RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 24a0160p.06
UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT
┐ CONNIE REGULI, │ Plaintiff-Appellant, │ > No. 23-5925 │ v. │ │ LORI RUSS; BRENTWOOD POLICE DEPARTMENT; CITY │ OF BRENTWOOD, TENNESSEE, │ Defendants-Appellees. │ ┘
Appeal from the United States District Court for the Middle District of Tennessee at Nashville. No. 3:22-cv-00896—Aleta Arthur Trauger, District Judge.
Decided and Filed: July 31, 2024
Before: GIBBONS, WHITE, and MURPHY, Circuit Judges. _________________
COUNSEL
ON BRIEF: Connie Reguli, Brentwood, Tennessee, in pro. per. Cassandra M. Crane, FARRAR ǀ BATES ǀ BEREXA, Brentwood, Tennessee, for Appellees.
The court delivered a PER CURIAM opinion. WHITE, J. (pg. 14), delivered a separate concurring opinion. MURPHY, J. (pp. 15–18), also delivered a separate concurring opinion in which GIBBONS, J., joined.
_________________
OPINION _________________
PER CURIAM. In January 2019, Detective Lori Russ searched Connie Reguli’s private Facebook records allegedly because Russ disliked Reguli’s criticism of the police. Reguli learned of this search a year later when preparing for her criminal trial. She did not sue over the No. 23-5925 Reguli v. Russ, et al. Page 2
search at that time. Much later, however, Reguli learned that her speech had motivated the search when Russ seemed to admit as much at Reguli’s sentencing in July 2022. That November, Reguli filed a First Amendment retaliation claim against Russ and her employer under 42 U.S.C. § 1983. But Reguli’s § 1983 claim triggered a short one-year statute of limitations under Tennessee law. So the district court dismissed Reguli’s claim as untimely. The court reasoned that this claim had accrued when Reguli learned of Russ’s search—not when she learned of Russ’s motivation for it. We agree and affirm.
I
Connie Reguli has practiced law in Tennessee for years. She started out as a local prosecutor in a district attorney’s office with a focus on domestic-violence and child-abuse cases. After switching to a private civil-rights practice, she regularly represented “families” in proceedings to remove children from their homes initiated by the Tennessee Department of Children’s Services (which goes by “DCS”). Compl., R.1, PageID 2. In this role, Reguli acquired a “distrust of government agents,” especially DCS employees. Id. She conveyed her “disdain” for the government on her public Facebook page, which developed a following of some 17,000 people. Id. But she eventually grew suspicious that government agents were monitoring her page, so she set it to “private.” Id.
In August 2018, DCS employees believed that Reguli’s zealous advocacy had crossed the line into criminal misconduct. The evidence from Reguli’s criminal trial recorded the events of that month. See State v. Reguli, 2024 WL 913212, at *1 (Tenn. Crim. App. Mar. 4, 2024). In early August, DCS began to investigate Wendy Hancock over the care she was providing her two children. Id. Reguli agreed to represent Hancock during the investigation. Id. Things escalated quickly. On August 13, DCS filed an ex parte petition to take custody of Hancock’s children. Id. A juvenile court granted this petition. See id. But DCS could not find Hancock or her 12- year-old daughter because they were staying at a hotel. See id.
Two days later, the Tennessee Bureau of Investigation issued a “missing and endangered child alert” for Hancock’s daughter. See id. While at the hotel, her daughter saw this alert on her phone and showed it to her mother and Reguli. See id. Because Reguli had visited the No. 23-5925 Reguli v. Russ, et al. Page 3
county clerk’s office earlier that day, she also knew of the court order granting custody of Hancock’s daughter to DCS. See id. Despite this knowledge, Reguli let Hancock and her daughter stay at her house in Brentwood, Tennessee. See id. Hancock disabled her phone and her daughter’s phone to avoid detection. See id. Reguli gave Hancock a new phone to use. See id.
But Hancock’s daughter had posted on social media shortly before Hancock disabled her phone. See id. After spotting this post, the police “pinged” the daughter’s phone and learned of its location at Reguli’s house. See id. They discovered Hancock and her daughter there. See id.
The next month, a DCS attorney referred Hancock and Reguli to the Brentwood Police Department. The attorney asked the police to look into whether the women had illegally interfered with DCS’s custody of Hancock’s daughter. Russ, a detective with the Brentwood police, oversaw the investigation.
On December 4, Russ sought a warrant to search Reguli’s Facebook records. Russ requested, among other things, Reguli’s “subscriber information,” her “chat logs or private messages,” and the “images and videos” that she had uploaded. Aff., R.1-1, PageID 27. In her affidavit seeking the warrant, Russ described Reguli’s conduct in August. Russ also noted that Reguli had posted videos to her Facebook page describing the August events and admitting her knowledge that DCS had taken custody of Hancock’s daughter. And Russ suggested that Reguli’s other Facebook records, such as her private messages, might contain evidence of her role in the alleged custodial-interference crime.
A judge issued the search warrant later that day. In January 2019, Russ obtained over 20,000 pages of Reguli’s Facebook records dating from August to December 2018.
