in legislative hearings. This opportunity may help explain why 12 out of 13 federal district courts invalidated the bans, finding that they prohibited methods that protected women’s health.

Appellate courts, on the other hand, lack many of these advantages. There are no formal rules governing fact finding at the appellate level. Advocacy groups and others submit amicus briefs at the appellate level that contain factual assertions that are unvetted by the adversary process. In addition, Supreme Court justices and appellate judges freely turn to their own research, including Internet searches. This informal, unscreened fact finding deprives the parties of the opportunity to contest or develop facts “found” by the appellate court. There is no reason to think that this system is better at resolving social fact disputes than the tried-and-true process of a trial.

The Supreme Court’s opinion upholding the federal Partial-Birth Abortion Ban Act illustrates the problems that occur when appellate courts independently review social facts. In Gonzales v. Carhart, Justice Kennedy asserted, admittedly without “reliable data,” that women are emotionally traumatized by abortion.9 The “authority” Justice Kennedy cited was an amicus brief submitted by Sandra Cano, the plaintiff from Doe v. Bolton10 who now claims to regret her abortion, and “180 post-abortive women who have suffered the adverse emotional and psychological effects of abortion.”11 The brief was part of a deliberate campaign of anti-abortion-rights groups to persuade Justice Kennedy to reverse his position on the constitutional status of abortion. The asserted social “fact” was not relevant to the case, was of questionable scientific validity, and surely would not have survived the rigors of a trial. Justice Kennedy even conceded it lacked a sufficient scientific foundation. Yet it now carries the Supreme Court’s imprimatur and is being cited by lawyers trying to convince courts to uphold laws that require women to be told that abortions cause women emotional harm.

In short, the pat rule that social facts should be independently reviewed, while adjudicative facts merit the deference of the “clearly erroneous” standard, does not hold up under scrutiny. A likely reason the Supreme Court has never definitively declared Rule 52(a)(6) inapplicable to all social fact finding is that such a rule seems too rigid. The justices surely sense that at times, deference is appropriate. Yet no one has offered a more nuanced rule as to when deference is appropriate and when it is not. This article aims to present such a rule.

The article first establishes the basic groundwork for understanding appellate review of social facts. It describes the general appellate standard of review for trial judge fact finding, the “clearly erroneous” standard, and the social facts exception. It goes on to identify and examine the various rationales underlying the social facts exception and concludes that none is sufficient to justify excluding social facts from Rule 52(a) (6). The article then proposes a framework for appellate review of social facts that incorporates the “clearly erroneous” standard.

More from CUNY Law magazine Spring 2013 »


1 See, e.g., United States v. Singleterry, 29 F.3d 733, 740 (1st Cir. 1994); Dunagin v. Oxford, 718 F.2d 738, 748 n.8 (5th Cir. 1983) (en banc) (plurality opinion).

2 See, e.g., Bryan Adamson, “Critical Error: Courts’ Refusal to Recognize Intentional Race Discrimination Findings as Constitutional Facts,” 28 Yale Law & Policy Review 1, 17 n.71 (2009).

3 See, e.g., McPherson v. Rankin, 786 F.2d 1233, 1237 (5th Cir. 1986) (“The precise fit of the clearly erroneous standard of review and our duty to make ‘an independent judgment on the facts of the case[] is not altogether clear.”), aff ‘d, 483 U.S. 378 (1987).

4 704 F. Supp. 2d 921 (N.D. Cal. 2010), aff ‘d sub nom. Perry v. Brown, 671 F.3d 1052, reh’g en banc denied, 681 F.3d 1065 (9th Cir. 2012), cert. granted sub nom. Hollingsworth v. Perry, 133 S. Ct. 786 (2012).

5 See, e.g., Andrew Koppelman, “Power in the Facts,” New York Times Room for Debate (Aug. 4, 2010, 10:03 PM), 2010/8/4/gay-marriage-and-the-constitution/judge-walkers-factual-findings; Dahlia Lithwick, “A Brilliant Ruling,” Slate (Aug. 4, 2010, 9:27 PM ET), ruling.html; Mark Ambinder, “Prop 8 Overturned: The Facts, Not the Law, Matter,” The Atlantic (Aug. 4, 2010, 5:17 PM ET), archive/2010/08/prop-8-overturned-the-facts-not-the-law-matter/60957/.

6 Dave Hoffman, “There Are Facts, and Then There Are Constitutional Facts,” Concurring Opinions (Aug. 5, 2010, 1:07 AM); Orin Kerr, “How Much Do the Factual Findings Matter in Perry v. Schwarzenegger?” The Volokh Conspiracy (Aug. 5, 2010, 2:42 AM), how-much-do-the-factual-findings-matter-in-perry-v-schwarzenegger/. Hoffman’s post reflects a common confusion or blurring between social (or “legislative”) facts and “constitutional facts.” See infra Part II.A.2,3. Here he seems to use the term to mean social facts, not constitutional facts in the sense in which the Supreme Court usually uses this term. Id.

7 Perry v. Brown, 671 F.3d 1052 (reciting Rule 52(a)(6) but expressing uncertainty as to whether the district judge’s findings of social fact were best characterized as “adjudicative” or “legislative” and whether the “clearly erroneous” standard should apply), reh’g en banc denied, 681 F.3d 1065 (9th Cir. 2012).

8 “An Act Relating to Partial-Birth Abortions: H. Judiciary Comm. Meeting on H.B. 65,” 20th Leg. (Alaska 1997), available at basis/get_single_minute.asp?ch=H&beg_line=0965&end_line=2183& sessio n=20&comm=JUD&date=19970307&time=1308.

9 550 U.S. 124, 159 (2007).

10 Doe v. Bolton, 410 U.S. 179 (1973).

11 Brief of Sandra Cano et al. as Amici Curiae in Support of Petitioner at 2, Gonzales v. Carhart, 127 S. Ct. 1610 (2007) (No. 05-380), 2006 WL 1436684.