RECOMMENDED FOR FULL-TEXT PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 18a0272p.06
UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT
RAMON JASSO ARANGURE, ┐ Petitioner, │ │ > No. 18-3076 v. │ │ │ MATTHEW G. WHITAKER, Acting Attorney General, │ Respondent. │ ┘
On Petition for Review from the Board of Immigration Appeals; No. A 056 333 337.
Argued: November 27, 2018
Decided and Filed: December 18, 2018
Before: THAPAR, BUSH, and NALBANDIAN, Circuit Judges.
_________________
COUNSEL
ARGUED: Benjamin Casper Sanchez, Paul Dimick, Zachary Hofeld, UNIVERSITY OF MINNESOTA, Minneapolis, Minnesota, for Petitioner. Song E. Park, UNITED STATES DEPARTMENT OF JUSTICE, Washington, D.C., for Respondent. ON BRIEF: Benjamin Casper Sanchez, UNIVERSITY OF MINNESOTA, Minneapolis, Minnesota, Russell Abrutyn, ABRUTYN LAW, PLLC, Berkley, Michigan, for Petitioner. Song E. Park, UNITED STATES DEPARTMENT OF JUSTICE, Washington, D.C., for Respondent. Javier N. Maldonado, LAW OFFICE OF JAVIER N. MALDONADO, San Antonio, Texas, Khaled Alrabe, NATIONAL IMMIGRATION PROJECT OF THE NATIONAL LAWYERS GUILD, Boston, Massachusetts, for Amici Curiae. No. 18-3076 Jasso Arangure v. Whitaker Page 2
OPINION _________________
THAPAR, Circuit Judge. Courts have always had an “emphatic[]” duty “to say what the law is.” Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177 (1803). But all too often, courts abdicate this duty by rushing to find statutes ambiguous, rather than performing a full interpretive analysis. When dealing with agencies, this abdication by ambiguity is even more tempting—and even more problematic. Because, under Chevron, ambiguity means courts get to outsource their “emphatic” duty by deferring to an agency’s interpretation. But even Chevron itself reminds courts that they must do their job before applying deference: they must first exhaust the “traditional tools” of statutory interpretation and “reject administrative constructions” that are contrary to the clear meaning of the statute. Chevron USA, Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 843 n.9 (1984). First and foremost, this means courts must analyze the statutory text. But when the text standing alone does not supply an answer, courts must consider canons of interpretation. Here, a canon makes the statute’s meaning clear. Thus, we reject the agency’s contrary interpretation.
I.
In 2003, the United States granted Jasso lawful permanent resident status.1 Over a decade later, he pled guilty to first-degree home invasion in Michigan. See Mich. Comp. Laws (MCL) § 750.110a(2). Shortly thereafter, DHS began a removal proceeding.
DHS argued that Jasso’s home-invasion conviction was a “crime of violence,” making him removable under the Immigration and Nationality Act (“INA”). See 8 U.S.C. §§ 1101(a)(43)(F), 1227(a)(2)(A)(iii). At the time, the statute defined a “crime of violence” with both an elements clause and a residual clause. 18 U.S.C. § 16. The Immigration Judge found that Jasso’s home-invasion conviction was a crime of violence under the residual clause. Jasso appealed to the Board of Immigration Appeals (“Board”), but, in the interim, this court found the residual clause unconstitutionally vague. Shuti v. Lynch, 828 F.3d 440, 446 (6th Cir. 2016).
1 In his brief, Ramon Jasso Arangure refers to himself as Jasso, so we do the same. No. 18-3076 Jasso Arangure v. Whitaker Page 3
Since Jasso’s removal order hinged on the residual clause, the Board remanded to the Immigration Judge for a new removability determination. In light of Shuti, the judge terminated the proceeding. In doing so, he explained that the termination was “without prejudice” and warned Jasso that DHS could still “recharge under a different theory.” AR 134.
DHS accepted the invitation two days later and initiated a second removal proceeding against Jasso relying on a different statutory subsection. This time DHS argued that Jasso’s home-invasion conviction was a “burglary offense” rather than a “crime of violence.” 8 U.S.C. §§ 1101(a)(43)(G), 1227(a)(2)(A)(iii). The Immigration Judge agreed and also rejected Jasso’s argument that res judicata barred the second proceeding. The Board affirmed, concluding that the doctrine of res judicata does not apply in removal proceedings involving aggravated felons (hereinafter “removal proceedings”). Matter of Jasso Arangure, 27 I&N Dec. 178, 186 (BIA 2017).
Now Jasso appeals to this court. His appeal raises three issues: (1) whether the doctrine of res judicata applies in removal proceedings (which requires an assessment of whether the Board’s conclusion is entitled to Chevron deference); (2) if res judicata does apply, whether the elements are met here such that the second removal proceeding is barred; and (3) if res judicata does not bar the second removal proceeding, whether it was right on the merits (i.e., whether Jasso’s home-invasion conviction qualifies as a “burglary offense” under the INA). We have jurisdiction to review the questions of law raised in Jasso’s petition, 8 U.S.C. § 1252(a)(2)(D), and do so de novo, Sad v. INS, 246 F.3d 811, 814 (6th Cir. 2001).