The same month, Reguli first learned of the police investigation. She requested all records related to the investigation from the Brentwood Police Department. In response, the department refused to provide any information due to the pending investigation.
In July 2019, a Tennessee grand jury indicted Hancock on a count of custodial interference and Reguli on a count of facilitating Hancock’s offense and two counts of being an No. 23-5925 Reguli v. Russ, et al. Page 4
accessory after the fact. See Reguli, 2024 WL 913212, at *2. In January 2020, Reguli obtained discovery from the prosecutor and learned for the first time that Russ had obtained a warrant back in 2018 to search her Facebook records.
The trial court held separate trials for Hancock and Reguli. The women’s respective juries found them guilty as charged. When testifying at Reguli’s trial, Russ did not introduce or otherwise use any of Reguli’s Facebook records.
The trial court sentenced Reguli on June 24, 2022. Russ testified at Reguli’s sentencing. The prosecution asked Russ why she had obtained the Facebook records. Russ’s answer suggested that Reguli’s speech had at least partially motivated this request:
Because of the things that she would put out on Facebook[,] Facebook Live, YouTube. She was very blatantly mocking of the whole process. From the beginning up until even after she was convicted, she was very blasé. She mocked the police. She accused myself personally of being involved in some sort of scheme and essentially questioning my integrity as a police officer and my involvement in saying that this was . . . And it’s been a consistent theme the whole time about it’s a conspiracy against her, that we’re after her because she is a squeaky wheel. And I felt like it was relevant to her actions.
Russ Tr., R.1-1, PageID 37. On cross-examination, Reguli’s attorney also asked Russ why she continued to monitor Reguli’s Facebook page even after the trial. (It is unclear whether Reguli’s Facebook page was still set to private and how Russ accessed her posts.) Russ reiterated her distaste for what she thought were Reguli’s unwarranted criticisms of her and her associates:
Because Ms. Reguli has repeatedly besmirched the people that were involved in this case and myself included. She posted things to the Brentwood Police Department’s Facebook page about me. I wanted to know if she was talking about me personally, because it’s an attack on my integrity that she would imply or blatantly say that I was acting in some sort of conspiracy against her with all of these other people that are involved, when I didn’t know her. My job is to find facts and whether it is to exonerate someone or to prosecute someone. And it’s insulting. So I wanted to know what she was saying personally about me.
Id., PageID 38. Russ then admitted that “it was kind of personal[.]” Id. The trial court ultimately sentenced Reguli to 30 days in jail and three years of probation. Reguli, 2024 WL 913212, at *1. No. 23-5925 Reguli v. Russ, et al. Page 5
On appeal, the Tennessee Court of Criminal Appeals reversed Reguli’s conviction and dismissed her criminal case. See id. Tennessee’s custodial-interference law makes it unlawful for a noncustodial parent of a child under 18 years old to “[d]etain the child . . . after the expiration of the . . . parent[’s] . . . lawful period of visitation, with the intent to violate . . . a court order regarding the custody or care of the child[.]” Tenn. Code Ann. § 39-13-306(a)(2) (emphasis added). The court interpreted the italicized language as requiring prosecutors to prove that a court had granted visitation rights to a parent and that the parent had detained the child after these rights had expired. Reguli, 2024 WL 913212, at *3. Although the court found Hancock’s conduct “troubling,” she had not been meeting with her daughter pursuant to a “visitation order” when she kept her at Reguli’s house in violation of the order granting custody to DCS. See id. at *3–4. The court thus held that Hancock had not violated this custodial- interference law. See id. So Reguli could not have facilitated that offense or been an accessory after the fact. See id.
On November 7, 2022, Reguli brought this federal lawsuit against Russ, the City of Brentwood, and its police department under 42 U.S.C. § 1983. As relevant to this appeal, Reguli’s complaint alleged that Russ had violated the First Amendment by obtaining a warrant to search Reguli’s Facebook records in retaliation for Reguli’s critical commentary about Russ, the police, and DCS.
A magistrate judge suggested that Reguli did not timely file her suit. Reguli v. Russ, 2023 WL 6690948, at *6–9 (M.D. Tenn. Aug. 22, 2023). The district court agreed and dismissed Reguli’s complaint. Reguli v. Russ, 2023 WL 6129503, at *8–12 (M.D. Tenn. Sept. 19, 2023).
II
Reguli appeals the dismissal of her First Amendment retaliation claim under § 1983. We review the district court’s decision de novo, accepting the complaint’s well-pleaded factual allegations as true. See Baltrusaitis v. UAW, 86 F.4th 1168, 1174 (6th Cir. 2023). Because a defendant bears the burden of proving a statute-of-limitations defense, a plaintiff’s complaint need not allege facts showing that a claim is timely. See Cataldo v. U.S. Steel Corp., 676 F.3d No. 23-5925 Reguli v. Russ, et al. Page 6
542, 547 (6th Cir. 2012). But plaintiffs can plead themselves out of court on statute-of- limitations grounds if the complaint alleges facts showing that they did not sue in time. See id.; see also Baltrusaitis, 86 F.4th at 1178. The key factual allegations in Reguli’s complaint do just that.
