CARLOS MONTEIRO SILVA, Petitioner, v. MERRICK B. GARLAND, Attorney General, Respondent.
No. 20-1593
United States Court of Appeals For the First Circuit
February 28, 2022
PETITION FOR REVIEW OF AN ORDER OF THE BOARD OF IMMIGRATION APPEALS
Before Howard, Chief Judge, Lynch and Barron, Circuit Judges.
Kerry E. Doyle, with whom Graves and Doyle was on brief, for petitioner.
Evan P. Schultz, Trial Attorney, Office of Immigration Litigation, Civil Division, with whom Brian Boynton, Acting Assistant Attorney General, Civil Division, and Stephen J. Flynn, Assistant Director, Office of Immigration Litigation, were on brief, for respondent.
Silva argues that the IJ and the BIA erred by applying the categorical approach to determine that his state conviction for accessory after the fact was “an offense relating to obstruction of justice.”
We make two holdings, each of which provides a basis for denying the petition. First, we follow the mode of analysis employed by the Supreme Court in Esquivel-Quintana v. Sessions, 137 S. Ct. 1562 (2017), and so apply “the normal tools of statutory interpretation,” id. at 1569. We hold the generic federal definition of “an offense relating to obstruction of justice” unambiguously does not require a nexus to a pending or ongoing investigation or judicial proceeding. Alternatively, we also hold, employing Chevron analysis, that the BIA‘s interpretation must be sustained. Consonant with these holdings, we conclude that the IJ and BIA properly concluded that Silva‘s Massachusetts conviction for accessory after the fact is categorically an offense relating to obstruction of justice and so rendered him removable as an aggravated felon. We deny Silva‘s petition using each mode of analysis.
I.
A. Massachusetts State Conviction
Silva is a native and citizen of Cape Verde who was admitted to the United States in 1989 as a lawful permanent resident.
In September 2017, Silva pleaded guilty in Massachusetts to accessory after the fact in violation of
[w]hoever, after the commission of a felony, harbors, conceals, maintains or assists the principal felon or accessory before
the fact, or gives such offender any other aid, knowing that he has committed a felony or has been accessory thereto before the fact, with intent that he shall avoid or escape detention, arrest, trial or punishment, shall be an accessory after the fact.
B. Removal Proceedings and BIA Decision
In January 2018, the U.S. Department of Homeland Security (“DHS“) initiated removal proceedings against Silva.3 In March 2019, DHS amended the charges of removal against Silva. Those amended charges stated that Silva was removable for, among other things, having been convicted of an aggravated felony in the form of “an offense relating to obstruction of justice . . . for which the term of imprisonment was at least one year.” See
In a written decision in May 2019, the IJ held that Silva‘s Massachusetts accessory-after-the-fact conviction was categorically an offense relating to obstruction of justice under the INA and so was a proper ground for removal as an aggravated felony.5 The IJ relied on the
Silva then filed applications for asylum, withholding of removal, and protection under the CAT. In a written decision in December 2019, the IJ denied all three forms of relief from removal. The IJ concluded that Silva‘s aggravated felony conviction for accessory after the fact rendered him statutorily ineligible for asylum, see
Silva appealed to the BIA. In May 2020, the BIA denied that appeal. The BIA, referencing Matter of Valenzuela Gallardo, adopted and affirmed the IJ‘s decision, holding that Silva‘s Massachusetts accessory-after-the-fact conviction was categorically an aggravated felony relating to obstruction of justice. The BIA also held that the IJ did not clearly err in finding that Silva‘s Massachusetts accessory-after-the-fact conviction was a particularly serious crime for purposes of withholding of removal, and affirmed the denial of all forms of relief.
This timely petition for review followed.
II.