II.
Res judicata “preclude[s] parties from contesting matters that they have had a full and fair opportunity to litigate.” Montana v. United States, 440 U.S. 147, 153 (1979). While there are two types of res judicata—issue preclusion and claim preclusion, Taylor v. Sturgell, 553 U.S. 880, 892 (2008)—only claim preclusion is relevant here. Claim preclusion prevents a party from litigating matters that should have been raised in an earlier case but were not. Id. Jasso argues that claim preclusion should have barred the second removal proceeding against him. The INA does not specify whether res judicata governs removal proceedings, so Jasso relies on the No. 18-3076 Jasso Arangure v. Whitaker Page 4
common-law presumption canon: courts presume that general statutory language incorporates common-law principles—like res judicata—unless there is a clear indication to the contrary. Astoria Fed. Sav. & Loan Ass’n v. Solimino, 501 U.S. 104, 108 (1991). The Board rejected this argument below, concluding that res judicata does not apply in removal proceedings because of “Congress’ clear intent to remove criminal aliens.” Matter of Jasso, 27 I&N Dec. at 183. On appeal, the government argues that the Board’s interpretation is entitled to deference. See Chevron, 467 U.S. at 844.
A.
The Board is eligible for Chevron deference when it interprets the INA. Negusie v. Holder, 555 U.S. 511, 516–17 (2009). But eligibility is not entitlement. Courts must assess whether an agency is actually entitled to Chevron deference through a two-part test. City of Arlington v. FCC, 569 U.S. 290, 296 (2013) (citing Chevron, 467 U.S. at 842–43). First, courts must determine whether the statute is ambiguous, “applying the ordinary tools of statutory construction.”2 Id. If the statute is unambiguous, then the court applies it as-written; “that is the end of the matter.” Id. If the statute is ambiguous, however, then the court moves to step two: defer to the agency’s construction if it is “permissible”—i.e., “within the bounds of reasonable interpretation.” Id.
Chevron’s first step is grounded in a recognition that “[t]he judiciary is the final authority on issues of statutory construction.” Chevron, 467 U.S. at 843 n.9. This means courts must do their best to determine the statute’s meaning before giving up, finding ambiguity, and deferring to the agency. When courts find ambiguity where none exists, they are abdicating their judicial duty. Cf. Kent Barnett & Christopher J. Walker, Chevron in the Circuit Courts, 116 Mich. L. Rev. 1, 33–34 (2017) (concluding that circuit courts find ambiguity at Chevron step one 70% of the time, based on a sample of over 1,000 cases). This abdication by ambiguity impermissibly
2 The Chevron Court said that statutory silence also triggers deference. 467 U.S. at 843. But Chevron’s theory of implicit delegation only applies to certain kinds of silences—those where we can plausibly infer Congress intentionally left a statutory gap for the agency to fill. Id. at 843–44. As discussed below, that is not the silence at issue in this case. Here, it is statutory silence in the face of existing common law. No. 18-3076 Jasso Arangure v. Whitaker Page 5
expands an already-questionable Chevron doctrine.3 See Voices for Int’l Bus. & Educ., Inc. v. NLRB, 905 F.3d 770, 780–81 (5th Cir. 2018) (Ho, J., concurring) (“Finding ambiguity where it does not exist—granting deference where it is not warranted . . . misuse[s] Chevron” and “abrogates separation of powers without even the fig leaf of Congressional authorization.”). Unsurprisingly, when courts neglect their duty, the Supreme Court has not hesitated to reverse. See, e.g., Pereira, 138 S. Ct. at 2113–14 (“[T]he Court need not resort to Chevron deference, as some lower courts have done, for Congress has supplied a clear and unambiguous answer to the interpretive question at hand.”); id. at 2120 (Kennedy, J., concurring) (chiding lower courts for “engag[ing] in cursory analysis” in Chevron step one and rushing to “reflexive deference”); Kingdomware Techs., Inc. v. United States, 136 S. Ct. 1969, 1979 (2016) (reversing lower court’s Chevron-based decision because the statute was unambiguous); United States v. LaBonte, 520 U.S. 751, 762 (1997) (same). In short, Chevron’s first step is not a free pass.
Thus, we must faithfully apply the “traditional tools of statutory construction” before concluding that the INA is ambiguous. Chevron, 467 U.S. at 843 n.9. That starts with an analysis of the statutory text. But here, the INA’s text is silent as to res judicata. Silence, however, does not necessarily connote ambiguity, nor does it automatically mean that a court can proceed to Chevron step two. See Ry. Labor Execs.’ Ass’n v. Nat’l Mediation Bd., 29 F.3d 655, 671 (D.C. Cir. 1994) (en banc). “[L]egal interpretation [is] more than just a linguistic exercise”—it includes the canons. Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts xxvii (2012). Canons are general background principles that courts have developed over time to guide statutory interpretation—the “interpretive rules of the road.” Kavanaugh, supra, at 2121; see also Black’s Law Dictionary (10th ed. 2014).