A
Section 1983 gives plaintiffs a right to seek damages from any “person” who, while acting “under color of” state law, “subjects” the plaintiffs “to the deprivation of any rights, privileges, or immunities secured by the Constitution[.]” 42 U.S.C. § 1983. This provision lacks its own statute of limitations. Courts have filled in this gap with a mix of federal and state rules. See 42 U.S.C. § 1988(a); Dibrell v. City of Knoxville, 984 F.3d 1156, 1160–61 (6th Cir. 2021).
At the outset, state law determines the length of § 1983’s statute of limitations. The Supreme Court has held that the statute incorporates the limitations period for personal-injury torts from the State where the events occurred. See Wallace v. Kato, 549 U.S. 384, 387−88 (2007).
Conversely, federal law determines when a § 1983 claim accrues to trigger the running of this state statute of limitations. See id. at 388. The Supreme Court has explained that the “standard” accrual “rule” starts a limitations period when “the plaintiff has ‘a complete and present cause of action.’” Bay Area Laundry & Dry Cleaning Pension Tr. Fund v. Ferbar Corp. of Cal., Inc., 522 U.S. 192, 201 (1997) (quoting Rawlings v. Ray, 312 U.S. 96, 98 (1941)); see Rotkiske v. Klemm, 589 U.S. 8, 13 (2019). Put differently, this “injury-occurrence” or “occurrence” rule triggers the limitations period on the first day that every element of a claim has occurred such that the plaintiff may sue in court over the claim. See Wallace, 549 U.S. at 388. The Supreme Court has recited this rule in three § 1983 cases. See Reed v. Goertz, 598 U.S. 230, 235–36 (2023); McDonough v. Smith, 588 U.S. 109, 114–15 (2019); Wallace, 549 U.S. at 388.
But our § 1983 cases have taken a different approach. We have suggested that the statute adopts a “discovery rule,” not an “occurrence rule.” See Dibrell, 984 F.3d at 1162; Sharpe v. Cureton, 319 F.3d 259, 266 (6th Cir. 2003). This rule postpones the limitations period to the date that the plaintiff discovered, or reasonably should have discovered, basic facts about the No. 23-5925 Reguli v. Russ, et al. Page 7
claim. See Johnson v. Memphis Light Gas & Water Div., 777 F.3d 838, 843 (6th Cir. 2015); Dixon v. Anderson, 928 F.2d 212, 215 (6th Cir. 1991); Sevier v. Turner, 742 F.2d 262, 273 (6th Cir. 1984).
B
Applying this law, the parties agree on some things and disagree on others. They agree that § 1983 incorporates a one-year statute of limitations from Tennessee. See Tenn. Code Ann. § 28-3-104(a)(1)(A)−(B); Dibrell, 984 F.3d at 1161. They also agree—and so we may assume— that the discovery rule applies given our caselaw. Nevertheless, the parties disagree over the date that Reguli’s First Amendment retaliation claim accrued under this rule. Russ would start the clock in January 2020 when Reguli discovered that Russ had searched her Facebook accounts. Under this view, Reguli filed her suit (in November 2022) well outside the statute of limitations. Reguli, by contrast, would start the clock in June 2022 when she learned at her sentencing that her speech had motivated Russ to engage in this search. Under this view, she timely sued.
At bottom, this debate over the proper accrual date arises from the parties’ conflicting answers to two questions: When did Reguli have a “complete and present” First Amendment retaliation claim that could even trigger the discovery rule’s inquiry into her knowledge about the claim’s existence? And what facts must Reguli have discovered about this claim to start the limitations period under that discovery rule? Russ has better answers to both questions.
1. When did Reguli have a “complete and present” retaliation claim?
Whether a plaintiff has a “complete and present cause of action” under § 1983 turns on the “specific constitutional right” at issue. Reed, 598 U.S. at 235–36. So we start by identifying the elements of Reguli’s claim. The First Amendment (as incorporated by the Fourteenth) prohibits state actors from “abridging the freedom of speech[.]” U.S. Const. amend. I; Stromberg v. California, 283 U.S. 359, 368 (1931). The Supreme Court has read this text to bar state actors from taking “retaliatory actions” against a party because of the party’s “protected speech.” Hous. Cmty. Coll. Sys. v. Wilson, 595 U.S. 468, 474 (2022) (quoting Nieves v. Bartlett, 587 U.S. 391, 398 (2019)). This sort of retaliation claim has three elements. See Lemaster v. No. 23-5925 Reguli v. Russ, et al. Page 8
Lawrence County, 65 F.4th 302, 307–10 (6th Cir. 2023). First, the plaintiff must have engaged in speech that the First Amendment protects—such as criticism of the police. See Rudd v. City of Norton Shores, 977 F.3d 503, 513–14 (6th Cir. 2020). Second, the state actor must have taken an “adverse” (that is, harmful) action against the plaintiff—such as “an arrest, a prosecution, or a dismissal from governmental employment.” Hous. Cmty. Coll. Sys., 595 U.S. at 477; Lemaster, 65 F.4th at 307. Third, a “causal connection” must exist between these elements—such that the protected speech motivated (indeed, qualified as a but-for cause of) the adverse action. Rudd, 977 F.3d at 515.