Silva first argues that the Court should give no deference to the BIA‘s interpretation in Matter of Valenzuela Gallardo of “an offense relating to obstruction of justice” for various reasons. He also argues that, even if his Massachusetts accessory-after-the-fact conviction is an aggravated felony relating to obstruction of justice, the BIA erred in determining that the conviction is a “particularly serious crime” for purposes of barring him from withholding of removal.7
Where, as here, “the BIA adopts and affirms an IJ‘s decision, we review the IJ‘s decision ‘to the extent of the adoption, and the BIA‘s decision as to [any] additional ground.‘” Sunoto v. Gonzales, 504 F.3d 56, 59-60 (1st Cir. 2007) (alteration in original) (quoting Berrio-Barrera v. Gonzales, 460 F.3d 163, 167 (1st Cir. 2006)). We review de novo the BIA‘s legal conclusions, including its determination that Silva‘s Massachusetts accessory-after-the-fact conviction is an aggravated felony. See Lecky v. Holder, 723 F.3d 1, 4 (1st Cir. 2013). We review the BIA‘s factual findings under a deferential standard, upholding them “as long as they are ‘supported by reasonable, substantial, and probative evidence on the record considered as a whole.‘” Sanabria Morales v. Barr, 967 F.3d 15, 19 (1st Cir. 2020) (quoting Thapaliya v. Holder, 750 F.3d 56, 59 (1st Cir. 2014)).
A. Applying the Tools of Statutory Interpretation as Applied by the Supreme Court in Esquivel-Quintana, the Conviction for Accessory After the Fact Is an Aggravated Felony “Relating to Obstruction of Justice” Under the INA
We turn first to Silva‘s argument that his Massachusetts accessory-after-the-fact conviction is not categorically “an offense relating to obstruction of justice” because, he alleges, it does not require a nexus to a pending or ongoing investigation or judicial proceeding. In Esquivel-Quintana, the Supreme Court addressed a similar issue of whether the petitioner‘s state conviction for unlawful sexual intercourse with a minor was categorically an offense for “sexual abuse of a minor” under
We initially, as we said, follow the same analytical path as the Supreme Court in Esquivel-Quintana and consider the same factors to determine the generic federal definition of “an offense relating to obstruction of justice.”
1. The Text of the Statute as a Whole Shows That the Generic Federal Definition of “An Offense Relating to Obstruction of Justice” Must Be Construed to Include Accessory After the Fact
We begin with the text of the INA. See id. at 1568-69; United States v. De la Cruz, 998 F.3d 508, 513 (1st Cir. 2021). “[W]e accord the statutory text its ordinary meaning by reference to the specific context in which that language is used, and the broader context of the statute
a. The Text of § 1101(a)(43)(S) Supports That Accessory After the Fact Is “An Offense Relating to Obstruction of Justice”
Section 1101(a)(43)(S), which provides that “an offense relating to obstruction of justice” is an aggravated felony, was added to the INA in 1996. See Antiterrorism and Effective Death Penalty Act of 1996, Pub. L. No. 104-132, § 440(e)(8), 110 Stat. 1214, 1278; Illegal Immigration Reform and Immigrant Responsibility Act of 1996, Pub. L. No. 104-208, § 321(a)(11), 110 Stat. 3009, 3009-628. The INA does not define “obstruction of justice.” So we turn to reliable contemporaneous dictionaries to determine that term‘s ordinary meaning. See Esquivel-Quintana, 137 S. Ct. at 1569-70; De la Cruz, 998 F.3d at 515.
Black‘s Law Dictionary defined “obstructing justice” as “[i]mpeding or obstructing those who seek justice in a court, or those who have duties or powers of administering justice therein,” “attempt[ing] to prevent, or . . . prevent[ing], the execution of lawful process,” or “obstructing the administration of justice in any way -- as by hindering witnesses from appearing, assaulting process servers, influencing jurors, obstructing court orders or criminal investigations.” Obstructing Justice, Black‘s Law Dictionary 1077 (6th ed. 1990). Another dictionary at that time defined “obstruction of justice” as “the crime or act of willfully interfering with the process of justice and law esp. by influencing, threatening, harming, or impeding a witness, potential witness, juror, or judicial officer or legal officer or by furnishing false information in or otherwise impeding an investigation or legal process.” Obstruction of Justice, Merriam-Webster‘s Dictionary of Law 337 (1996).