The common-law presumption canon is at issue here. Under this canon, courts presume that general statutory language incorporates established common-law principles (like res
3 Many members of the Supreme Court have called Chevron into question. See Pereira v. Sessions, 138 S. Ct. 2105, 2121 (2018) (“Given the concerns raised by some Members of this Court . . . it seems necessary and appropriate to reconsider, in an appropriate case, the premises that underlie Chevron and how courts have implemented that decision.”) (Kennedy, J., concurring) (internal citations omitted) (citing City of Arlington, 569 U.S. at 327 (Roberts, C.J., dissenting); Michigan v. EPA, 135 S. Ct. 2699, 2712–14 (2015) (Thomas, J., concurring); Gutierrez–Brizuela v. Lynch, 834 F.3d 1142, 1149–58 (10th Cir. 2016) (Gorsuch, J., concurring)); see also Brett M. Kavanaugh, Fixing Statutory Interpretation, 129 Harv. L. Rev. 2118, 2144, 2153–54 (2016) (book review) (advocating for courts to strive for the “best reading” of a statute rather than find ambiguity). No. 18-3076 Jasso Arangure v. Whitaker Page 6
judicata) unless “a statutory purpose to the contrary is evident.” Astoria, 501 U.S. at 108 (quoting Isbrandtsen Co. v. Johnson, 343 U.S. 779, 783 (1952)). If this canon applies here, then there is no ambiguity, and Jasso is correct that res judicata governs removal proceedings. But if the canon does not apply, then the INA is ambiguous, and Chevron deference would resolve the statute’s silence against Jasso. Thus, we must first address whether the common-law presumption canon is a “traditional tool” that applies in step one. If it is, then the INA is not ambiguous despite its silence, and no Chevron deference is needed.
B.
Unfortunately, the Supreme Court has never explicitly described which canons are “traditional tools” and which are not. Thus, “[t]he relationship between Chevron deference and the canons . . . remains ‘one of the most uncertain aspects of the Chevron doctrine.’” Caleb Nelson, Statutory Interpretation and Decision Theory, 74 U. Chi. L. Rev. 329, 347 (2007) (book review) (quoting Curtis A. Bradley, Chevron Deference and Foreign Affairs, 86 Va. L. Rev. 649, 675 (2000)); see also Abbe R. Gluck, What 30 Years of Chevron Teach us About the Rest of Statutory Interpretation, 83 Fordham L. Rev. 607, 618–19 (2014). This lack of instruction has led to some uncertainty in the lower courts. Compare Valenzuela Gallardo v. Lynch, 818 F.3d 808, 816 (9th Cir. 2016) (“[T]he canon of constitutional avoidance is highly relevant at Chevron step one.” (internal quotation marks omitted)), with Olmos v. Holder, 780 F.3d 1313, 1321 (10th Cir. 2015) (“[T]he canon of constitutional avoidance does not bear on our inquiry at [Chevron] step one.”); compare Rancheria v. Jewell, 776 F.3d 706, 713 (9th Cir. 2015) (“[A]n agency’s legal authority to interpret a statute appears to trump any practice of construing ambiguous statutory provisions in favor of Indians.”), with Cobell v. Norton, 240 F.3d 1081, 1100–01 (D.C. Cir. 2001) (opposite). But this uncertainty has a solution. Though the Supreme Court has not explicitly described which canons qualify as “traditional tools,” it has decided a number of cases that provide a good outline to guide the analysis.
The Supreme Court has repeatedly applied canons at step one. See, e.g., Epic Sys. Corp. v. Lewis, 138 S. Ct. 1612, 1617, 1625–30 (2018) (ejusdem generis, expressio unius, and the presumption against implied repeals); Nat’l Ass’n of Home Builders v. Defs. of Wildlife, 551 U.S. 644, 666 (2007) (presumption against implied repeals); INS v. St. Cyr, 533 U.S. 289, 319 n.45 No. 18-3076 Jasso Arangure v. Whitaker Page 7
(2001) (presumption against retroactivity); Solid Waste Agency of N. Cook Cty. v. U.S. Army Corps of Eng’rs, 531 U.S. 159, 172–73 (2001) (constitutional avoidance canon and presumption against preemption); Nat’l Credit Union Admin. v. First Nat’l Bank & Tr. Co., 522 U.S. 479, 501–02 (1998) (presumption of consistent usage); Dole v. United Steelworkers of Am., 494 U.S. 26, 35–36, 42–43 (1990) (noscitur a sociis); Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Const. Trades Council, 485 U.S. 568, 574–75 (1988) (constitutional avoidance canon). These cases suggest canons are “traditional tools” of statutory interpretation that take precedence over Chevron deference. Indeed, the Court has come close to saying exactly that. St. Cyr, 533 U.S. at 320 n.45 (“Because a statute that is ambiguous with respect to retroactive application is construed under our precedent to be unambiguously prospective . . . there is, for Chevron purposes, no ambiguity in such a statute for an agency to resolve.” (internal citation omitted)). And recently, the Supreme Court flatly stated: if “the canons supply an answer, ‘Chevron leaves the stage.’” Epic, 138 S. Ct. at 1630 (quoting NLRB v. Alt. Entm’t, Inc., 858 F.3d 393, 417 (6th Cir. 2017) (Sutton, J., concurring in part and dissenting in part)). One commentator surveyed the relevant precedent and deemed the Court’s approach “Canons Trump Deference.” Kenneth A. Bamberger, Normative Canons in the Review of Administrative Policymaking, 118 Yale L.J. 64, 77 (2008); see also Caleb Nelson, Statutory Interpretation 729–33 (1st ed. 2010). It seems that the Court has a “canons first” rule, even if it has not said so explicitly.4
Several reasons support the idea that canons belong in Chevron step one. First and foremost, canons can assist a court in determining a statute’s meaning. Caleb Nelson, What is Textualism?, 91 Va. L. Rev. 347, 383–86 (2005). Canons that serve this goal can be thought of as “descriptive”—in contrast to “normative” canons that are instead designed to favor certain substantive policies. Id. at 393–94; see also Black’s Law Dictionary (10th ed. 2014); Jonathan D. Urick, Note, Chevron and Constitutional Doubt, 99 Va. L. Rev. 375, 409 (2013). Within the