Under these elements, Reguli had a “complete and present cause of action” when Russ obtained Reguli’s Facebook records in early 2019. Reed, 598 U.S. at 235 (quoting Bay Area Laundry, 522 U.S. at 201). Had Reguli engaged in her “protected” speech by then? Rudd, 977 F.3d at 513. Yes, the complaint alleges that she regularly used Facebook to criticize the police, including soon after the events of August 2018. Compl., R.1, PageID 2, 8. Had Russ engaged in the “adverse action” by then? Rudd, 977 F.3d at 513. Yes, the complaint suggests that Russ requested a search warrant for Reguli’s Facebook records in December 2018 and obtained them in January 2019. Compl., R.1, PageID 10. Lastly, did a “causal connection” exist between Reguli’s speech and Russ’s search? Rudd, 977 F.3d at 513. Yes, the complaint suggests that Russ initiated the Facebook search “[b]ecause of the things that [Reguli] would put out on Facebook.” Compl., R.1, PageID 13. And while evidence of Russ’s allegedly improper motive came out later during Reguli’s sentencing, the motive itself necessarily existed at the time of the search. After all, Russ’s subjective intent for taking a past act could not have changed after the act. Unsurprisingly, then, the Supreme Court has recognized for a similar retaliation claim that the plaintiff typically has a complete cause of action “when the retaliatory action occurs.” Graham Cnty. Soil & Water Conservation Dist. v. United States ex rel. Wilson, 545 U.S. 409, 419 (2005).
In response, Reguli suggests that First Amendment retaliation claims have a fourth element. According to Reguli, plaintiffs must have “suffered” not just an adverse action in retaliation for their speech but also a “chilling effect” on that speech before they have actionable claims. Appellant’s Br. 42, 48. And here, Reguli says, she could not have suffered any “chill” No. 23-5925 Reguli v. Russ, et al. Page 9
until she learned at sentencing of Russ’s unlawful motivation. She is mistaken. Nothing in the First Amendment’s text nor in the caselaw interpreting it suggests this element. That is for good reason. The element would permit officials to punish speech under the First Amendment whenever their harmful acts did not “chill” courageous speakers from speaking. Counterman v. Colorado, 600 U.S. 66, 75 (2023). But the First Amendment protects the stout no less than the timid.
An action’s “chilling effect” instead plays only a winnowing role in deciding whether plaintiffs have adequately alleged the second (adverse action) element of their claim. An official’s conduct qualifies as sufficiently “adverse”—and so actionable under the First Amendment—only if it would “chill” (that is, deter) “a person of ordinary firmness” from speaking. Hous. Cmty. Coll. Sys., 595 U.S. at 477 (quoting Nieves, 587 U.S. at 397); see Rudd, 977 F.3d at 514. For example, a plaintiff may not sue over a state official’s “mere frown” because facial expressions alone would not deter expression. Hous. Cmty. Coll. Sys., 595 U.S. at 477. If we reached the merits of Reguli’s claim, then, we would have to ask whether Russ’s alleged adverse action (her invasion of Reguli’s privacy through a search of her Facebook records) would deter an “ordinary citizen” from engaging in protected expression. Rudd, 977 F.3d at 514. If it would, however, Reguli would not have to establish—as another element—that this conduct subjectively chilled her speech.
To support a contrary view, Reguli relies on Bloch v. Ribar, 156 F.3d 673 (6th Cir. 1998), and Barrett v. Harrington, 130 F.3d 246 (6th Cir. 1997). But she misreads these two cases. In both, an official’s speech qualified as the “adverse action” that the plaintiff sued over. The plaintiff in Bloch—a rape victim—alleged that a sheriff revealed confidential details of the rape in retaliation for her criticism of him. 156 F.3d at 676, 678–81. The plaintiff in Barrett—a litigant—alleged that a state judge made false statements about him in retaliation for his criticism of her. 130 F.3d at 249–51, 262–63. So the date of the harmful speech in these cases (the public disclosures or defamatory statements) qualified as the date of the adverse actions. Reguli, by contrast, nowhere claims that Russ gave her sentencing testimony in retaliation for Reguli’s speech. Rather, Reguli claims that Russ searched her Facebook records in retaliation for that No. 23-5925 Reguli v. Russ, et al. Page 10
speech. Here, then, Russ’s search—not her speech—is the adverse action. So the date of the search (early 2019) dictates when Reguli had a complete and present case of action.
We end with a disclaimer. The Supreme Court has held that if a § 1983 claim “would necessarily imply the invalidity” of a criminal conviction, a plaintiff cannot bring the claim until successfully overturning that conviction. Heck v. Humphrey, 512 U.S. 477, 487 (1994). Or, to put this principle in statute-of-limitations language, the plaintiff does not have a “complete and present cause of action” until that time. McDonough, 588 U.S. at 119 (quoting Wallace, 549 U.S. at 388). Here, however, Reguli does not assert that her First Amendment claim would have implied the invalidity of her criminal conviction (before she successfully overturned it on other grounds). Indeed, her complaint alleged that the Facebook evidence played no role in her criminal trial. Compl., R.1, PageID 12. So our opinion should not be taken to impliedly resolve any Heck issues.