Indeed, Bryan Garner, a noted legal commentator often cited by the Supreme Court, see, e.g., Esquivel-Quintana, 137 S. Ct. at 1569; Van Buren v. United States, 141 S. Ct. 1648, 1657 (2021); Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1749 (2020); Lockhart v. United States, 577 U.S. 347, 351 (2016), defined “obstruction of justice” as “interference with the orderly administration of law” and stated that it was “a broad phrase that captures every willful act of corruption, intimidation, or force that tends somehow to impair the machinery of the civil or criminal law,” Obstruction of Justice, B. Garner, A Dictionary of Modern Legal Usage 611 (2d ed. 1995). These definitions did not explicitly require that the obstructive conduct be committed in relation to a pending or ongoing investigation or judicial proceeding. Acting as an accessory to a crime after the fact by providing some sort of aid to the principal with the intent that the principal evade capture, trial, or punishment, even if there is not then a pending or ongoing investigation or proceeding, certainly fits within these definitions of “obstruction of justice.”
Furthermore, the relevant statutory term to be interpreted is not just “obstruction of justice“; rather, it is “an offense
b. The Text and Structure of Surrounding Provisions of the INA Support That Accessory After the Fact Is “An Offense Relating to Obstruction of Justice”
In addition to the text of the specific provision at issue, we also consider the text and structure of the INA as a whole. See Esquivel-Quintana, 137 S. Ct. at 1571 (considering the INA‘s disparate use of cross-references to other statutes in interpreting a particular provision). Most of the offenses listed as “aggravated felon[ies]” under
Congress easily could have included cross-references in
2. The Statute‘s Relationship to Other Federal Statutes Confirms That Accessory After the Fact Is “An Offense Relating to Obstruction of Justice”
In addition to the text of
a. The Federal Accessory-After-The-Fact Statute and the Federal Bribery Statute Support That the Generic Federal Definition Does Not Require a Nexus to a Pending or Ongoing Investigation or Judicial Proceeding
The federal accessory-after-the-fact statute,
defendant‘s conviction for accessory after the fact where he provided aid to his co-defendants immediately after they robbed a bank and helped them to escape apprehension).
This understanding of the federal accessory-after-the-fact statute as relating to obstruction of justice is bolstered by case law from the federal circuit courts of appeal before 1996. See United States v. Brown, 33 F.3d 1002, 1004 (8th Cir. 1994) (stating that “[t]he gist of being an accessory after the fact lies essentially in obstructing justice by rendering assistance to hinder or prevent the arrest of the offender after he [or she] has committed the crime” (alteration in original) (emphasis added) (quoting United States v. Barlow, 470 F.2d 1245, 1252-53 (D.C. Cir. 1972))); United States v. Huppert, 917 F.2d 507, 510 (11th Cir. 1990) (same), superseded by statute in other part, U.S. Sent‘g Guidelines Manual § 2J1.2(c)(1), cmt. (U.S. Sent‘g Comm‘n 1991); United States v. Balano, 618 F.2d 624, 631 (10th Cir. 1979) (same), abrogated in other part by Richardson v. United States, 468 U.S. 317 (1984); United States v. Willis, 559 F.2d 443, 444 (5th Cir. 1977) (same); Gov‘t of V.I. v. Aquino, 378 F.2d 540, 553 & n.19 (3d Cir. 1967) (same); see also Pugin v. Garland, 19 F. 4th 437, 447 n.8 (4th Cir. 2021) (collecting cases). This consistent interpretation of the federal accessory-after-the-fact statute when
Another federal statute we consider is the statute related to bribery of a witness.11 See
b. Several Obstruction of Justice Offenses Under 18 U.S.C. Chapter 73 Do Not Require a Nexus to a Pending or Ongoing Investigation or Judicial Proceeding
Silva argues that we should look exclusively to Chapter 73 of Title 18 of the U.S. Code, which is titled “Obstruction of Justice,” to determine the generic federal definition of “obstruction of justice.” He asserts that Chapter 73 cannot be reviewed for its components but must be read as a whole to require a nexus to a pending or ongoing investigation or judicial proceeding. We disagree. Chapter 73 contains many offenses and cannot be read as a whole.
Section 1512, which is titled “[t]ampering with a witness, victim, or an informant,” specifically provided in 1996 that “[f]or the purposes of this section . . . an official proceeding need not be pending or about to be instituted at the time of the offense.”12
that an investigation or judicial proceeding be pending or ongoing. Id. § 1511 (1996).