4 The Supreme Court has not expressly addressed the hierarchy between canons in the Chevron analysis (or outside it), i.e., which canons lower courts should consider before others. And the legitimacy of some canons has been called into question. For instance, then-Professor Amy Coney Barrett developed thoughtful arguments for why some canons are more consistent with the judicial role. And she suggests other canons should be avoided altogether. See generally Amy Coney Barrett, Substantive Canons and Faithful Agency, 90 B.U. L. Rev. 109 (2010). In this case, the parties have not asked us to consider the legitimacy of the canons—and for good reason. The proper court to evaluate the legitimacy of canons is the Supreme Court. Given the appropriate opportunity, the Court should consider Judge Barrett’s thoughtful critique. No. 18-3076 Jasso Arangure v. Whitaker Page 8
family of descriptive canons, linguistic canons are the strongest species. These canons simply “reflect[] accepted notions of diction, grammar, and syntax.” Black’s Law Dictionary (10th ed. 2014). Examples include the expressio unius canon (the expression of one thing implies the exclusion of others); the general/specific canon (a specific provision prevails over a general one if the two are in conflict); and the presumption of consistent usage (“Courts presume that the same words in the same statute mean the same thing.”). In re Jackson Masonry, LLC, 906 F.3d 494, 501 (6th Cir. 2018); Scalia & Garner, supra, §§ 10, 25, 28. These are all just “ordinary principles that laymen as well as lawyers use to interpret communications.” Nelson, What is Textualism, supra, at 383. Accordingly, linguistic canons are just specific applications of the basic goal of interpretation: finding the ordinary meaning of statutory text. These canons clearly and exclusively serve descriptive, rather than normative, purposes. Therefore, they belong in step one, which has the same goal: determining the meaning of the statute. Chevron, 467 U.S. at 842–43. On this, there is broad agreement. See Cass R. Sunstein, Law and Administration After Chevron, 90 Colum. L. Rev. 2071, 2120 (1990) (“When the relevant interpretive norm is part of an effort to discern legislative instructions, Chevron is uncontroversially subordinate to that norm.”).
Canons based on legislative practice rather than linguistics can also be viewed as descriptive. When the statute’s text (as interpreted using linguistic canons) provides no clear answer to an interpretive question, courts presume that the legislature intends to follow its typical patterns of behavior. Nelson, What is Textualism, supra, at 389. These canons operate like default rules for interpreting contracts. For example, the reenactment canon: if Congress amends or reenacts a provision, a significant change in language is presumed to entail a change in meaning. Scalia & Garner, supra, § 40. The presumption against retroactivity is another example: based on longstanding historical practice and assumptions about our constitutional order, courts presume that legislation is not meant to cover events that occurred before its enactment. Id. at § 41. To the extent these sorts of canons have descriptive value, that is a sufficient basis for putting them in Chevron’s first step.
Descriptive value, however, may be an insufficient explanation for why the Court has consistently applied these legislative-practice canons in step one. To start, such canons are on No. 18-3076 Jasso Arangure v. Whitaker Page 9
less solid descriptive footing than their linguistic counterparts. See Sunstein, Law and Administration, supra, at 2109–10. For instance, the canon of constitutional avoidance “[p]erhaps . . . was originally based[] on a genuine assessment of probable meaning” but “[a] more plausible basis for the rule is that it represents judicial policy . . . that courts should minimize the occasions on which they confront and perhaps contradict the legislative branch.” Scalia & Garner, supra, § 38; see also Kavanaugh, supra, at 2145–49 (arguing that the constitutional avoidance canon should be “jettisoned” entirely). But see Urick, supra, at 376 (arguing that the constitutional avoidance canon should be deployed before Chevron deference based on a textualist account of the canon “rooted in historical practice”). Second, to the extent these canons have descriptive value, it is often hard to disentangle their descriptive motivations from their normative ones. See, e.g., Nelson, What is Textualism, supra, at 395 n.143 (noting that the presumption against retroactivity “is partly descriptive, but normative judgments about the unfairness of retroactive legislation probably give it some extra force”). So, for these canons, their potential descriptive value may be insufficient, standing alone, to justify deploying them in step one. Nonetheless, the Supreme Court has repeatedly invoked them there. See, e.g., Solid Waste Agency, 531 U.S. at 172–73 (constitutional avoidance canon and presumption against preemption); St. Cyr, 533 U.S. at 319 n.45 (presumption against retroactivity); DeBartolo Corp., 485 U.S. at 574–75 (constitutional avoidance canon). To understand why, we need to consider the justifications for Chevron deference itself. (Justifications that, for the sake of argument, are taken as a given. But see supra note 3.)