2. Did the discovery rule require Reguli to know of Russ’s motivation?
Because Russ agrees that the discovery rule applies, she must show more than the mere existence of a complete and present cause of action to trigger the limitations period. She must also show that Reguli knew of this cause of action (or at least reasonably should have known of it). Yet our cases have sent mixed messages on the specific facts that § 1983 plaintiffs must know to start the limitations period under the discovery rule. In dozens of cases dating back decades, we have said that the limitations period begins on the date that “the plaintiff knows or has reason to know of the injury which is the basis of his action.” Sevier, 742 F.2d at 273 (emphasis added); see, e.g., Endres v. Ne. Ohio Med. Univ., 938 F.3d 281, 292 (6th Cir. 2019); Johnson, 777 F.3d at 843; Sharpe, 319 F.3d at 266; Dixon, 928 F.2d at 215; see also Rotella v. Wood, 528 U.S. 549, 555 (2000). So, for example, we have held that a § 1983 plaintiff’s excessive-force claim accrued on the day of the force because the plaintiff knew of the injury at that point—even if he did not know the identity of the officer who had used the force and caused this injury. See Miller v. Cocke County, 2022 WL 103143, at *2 (6th Cir. Jan. 11, 2022); see also Ruiz-Bueno v. Maxim Healthcare Servs., Inc., 659 F. App’x 830, 834 (6th Cir. 2016); Dowdy v. Prison Health Servs., 21 F. App’x 433, 434–35 (6th Cir. 2001) (order). No. 23-5925 Reguli v. Russ, et al. Page 11
In another § 1983 case, though, we suggested that the discovery rule starts the limitations period when the plaintiff “knows, or in the exercise of due diligence should have known, both his injury and the cause of that injury.” Bishop v. Child.’s Ctr. for Developmental Enrichment, 618 F.3d 533, 536 (6th Cir. 2010) (quoting Campbell v. Grand Trunk W. R.R., 238 F.3d 772, 775 (6th Cir. 2001)) (emphasis added); see Snyder-Hill v. Ohio State Univ., 48 F.4th 686, 701 (6th Cir. 2022). This precedent relies on United States v. Kubrick, 444 U.S. 111 (1979). Kubrick addressed when a medical-malpractice claim accrued under the Federal Tort Claims Act. Id. at 113. Although the Court recited the general rule that a claim accrues “at the time of the plaintiff’s injury,” it suggested that the limitations period should not start in medical-malpractice cases “until the plaintiff has discovered both his injury and its cause.” Id. at 119–20. So perhaps Kubrick suggests that the discovery rule’s elements depend on the claim at issue. Cf. Miller, 2022 WL 103143, at *2.
Regardless, we need not reconcile these views in this case. Under either approach to the discovery rule, our cases leave no doubt that a statute of limitations can start to run even if a § 1983 plaintiff lacks knowledge of every element of the claim. Kubrick itself held as much. The Supreme Court found the medical-malpractice claim untimely because the plaintiff knew of both his injury (hearing loss) and its cause (an antibiotic that a hospital had provided). See id. at 122–24. The Court rejected the claim that the discovery rule should delay the limitations period until the plaintiff also knew that the hospital had breached its duty of care. Id. at 123. In other words, the plaintiff did not need to know of this negligence element to start the limitations period. The Supreme Court’s decision in Rotella followed the same approach when finding a civil claim under the Racketeer Influenced and Corrupt Organizations Act (RICO) untimely. There, a patient argued that a psychiatric hospital had injured him in violation of RICO by admitting him to maximize its profits rather than to provide needed care. Rotella, 528 U.S. at 551–52. RICO has a four-year statute of limitations, yet the plaintiff had discovered this injury over a decade before he sued. Id. at 552. He nevertheless argued that RICO required him to show that the hospital had engaged in a “pattern of racketeering activity” and that he reasonably lacked knowledge of this “pattern” element until just a few years before the suit. Id. at 552–53. So he claimed that he had sued in time. The Court disagreed. It reasoned that the clock starts on No. 23-5925 Reguli v. Russ, et al. Page 12
the “discovery of the injury, not discovery of the other elements of a claim[.]” Id. at 555. The plaintiff thus did not need to know of the pattern element to trigger RICO’s limitations period.
This precedent dooms Reguli’s First Amendment retaliation claim. Whether she needed to know about only her injury or about both her injury and its cause, she did not file her claim within the one-year statute of limitations. Her complaint alleged that she discovered her injury— the invasion of privacy from the Facebook search—“in about January 2020.” Compl., R.1, PageID 11. And her complaint alleged that she learned of this injury’s cause—“Detective Lori Russ”—at the same time. Id. This knowledge started the limitations period in January 2020, whether or not Reguli knew of the “other elements of” her claim. Rotella, 528 U.S. at 555.