Indeed, as the Fourth Circuit recently held, obstruction of justice for the purposes of
c. Congress Intended Obstruction of Justice to be Read More Broadly Than Other Aggravated Felonies
Congress specifically included cross-references to other federal statutes in defining other offenses that constitute aggravated felonies under
3. Other Sources of the Statute‘s Meaning Support That Accessory After the Fact Is “An Offense Relating to Obstruction of Justice”
In addition to the text and structure of the statute and closely related federal statutes, we also consider other indicia identified in Esquivel-Quintana in determining the generic federal definition of “an offense relating to obstruction of justice.”
a. A Consensus of State Obstruction of Justice Statutes Confirm That “An Offense Relating to Obstruction of Justice” Does Not Require a Nexus to a Pending or Ongoing Investigation or Judicial Proceeding
One of those other sources we consider is a consensus of “state criminal codes” which “we look to . . . for additional evidence about the generic meaning of” an offense. Esquivel-Quintana, 137 S. Ct. at 1571; see also id. at 1571 n.3 (noting that a multijurisdictional analysis of state criminal codes can aid interpretation by providing “useful context“). When
And the obstruction of justice statutes in four of these latter states proscribed conduct which would have been punishable under the federal accessory-after-the-fact statute. See
b. The Model Penal Code and Federal Sentencing Guidelines Further Support That Accessory After the Fact Is “An Offense Relating to Obstruction of Justice”
The Model Penal Code and Federal Sentencing Guidelines also support that “an offense relating to obstruction of justice” includes accessory after the fact. See id. at 1571 (citing the Model Penal Code as evidence of the generic meaning of a term in the INA). When
The Federal Sentencing Guideline for “Obstruction of Justice” in effect in 1996 cross-referenced the guideline for “Accessory After the Fact.” See U.S. Sent‘g Guidelines Manual § 2J1.2(c) (U.S. Sent‘g Comm‘n 1995). The commentary to the “Obstruction of Justice” guideline stated that “[b]ecause the conduct covered by this guideline is frequently part of an effort to avoid punishment for an offense that the defendant has committed or to assist another person to escape punishment for an offense, a cross reference to [the Accessory After the Fact guideline] is provided.” Id. § 2J1.2 cmt. background (emphasis added). The commentary and cross-reference are further support that there was an understood connection between accessory after the fact and obstruction of justice when
4. Even Assuming Arguendo That There Must Be a Nexus to an Investigation for the Normal Accessory-After-the-Fact Crime, That Nexus Is Satisfied
Assuming arguendo that
[w]hoever, after the commission of a felony, harbors, conceals, maintains or assists the principal felon or accessory before the fact, or gives such offender any other aid, knowing that he has committed a felony or has been accessory thereto before the fact, with intent that he shall avoid or escape detention, arrest, trial or punishment, shall be an accessory after the fact.
To be convicted under that statute, the accessory must act with specific intent to enable a felon to “avoid or escape detention, arrest, trial, or punishment.” Absent an investigation, there can be no prosecution and no detention, arrest, trial, or punishment to avoid or escape.
In determining the least culpable conduct that violates the statue, Moncrieffe v. Holder directs us to “focus on the minimum conduct criminalized by the state statue,” rather than “apply ‘legal imagination’ to the state offense.” 569 U.S. 184, 191 (2013). Unlike the dissent, we look not to hypothetical conduct but to the “the least serious conduct for which there is a ‘realistic probability’ of charge and conviction.” United States v. Starks, 861 F.3d 306, 315 (1st Cir. 2017) (quoting id.). Looking at the mine run of prosecutions, we see none for aiding and abetting less serious felonies. When we look at the seventy Massachusetts state court opinions on Westlaw that cite
These offenses are also markedly different from the tax fraud scheme in United States v. Marinello, on which our dissenting colleague puts so much weight, in another key way. Persons are often able to skirt tax laws and go undetected for some time. Indeed, by definition, a person who commits tax evasion attempts to conceal his acts and evade detection. United States v. Stierhoff, 549 F.3d 19, 25 (1st Cir. 2008) (elements of tax evasion include “an affirmative act of evasion or attempted evasion“); see, e.g., United States v. Marek, 548 F.3d 147, 150–51 (1st Cir. 2008) (defendant falsified invoices); United States v. Lavoie, 433 F.3d 95, 98-100 (1st Cir. 2005) (defendant intentionally underreported revenue). An investigation of a tax offense would not be in the offing upon the commission of the offense. By contrast, consider murder, the underlying crime here. Following a homicide, more often than not the murder is reported, or a body is found, or a person is reported missing. A law enforcement investigation is reasonably foreseeable. Also consider robbery or extortion, crimes which underlie other actual Massachusetts accessory-after-the-fact prosecutions. These crimes involve immediate harm to a victim, unlike with tax crimes, where the government must uncover an offense against the fisc. In such situations, it is more likely that a law enforcement investigation is in the offing.