One purported rationale for Chevron deference is that agencies are more likely to get the answer right, given their expertise. Chevron, 467 U.S. at 844–45, 865. But many canons reverse that rationale: courts are more likely to get the answer right. For instance, canons often call for an analysis of the statute as a whole in light of some other source of legal authority (common law, other statutes, or the Constitution). Courts are better equipped and more experienced than agencies in answering these types of questions. See Epic, 138 S. Ct. at 1629 (applying the presumption against implied repeals at Chevron step one because “the ‘reconciliation’ of distinct statutory regimes ‘is a matter for the courts,’ not agencies” (quoting Gordon v. N.Y. Stock Exch. Inc., 422 U.S. 659, 685–86 (1975))). And these questions are far afield from the prototypical Chevron situation: an agency’s application of law to fact. See INS v. Cardoza-Fonseca, 480 No. 18-3076 Jasso Arangure v. Whitaker Page 10
U.S. 421, 446–48 (1987) (stating that whether two statutory standards are identical “is a pure question of statutory construction for the courts to decide . . . quite different from the question of interpretation that arises in each case in which the agency is required to apply either or both standards to a particular set of facts”). In short, courts, not agencies, are the true experts in applying many of the canons.
Another rationale for Chevron deference is that statutory ambiguities are an implicit delegation to agencies. Chevron, 467 U.S. at 843–44. In contrast, many canons can be understood to represent important choices that Congress generally does not delegate to agencies and instead must answer for itself—“nondelegation canons.” Cass R. Sunstein, Nondelegation Canons, 67 U. Chi. L. Rev. 315, 330 (2000); see also Sunstein, Law and Administration, supra, at 2111–15. These canons require a clear statement from Congress if it wants to impair certain fundamental values derived from the Constitution or “time-honored notions of equity and comity.” Sunstein, Law and Administration, supra, at 2114; see, e.g., St. Cyr, 533 U.S at 316 (“Requiring clear intent [that legislation be retroactive] assures that Congress itself has affirmatively considered the potential unfairness of retroactive application and determined that it is an acceptable price to pay for the countervailing benefits.” (quoting Landgraf v. USI Film Prods., 511 U.S. 244, 272–73 (1994))). In other words, the presumption against retroactivity means that Congress must make the ultimate choice on whether a statute will apply retroactively—either by explicitly saying “yes” or, through silence, implicitly saying “no.” Silence in this situation is a congressional choice, not a delegation to the agency. Accordingly, the Supreme Court has applied these “nondelegation canons” at step one to cabin agency discretion.
In sum, it is unsurprising that the Court has repeatedly applied canons in step one of the Chevron framework. Canons can (i) be useful tools in assessing statutory meaning, (ii) require analysis that courts are better at, and (iii) protect fundamental constitutional or historical values. Most well-established canons have at least one of these characteristics, and many combine several of them. But, as tempting as a simple rule might be, we cannot say that all canons belong in step one. A canon that reflects a pure policy judgment (i.e., of limited-to-no-descriptive value) that is well within an agency’s wheelhouse might not belong there, at least under current No. 18-3076 Jasso Arangure v. Whitaker Page 11
precedent. (Nor would it belong anywhere in a textualist’s interpretive process, for that matter.) Still, these canons should be viewed as the exception, not the rule. Based on the Court’s precedent, most canons are “traditional tools” of statutory interpretation that should be deployed in Chevron step one.
C.
The common-law presumption canon qualifies as a “traditional tool” of statutory interpretation. This canon presumes that the existing common law still applies unless the statute clearly indicates otherwise. United States v. Texas, 507 U.S. 529, 534 (1993); see also Scalia & Garner, supra, § 52. It is more descriptive than normative. The canon is not based on a normative judgment that the common law is better as a matter of policy. Rather, it is based on a descriptive judgment: Congress legislates against a common-law backdrop and presumably does not intend to reject that backdrop with general statutory language. See Astoria, 501 U.S at 108 (“Courts do not, of course, have free rein to impose rules of preclusion, as a matter of policy[;]” instead “the question is . . . whether it is intended by the legislature.”) Indeed, it is hard to imagine an interpretive “tool” more “traditional” than the centuries-old common-law presumption. See Fairfax’s Devisee v. Hunter’s Lessee, 11 U.S. (7 Cranch) 603, 623 (1812) (“The common law, therefore, ought not to be deemed to be repealed, unless the language of a statute be clear and explicit for this purpose.”); see also Riggs v. Palmer, 115 N.Y. 506, 511 (1889) (“[A]ll laws, as well as all contracts, may be controlled in their operation and effect by general, fundamental maxims of the common law.”). Finally, the canon calls for an analysis that courts are well-equipped to undertake: examining a common-law doctrine (and the reasons for it) and seeing whether it fits within the overall structure of a statute. In sum, the principles underlying the Court’s general “canons first” approach apply to the common-law presumption canon.