In response, Reguli argues that she knew of neither her injury nor its cause until her sentencing in June 2022. She first claims that she did not suffer her entire “injury”—including the “chilling effect” on her speech—until Russ admitted at sentencing that she had conducted the search because of Reguli’s criticism of the police. Even if Reguli could recover for the “mental anguish” that she allegedly felt on hearing Russ’s testimony, this theory would do Reguli no good. Bloch, 156 F.3d at 679 (citation omitted). Courts universally recognize that the discovery rule begins once a plaintiff learns of an injury from the defendant’s conduct—even if the plaintiff does not discover the “full extent of the injury” until later. Wallace, 549 U.S. at 391 (citation omitted) (emphasis added); see, e.g., Stephens v. Clash, 796 F.3d 281, 288 (3d Cir. 2015); Varnell v. Dora Consol. Sch. Dist., 756 F.3d 1208, 1216 (10th Cir. 2014); Goodhand v. United States, 40 F.3d 209, 212 (7th Cir. 1994). And Reguli learned of her primary injury well before sentencing.
Reguli’s “causation” theory fares no better. She argues that, until her sentencing, she lacked knowledge that Russ had engaged in the search because of her protected speech. And she says that the discovery rule required her to know of this causation element to trigger the limitations period. Yet the cases incorporating a causation inquiry into the discovery rule have indicated that plaintiffs must have known of “the defendant [that] caused their injury”—not the defendant’s subjective reasons for doing so. Snyder-Hill, 48 F.4th at 704 (emphasis added). And here, Reguli concedes that she had pinpointed Russ as the culprit in January 2020. No. 23-5925 Reguli v. Russ, et al. Page 13
An analogy confirms this conclusion. A defendant’s subjective motives for taking a harmful act routinely matter in the law. Plaintiffs, for example, often assert that public employers took employment actions against them for forbidden reasons—say, racial discrimination in violation of the Equal Protection Clause or political discrimination in violation of the First Amendment. See Martinez-Rivera v. Puerto Rico, 812 F.3d 69, 70, 75 (1st Cir. 2016); Lukovsky v. City & Cnty. of San Francisco, 535 F.3d 1044, 1046, 1048–51 (9th Cir. 2008); see also Chardon v. Fernandez, 454 U.S. 6, 6–8 (1981) (per curiam). And “numerous” courts have adopted a bright-line rule that these claims accrue when plaintiffs learn of the “employment action” that injured them. Lukovsky, 535 F.3d at 1049 (citing cases). Courts have also kept to this rule even when plaintiffs allege that they did not learn of the “discriminatory motive” until after the employment action. Id. at 1051; cf. Amini v. Oberlin Coll., 259 F.3d 493, 500–01 (6th Cir. 2001). Our logic follows the same path: we tie the discovery rule to the date that Reguli learned of Russ’s adverse action—not the date she learned of Russ’s “motive” for this action. Lukovsky, 535 F.3d at 1051.
Reguli lastly protests that she lacked the ability to sue before Russ testified at sentencing because plaintiffs cannot pursue litigation in court based on “mere conjecture” about a defendant’s subjective motives. Appellant’s Br. 51. Yet courts in the employment setting have considered this predicament and suggested an answer: they might toll the statute of limitations if a plaintiff could not have reasonably determined whether an unlawful motive existed before the limitations period expired. See Lukovsky, 535 F.3d at 1051 n.5; Amini, 259 F.3d at 500–01; see also Wallace, 549 U.S. at 394. But we need not consider this possibility here. Although Reguli raised a tolling argument in the district court, she did not renew the argument in this court. She has thus forfeited (or perhaps even waived) any tolling claim. See Bannister v. Knox Cnty. Bd. of Educ., 49 F.4th 1000, 1011–12 (6th Cir. 2022).
We affirm. No. 23-5925 Reguli v. Russ, et al. Page 14
__________________
CONCURRENCE __________________
HELENE N. WHITE, Circuit Judge, concurring. I agree that Reguli’s claim accrued under federal law when she had “a complete and present cause of action,” Maj. Op. at 6 (quoting Bay Area Laundry & Dry Cleaning Pension Tr. Fund v. Ferbar Corp. of Cal., Inc., 522 U.S. 192, 201 (1997))—when Russ obtained her Facebook records. Under the discovery rule, which the parties agree applies here, Russ must also “show that Reguli knew of this cause of action (or at least reasonably should have known of it).” Id. at 10. Whether this is so is generally a fact- intensive inquiry. Here, Reguli’s complaint establishes that she should have pursued a First Amendment retaliation claim when she received the discovery packet in January 2020 and perceived that the entire prosecution was unfounded and irregular. Because straightforward application of the discovery rule makes this clear, I find much of the majority’s discussion unnecessary. More specifically, I would not relegate the discovery of a retaliatory motive in cases where retaliation is an element of the claim to the domain of equitable tolling in all cases. This is especially so because equitable tolling, unlike accrual, is generally determined under state law. See Wallace v. Kato, 549 U.S. 384, 385 (2007); Johnson v. Memphis Light Gas & Water Div., 777 F.3d 838, 845 (6th Cir. 2015). No. 23-5925 Reguli v. Russ, et al. Page 15
MURPHY, Circuit Judge, concurring. We decide this case in the way that the parties have presented it to us. See United States v. Sineneng-Smith, 590 U.S. 371, 375–76 (2020). So we assume that the “discovery rule”—which delays the usual start date of a statute of limitations—applies to Connie Reguli’s First Amendment claim under 42 U.S.C. § 1983. But this assumption may well prove misplaced. As in another statute-of-limitations case that the Supreme Court recently decided, the parties debate the mechanics of the discovery rule without addressing “the logically antecedent question”: Should this rule apply at all? Warner Chappell Music, Inc. v. Nealy, 144 S. Ct. 1135, 1140 (2024) (Gorsuch, J., dissenting). Although our court has recited the discovery rule in many § 1983 cases, I have my doubts about our approach. Frankly, the Supreme Court’s caselaw on this question and our caselaw on it seem to be speaking different languages.