For the reasons stated above, it is clear that Silva‘s Massachusetts accessory-after-the-fact conviction is an aggravated felony under
5. The Dissent Misreads the BIA‘s Analysis
There is no basis to conclude that we have adopted a much looser nexus requirement than the BIA. Our dissenting colleague overreads the BIA‘s analysis in Matter of Valenzuela Gallardo to create a far tighter nexus requirement than the agency did.
The BIA held that generic obstruction of justice “cover[s] crimes involving (1) an affirmative and intentional attempt (2) that is motivated by a specific intent (3) to interfere with an investigation or proceeding that is ongoing, pending, or ‘reasonably foreseeable by the defendant.‘” Matter of Valenzuela Gallardo, 27 I. & N. Dec. at 456
To confirm that the dissent misconstrues the BIA‘s decision, we need look no further than the BIA implementation of Matter of Valenzuela Gallardo. In Matter of Cordero-Garcia, the BIA held that the California offense of dissuading a witness is categorically an offense related to obstruction of justice. 27 I. & N. 652, 653-54 (2019) (citing
B. Alternatively, If § 1101(a)(43)(S) Were Ambiguous, the BIA Would Prevail Under Chevron deference
Assuming, arguendo, that the statute is ambiguous and thus subject to Chevron deference,22
Chevron deference applies where an agency has followed suitably formal procedures, see United States v. Mead Corp., 533 U.S. 218, 229-30 (2001), to interpret a civil statute it administers, see Gonzales v. Oregon, 546 U.S. 243, 265 (2006). In applying Chevron, we follow a familiar two-step pathway. First, we determine whether the statute is ambiguous. Second, if it is, we defer to the responsible agency‘s reasonable interpretation of the ambiguous text. See Chevron, 467 U.S. at 842-43.
1. At “Step Zero,” the BIA‘s Interpretation Is Eligible for Chevron Deference
The BIA‘s administration of the INA falls within Mead‘s safe harbor for formal adjudications. See Mead, 533 U.S. at 226-27; see also Negusie v. Holder, 555 U.S. 511, 517 (2009). Its interpretation of
Additionally, Silva‘s contention that we cannot defer to the BIA‘s interpretation of a statute with criminal implications is misguided. We have previously deferred to the BIA‘s interpretation of the definition of aggravated felonies,23 see, e.g., Soto-Hernandez v. Holder, 729 F.3d 1, 5 (1st Cir. 2013); Lecky v. Holder, 723 F.3d 1, 5 (1st Cir. 2013), and continue to do so here.24
To hold otherwise would be flatly inconsistent with precedents both of this Court and the Supreme Court. In Chevron itself, the Supreme Court deferred to the EPA‘s interpretation of the Clean Air Act, even though knowing violation of an implementation plan under that act carried criminal penalties. See Guedes v. Bureau of Alcohol, Tobacco, Firearms & Explosives, 920 F.3d 1, 24 (D.C. Cir. 2019) (citing
2. Assuming § 1101(a)(43)(S) Is Ambiguous, the BIA‘s Interpretation of it Is Reasonable
We explained earlier why interpreting
In the realm of immigration, “Congressional powers are at their apex and judicial powers are at their nadir.” Hernandez-Lara v. Lyons, 10 F.4th 19, 54 (Lynch, J., dissenting). Congress has used that broad power to “charge[] the Attorney General with administering the INA.” Neguise, 555 U.S. at 516-17. Thus, “[j]udicial deference in the immigration context is of special importance, for executive officials ‘exercise especially sensitive political functions that implicate questions of foreign relations.‘” Id. at 517 (quoting INS v. Abudu, 485 U.S. 94, 110 (1988)). Compared to the executive, “[t]he judiciary is not well positioned to shoulder primary responsibility for assessing the . . . diplomatic repercussions” of sensitive immigration decisions. INS v. Aguirre-Aguirre, 526 U.S. 415, 425 (1999); see Griffiths v. INS, 243 F.3d 45, 49 (1st Cir. 2001). The BIA‘s exercise of its Congressionally assigned functions certainly was not unreasonable.