Thus, unsurprisingly, the common-law presumption is analogous to other canons that the Court has applied in Chevron step one. The presumption against implied repeals is one example. See, e.g., Epic, 138 S. Ct. at 1629. Courts presume that a statute does not repeal a prior statute in the same way (and for many of the same reasons) they presume that it does not displace the common law. See Scalia & Garner, supra, § 52. Likewise, the specific variant of the canon at No. 18-3076 Jasso Arangure v. Whitaker Page 12
issue here—the presumption that res judicata applies—implicates similar fairness and reliance concerns to the presumption against retroactivity. The presumption against retroactivity is “deeply rooted in our jurisprudence” because “[e]lementary considerations of fairness dictate that individuals should have an opportunity to know what the law is and to conform their conduct accordingly; settled expectations should not be lightly disrupted.” St. Cyr, 533 U.S. at 316 (quoting Landgraf, 511 U.S. at 265). In the same way, a concluded case tells the parties how the law applies to them, “settl[ing] expectations” and permitting them to “conform their conduct accordingly.” Id. Thus, in most instances it is disruptive and unfair to revive a case that has already finished. Res judicata helps avoid this outcome. See Federated Dep’t Stores, Inc. v. Moitie, 452 U.S. 394, 401 (1981) (“This Court has long recognized that . . . those who have contested an issue shall be bound by the result of the contest, and that matters once tried shall be considered forever settled as between the parties . . . [this is] a rule of fundamental and substantial justice.”). For all of these reasons, the common-law presumption canon is a perfect fit for Chevron step one.
D.
Applying the common-law presumption canon renders the INA unambiguous on the question this case presents. To restate the canon: because res judicata is a well-established common-law principle, it presumptively applies to an administrative adjudicatory scheme set up by a statute unless a “purpose to the contrary is evident.” Astoria, 501 U.S. at 108 (quoting Isbrandtsen, 343 U.S. at 783). Congress must make the contrary statutory purpose clear, either through explicit text or through an obvious inference from the statute’s structure.
Two cases illustrate what it takes to override the common-law presumption of res judicata. First, in Astoria, the Court found that Congress overrode the presumption in the Age Discrimination in Employment Act. Id. at 112–14; see 29 U.S.C. § 621 et seq. The Act required plaintiffs to exhaust state administrative remedies before filing a complaint in federal court. 29 U.S.C. § 633(b). The Court concluded that, if these administrative findings were given res judicata effect, later federal proceedings “would be strictly pro forma.” Id. at 111. State winners would not file in federal court because they already got what they wanted, and state losers would have their federal cases summarily dismissed because of res judicata. Id. Interpreting the No. 18-3076 Jasso Arangure v. Whitaker Page 13
statute’s exhaustion requirement in this way would violate the presumption against superfluity— rendering the federal right of action a nullity. Id. at 112.
A more recent case provides a good counterexample: B&B Hardware, Inc. v. Hargis Indus., 135 S. Ct. 1293 (2015). There, the Court concluded that res judicata principles apply to decisions of the Trademark Trial and Appeal Board (“Trademark Board”), an administrative body that adjudicates trademark registration disputes under the Lanham Act. Id. at 1299–300. The Court concluded that nothing in the text or structure of the statute overrode the presumption that common-law res judicata principles apply. Id. at 1305–06. Unlike the statute in Astoria, the Lanham Act does not make Trademark Board suits a prerequisite to federal claims. So giving preclusive effect to Trademark Board decisions would not render federal suits “strictly pro forma.” Id. at 1305 (quoting Astoria, 501 U.S. at 111). The court also rejected a more generalized policy argument that res judicata would lead to increased Trademark Board litigation that would “bog[] down the [trademark] registration process.” Id. at 1306.5
This case is more like B&B Hardware than Astoria. The immigration courts are a replacement, not a prerequisite, for a full adjudication in Article III courts. Removal proceedings are “plainly adjudicatory in character”—something the government does not seriously dispute. See Duvall v. Att’y Gen. of U.S., 436 F.3d 382, 390 (3d Cir. 2006). And finding that res judicata applies would not violate the presumption against superfluity: the government does not point to any provision that would be rendered a nullity or any other structural problems that res judicata would create. In fact, if anything, the INA’s structure militates against the government’s argument. DHS’s burden of proof in each removal proceeding—“clear and convincing” evidence—would be rendered “largely meaningless” if DHS could repeatedly bring one proceeding after another until it got the result it wanted. See id. at 387–88 (citing 8 U.S.C. § 1229a(b), (c)).