1. Start with the Supreme Court’s approach. In recent cases, the Court has repeatedly considered when different types of constitutional claims accrue to trigger § 1983’s statute of limitations. See Reed v. Goertz, 598 U.S. 230, 235–37 (2023); McDonough v. Smith, 588 U.S. 109, 115–25 (2019); Wallace v. Kato, 549 U.S. 384, 388 (2007); see also Manuel v. City of Joliet, 580 U.S. 357, 369–72 (2017). Each time, the Court has followed the same general framework. It has first pinpointed the “specific constitutional right” on which the § 1983 plaintiff relied—whether the Fourteenth Amendment’s Due Process Clause or the Fourth Amendment’s ban on unreasonable searches and seizures. Manuel, 580 U.S. at 370 (citation omitted); see Reed, 598 U.S. at 236; McDonough, 588 U.S. at 115; Wallace, 549 U.S. at 388.
The Court has next turned to the “common law of torts” to determine the “accrual” rules for the identified constitutional claim. Manuel, 580 U.S. at 370. It has explained that the common law “presumptively” adopted what the majority opinion calls the “occurrence” rule— not any type of discovery rule. See McDonough, 588 U.S. at 115. That is, the common law typically started the limitations period on the first day that plaintiffs had “a complete and present No. 23-5925 Reguli v. Russ, et al. Page 16
cause of action[.]” Id. (quoting Wallace, 549 U.S. at 388). Plaintiffs have such a cause of action when they can seek relief in court because a claim’s legal elements have all arisen (whether or not they know of this fact). See Wallace, 549 U.S. at 388; see also Petrella v. Metro-Goldwyn- Mayer, Inc., 572 U.S. 663, 670 & n.4 (2014). Plenty of evidence supports this understanding of the common law. Treatises, for example, routinely suggested that a “cause of action or suit arises when and as soon as the party has a right to apply to the proper tribunals for relief.” J.K. Angell, Treatise on the Limitations of Actions 34 (5th ed. 1869); see also 1 H.G. Wood, Limitation of Actions § 122a, at 684–85 (4th ed. 1916); Arthur Underhill, Principles of the Law of Torts 65–66 (1st Am. ed. 1881).
To be clear, the Court has recognized only a presumption in favor of this occurrence rule. It has applied the rule when nothing supported a departure from it. See Reed, 598 U.S. at 235– 36; McDonough, 588 U.S. at 116–20. But sometimes a constitutional claim (such as an unreasonable-seizure claim) might resemble a tort (such as a false-imprisonment claim) that followed a “distinctive” accrual test at common law. Wallace, 549 U.S. at 389. In that scenario, the Court might extend this distinctive test (not the occurrence rule) to the analogous constitutional claim. See id. at 389–90. And besides, the Court has added that the common law provides only a “guide” that courts may adjust to best fit the relevant right. Manuel, 580 U.S. at 370.
2. Now consider our approach. We first adopted a “discovery” rule for § 1983’s statute of limitations in a 1984 case about the Fifth Amendment privilege against self-incrimination. See Sevier v. Turner, 742 F.2d 262, 267, 273 (6th Cir. 1984). Bereft of any reasoning that would justify our choice of a discovery rule over an occurrence rule, Sevier devoted all of one sentence to this topic: “The statute of limitations commences to run when the plaintiff knows or has reason to know of the injury which is the basis of his action.” Id. at 273. Its support? Four out- of-circuit decisions. See id. Since then, we have cited Sevier’s lone sentence as our source to expand the discovery rule—in seemingly automatic fashion—to many other constitutional claims that plaintiffs have pursued under § 1983. See, e.g., Rodriguez v. City of Cleveland, 439 F. App’x 433, 458 (6th Cir. 2011) (Fourth Amendment); Eidson v. Tenn. Dep’t of Child.’s Servs., 510 F.3d 631, 635 (6th Cir. 2007) (Due Process Clause); Cooey v. Strickland, 479 F.3d No. 23-5925 Reguli v. Russ, et al. Page 17
412, 415–16 (6th Cir. 2007) (Eighth Amendment); Sharpe v. Cureton, 319 F.3d 259, 265–66 (6th Cir. 2003) (First Amendment); Kuhnle Bros. v. County of Geauga, 103 F.3d 516, 520 (6th Cir. 1997) (Takings Clause); Dixon v. Anderson, 928 F.2d 212, 215 (6th Cir. 1991) (Equal Protection Clause).