The government relies primarily on “the express intent of Congress to remove criminal noncitizens convicted of aggravated felony offenses,” which purportedly means that Congress
5 Because Jasso’s case concerns the preclusive effect of an administrative judgment on a subsequent administrative proceeding, rather than on a subsequent federal court proceeding, it does not implicate the Article III concerns expressed by the B&B Hardware dissenters. 135 S. Ct. at 1316 (Thomas, J., dissenting). No. 18-3076 Jasso Arangure v. Whitaker Page 14
would disfavor any obstacle to that goal, such as res judicata. Respondent Br. at 3–4.6 This argument illustrates the problems with purposivism; it suggests courts can simply ignore the enacted text and instead attempt to replace it with an amorphous “purpose” that happens to match with the outcome one party wants. But that has no limiting principle. Indeed, the government’s argument here would make the common-law presumption meaningless. Congress always wants the statutes it passes to be enforced, and res judicata could always be construed as an obstacle to that goal. So the government’s argument proves too much. And more fundamentally, statutes are motivated by many competing—and often contradictory—purposes. Congress addresses these purposes by negotiating, crafting, and enacting statutory text. It is that text that controls, not a court’s after-the-fact reevaluation of the purposes behind it.
In this case, the Chevron analysis begins and ends with step one. The common-law presumption of res judicata makes the INA unambiguous. Res judicata doctrine applies in removal proceedings.7
III.
The next question is whether res judicata—in this case claim preclusion—bars the second removal proceeding against Jasso. A party invoking claim preclusion must show a prior proceeding (1) litigated to a final judgment (2) arising out of the same factual occurrence as the current proceeding (3) involving the same parties (4) where the non-moving party could have raised the claim at issue. Wheeler v. Dayton Police Dep’t, 807 F.3d 764, 766 (6th Cir. 2015).
6 In making this argument, the government references the Fourth Circuit’s decision in Johnson v. Whitehead, 647 F.3d 120, 129 (4th Cir. 2010). That case arguably held that res judicata does not apply in removal proceedings, though it dealt specifically with issue preclusion rather than claim preclusion, and its holding may have been limited to the facts before it. Id. To the extent the Johnson court’s conclusion is inconsistent with ours, we respectfully reject it. In our view, that court did not fully consider the role of canons in the Chevron framework, and its analysis focused too much on the purposes behind the INA rather than the statute’s text and structure. 7 The government also argues that res judicata is an “extra procedural device[]” that agencies can use, or not use, at their discretion. See Vermont Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 546 (1978). But the Astoria line of cases makes it clear that the application of res judicata is a matter of statutory interpretation, not agency choice. See Channer v. DHS, 527 F.3d 275, 279–80 (2d Cir. 2008); Duvall, 436 F.3d at 389. For the same reason, we need not reach Jasso’s arguments for res judicata based on the immigration rule of lenity and the Due Process Clause. See Duvall, 436 F.3d at 387 (rejecting a similar due process-based argument). No. 18-3076 Jasso Arangure v. Whitaker Page 15
The government apparently does not dispute the latter two elements, and they are clearly met here. DHS and Jasso were the parties in both removal proceedings, and in the first removal proceeding DHS could have argued that Jasso’s conviction was a “burglary offense” instead of, or in addition to, a “crime of violence.” 8 U.S.C. § 1101(a)(43)(F)–(G). The parties only dispute whether the two proceedings arose out of the same factual occurrence and whether the second proceeding was litigated to a final judgment on the merits.
Factual occurrence. Here, in both the first and second removal proceedings, DHS claimed that Jasso was removable because of the same underlying fact: his Michigan home- invasion conviction. In arguing to the contrary, the government equates the underlying factual occurrence with the legal claims asserted based on it. True, in Jasso’s first removal proceeding DHS argued that the conviction was a “crime of violence” rather than a “burglary offense.” Id. But the two different statutory subsections were “two different theories of the case arising from the same factual situation”—the “precise” circumstances covered by claim preclusion. Wilkins v. Jakeway, 183 F.3d 528, 535 (6th Cir. 1999) (citing Restatement (Second) of Judgments §§ 24, 25 cmts. a, d). Further, it does not matter that the two theories relied on different aspects of the facts (whether Jasso’s conviction met the generic burglary definition versus whether it qualified as a “crime of violence”). Practically speaking, any two different legal theories will emphasize different aspects of the same facts or rely on some factual details rather than others. But a party’s “different shading[] of the facts” does not create two separate factual occurrences for claim preclusion. See Talismanic Props., LLC v. City of Tipp City, 742 F. App’x 129, 132 (6th Cir. 2018). Other circuits addressing claim preclusion in removal proceedings have taken the same view. See, e.g., Duhaney v. Att’y Gen. of U.S., 621 F.3d 340, 350 (3d Cir. 2010) (“[T]he relevant factual occurrence is the conviction or convictions giving rise to a charge of removability.”); Channer, 527 F.3d at 281. And indeed, a narrower view would make claim preclusion practically impossible to prove.
The government contends that several practical problems would arise from applying res judicata to removal proceedings. But these supposed problems all stem from a further misunderstanding of the factual occurrence. In a removal proceeding based on an “aggravated felony” under 8 U.S.C. § 1227(a)(2)(A)(iii), the factual occurrence is only the conviction (or No. 18-3076 Jasso Arangure v. Whitaker Page 16
convictions) that DHS chooses to raise. Thus, DHS does not have to “obtain all the conviction records for every conviction” and “bring all possible charges in connection with a single proceeding.” Respondent Br. at 26. Rather, res judicata merely requires that DHS marshal all the legal arguments it intends to make based on the convictions it has raised. If DHS loses, claim preclusion bars it from trying again with different legal arguments. But DHS is not prevented from trying again based on different convictions (i.e., different factual occurrences), even if those convictions could have been raised in the first proceeding. Cf. 8 C.F.R. § 1003.30 (“At any time during deportation or removal proceedings, additional or substituted charges of deportability and/or factual allegations may be lodged by [DHS] in writing.” (emphasis added)). And obviously, if a noncitizen is convicted of a new crime, claim preclusion would not bar a new removal proceeding based on that new conviction. DHS’s doomsday predictions would only be true if the factual occurrence was “whether the immigrant is removable on any ground.” But claim preclusion is not applied at such an abstract level.
Since Jasso’s first and second removal proceedings arose out of the same conviction for Michigan first-degree home invasion, the proceedings arose out of the same factual occurrence.
Finality. Jasso meets three of the elements of claim preclusion, so this case comes down to the fourth: whether the first removal proceeding was litigated to a final judgment on the merits.
A final judgment on the merits is one “that signifies the ‘death knell’ of the litigation . . . permanently foreclos[ing] a party” from trying again. Mitchell v. Chapman, 343 F.3d 811, 821– 22 (6th Cir. 2003) (citation omitted). Indeed, since a decision may address the merits but still not have that effect, the Supreme Court has suggested that the “on the merits” language in the classic formulation of claim preclusion is misleading. Semtek Int’l Inc. v. Lockheed Martin Corp., 531 U.S. 497, 501–03 (2001). What really matters is the effect of the judgment. And, critically here, a judgment dismissing a case “without prejudice” is not truly “final.” Generally speaking, a dismissal “without prejudice” means “a dismissal without barring the plaintiff from returning later, to the same court, with the same underlying claim.” Id. at 505; see also Restatement (Second) of Judgements § 20(1)(b). Such a judgment does not “permanently foreclose[]” a litigant from trying again, so it is not sufficiently “final” to be given res judicata effect. No. 18-3076 Jasso Arangure v. Whitaker Page 17
See Mitchell, 343 F.3d at 821–22; see also Johnson v. De Grandy, 512 U.S. 997, 1005 (1994) (rejecting res judicata for a state court judgment rendered “without prejudice to the right” of any future claimant to sue).
Here, the effect of the first judgment is unclear. The written order itself simply stated that the charge of removability “[wa]s not sustained” and the proceeding was “terminated.” AR 41 (emphasis removed). The order did not explain whether the termination was with or without prejudice. But, at the hearing immediately before the order was issued, the Immigration Judge purported to clear up any ambiguity. Reviewing a draft of the order, counsel for DHS asked the judge whether it was a “termination without prejudice” and noted that he was contemplating “other potential charges.” AR 133–34. The judge responded:
It is without prejudice. It is only an analysis of the charge as originally presented. So the Government is in its normal position at this point. If there are other charges of removability that the Government wishes to prefer, it may do so.
Id. at 134; see also id. at 135–37. But on appeal, the Board simply assumed for the sake of argument that the termination was with prejudice. Matter of Jasso, 27 I&N Dec. at 179 n.3. Because of this posture, the parties did not fully address the finality of the first termination in their briefs before this court.
At oral argument, counsel for Jasso claimed that the Immigration Judge lacked the authority to terminate the first removal proceeding without prejudice. He relied primarily on Matter of S-O-G & F-D-B-, a decision issued after the briefs had been filed. 27 I&N Dec. 462 (A.G. 2018). But we read Matter of S-O-G differently. It held that immigration judges can only terminate removal proceedings in specific circumstances delineated by the relevant statutes and regulations, but it did not specifically address their authority to make those terminations with or without prejudice. Id. at 462. Though some of the decision’s reasoning could support Jasso’s reading, it did not squarely address the key issue for this appeal.
Given the Board’s avoidance of this issue and the limited briefing, we are unable to decide whether the first removal proceeding against Jasso was litigated to a final judgment. Consequently, we cannot conclude whether the requirements for claim preclusion have been met. No. 18-3076 Jasso Arangure v. Whitaker Page 18
* * *
The outcome of this case depends on whether claim preclusion applies. And the application of claim preclusion depends on whether the first removal proceeding was dismissed with or without prejudice—an issue unclear from the record before us and never addressed by the Board. Therefore, the appropriate course is to remand so that the Board can consider it in the first instance. See Ruiz-Del-Cid v. Holder, 765 F.3d 635, 641–42 (6th Cir. 2014). Accordingly, we VACATE the decision of the Board and REMAND for further proceedings consistent with this opinion.