Consider a few ways that our current approach to this accrual question differs from the Supreme Court’s. To begin with, the Supreme Court has told us to start with the “specific constitutional right” at issue when determining the proper accrual test. See Reed, 598 U.S. at 236. But most of our cases start—and end—by adopting a discovery rule without considering the relevant constitutional right. Indeed, many opinions do not even identify the right that the § 1983 plaintiff sought to vindicate in the suit. See Johnson v. Memphis Light Gas & Water Div., 777 F.3d 838, 843 (6th Cir. 2015); Bishop v. Child.’s Ctr. for Developmental Enrichment, 618 F.3d 533, 536–37 (6th Cir. 2010); Roberson v. Tennessee, 399 F.3d 792, 794–96 (6th Cir. 2005); Hughes v. Vanderbilt Univ., 215 F.3d 543, 548–50 (6th Cir. 2000).
Next, the Supreme Court “presumptively” follows the occurrence rule in § 1983 cases both because the Court looks to “common-law tort principles” to interpret § 1983 and because States generally followed this rule to decide when the statute of limitations began for their torts. McDonough, 588 U.S. at 115 (quoting Wallace, 549 U.S. at 388). We, by contrast, have automatically (not presumptively) adhered to a discovery rule in our § 1983 cases. And none of our cases has even examined “common-law tort principles,” let alone attempted to ground our broad-brush discovery rule in those principles. Wallace, 549 U.S. at 388.
In a case that did not involve § 1983, we did suggest that our discovery rule in the § 1983 context could be seen as adhering to “a common-law accrual principle[.]” Snyder-Hill v. Ohio State Univ., 48 F.4th 686, 700 (6th Cir. 2022). But we cited no support for this dictum. And § 1983 does not give us license to adopt the rules that we think best as a policy matter. Rather, the Supreme Court has told us to interpret the statute in light of the “common-law principles that were well settled at the time of its enactment.” Nieves v. Bartlett, 587 U.S. 391, 405 (2019) (quoting Kalina v. Fletcher, 522 U.S. 118, 123 (1997)) (emphasis added). As far as I can tell, a broad discovery rule conflicts with these traditional principles. Even as late as 1939, the Restatement of Torts noted that many States continued to adhere to the view that “the statutory No. 23-5925 Reguli v. Russ, et al. Page 18
period runs from the time the tort was committed although the injured person had no knowledge or reason to know of it.” Restatement (First) of Torts § 899 cmt. (e) (Am. L. Inst. 1939) (emphasis added).
At the same time, the Supreme Court’s instructions that we should follow “distinctive” accrual rules when the common law would adopt them likely leaves some room for a discovery rule. Wallace, 549 U.S. at 389. Most notably, the common law at the time of § 1983’s enactment adopted a discovery rule that delayed the running of the statute of limitations when the plaintiff asserted a fraud claim. See Gabelli v. SEC, 568 U.S. 442, 449 (2013); Holmberg v. Armbrecht, 327 U.S. 392, 397 (1946); Bailey v. Glover, 88 U.S. 342, 348–49 (1874); Sherwood v. Sutton, 21 F. Cas. 1303, 1304–08 (C.C.D.N.H. 1828) (Story, J.). If a constitutional claim under § 1983 resembles such a fraud claim, this “distinctive” discovery rule might apply. Wallace, 549 U.S. at 389. Other unique rules might also exist. Even for non-fraud torts, for example, some common-law sources suggested that a defendant’s fraudulent effort to hide the tort could delay the start of the statute of limitations. See Hugh Fraser, Compendium of the Law of Torts 130 (1888). Perhaps § 1983 would incorporate this fraudulent-concealment rule. That said, the Supreme Court has at times described this rule as a tolling (not an accrual) doctrine. See Klehr v. A.O. Smith Corp., 521 U.S. 179, 194 (1997). And the Court has suggested that § 1983 incorporates state-by-state tolling rules rather than a catch-all federal rule (as in the accrual context). See Wallace, 549 U.S. at 395.
I will not belabor the point further. I hope that what I have said suffices to show the tension between our precedent and the Supreme Court’s. Eventually, the distinction between our automatic use of the discovery rule and the Supreme Court’s presumptive use of the occurrence rule will matter to the outcome of a § 1983 case. When it does, a panel must address whether our prior decisions bind us to the discovery rule as a precedential matter despite the Supreme Court’s caselaw. If a panel finds itself bound, perhaps the en banc court should take a fresh look at our discovery rule. After all, that rule arose long ago in a conclusory fashion well before the Supreme Court’s more recent (and reasoned) guidance on this issue.