C. The IJ and BIA Did Not Err in Determining That Silva‘s Massachusetts Accessory-After-the-Fact Conviction Was a “Particularly Serious Crime” for Purposes of Denying Withholding of Removal
Silva also argues that even if his Massachusetts conviction is an aggravated felony under
Under Matter of N-A-M-, 24 I. & N. Dec. 336 (B.I.A. 2007), overruled in part on other grounds by Blandino-Medina v. Holder, 712 F.3d 1338, 1347-48 (9th Cir. 2013), which Silva does not challenge, the BIA “examine[s] the nature of the conviction, the type of sentence imposed, and the circumstances and underlying facts of the conviction” in determining whether it is a particularly serious crime. Id. at 342. “[A]n offense is more likely to be considered particularly serious if it is against a person . . . .” Matter of R-A-M-, 25 I. & N. Dec. 657, 662 (B.I.A. 2012) (citing Matter of N-A-M-, 24 I. & N. Dec. at 343). And factual findings made by the BIA in determining whether a conviction is for “a particularly serious crime” are “conclusive unless any reasonable adjudicator would be compelled to conclude to the contrary,” which “is a ‘highly deferential’ standard.” Garland v. Ming Dai, 141 S. Ct. 1669, 1674, 1677 (2021) (first quoting
The record supports the BIA‘s determination that, based on his guilty plea to the Massachusetts indictment, Silva admitted to knowing that the individuals whom he assisted had committed a murder. Based on that finding and Silva‘s conduct in aiding known murderers to avoid or escape apprehension and punishment, the BIA did not commit any legal or other error in determining that the accessory-after-the-fact conviction was a particularly serious crime which rendered him ineligible for withholding of removal.
III.
Silva‘s petition for review of the BIA‘s final order of removal is denied.
-DISSENTING OPINION FOLLOWS-
BARRON, Circuit Judge, dissenting. Is the crime of being an accessory after the fact to a “felony” under Massachusetts law, see
I.
It helps first to explain exactly how the majority‘s construction of
Each of those cases rejected the construction of the phrase “an offense relating to obstruction of justice” in
The Third and Ninth Circuits held that the “obstruction of justice“-related offense referred to in that federal statute requires proof of a much tighter nexus than the BIA in Valenzuela Gallardo III had held that the offense requires between an official investigation or proceeding and the obstructive conduct. Specifically, those courts held that the required nexus needs to be between the obstructive conduct and an official investigation or proceeding that, at the time of that conduct, was at least pending and not, for example, merely reasonably foreseeable. Valenzuela Gallardo IV, 968 F.3d at 1068; Flores, 856 F.3d at 292-294.
Moreover, on that basis, the Third and Ninth Circuits rejected the BIA‘s determination in each of the cases before them that an accessory after the fact offense that did not require proof of such a tight nexus could be “an offense relating to obstruction of justice” within the meaning of
In Silva‘s case, the BIA embraced the same looser description of the nexus requirement that the BIA had embraced in Valenzuela Gallardo III and that the Third and Ninth Circuits rejected for not being tight enough. The BIA then ruled based on that understanding of the nexus requirement that Silva‘s state-law accessorial offense is an “offense relating to obstruction of justice” under
Silva now contends that we must reject the BIA‘s ruling in his case for essentially the reasons that Flores and Valenzuela Gallardo IV gave in ruling as they did in those cases. But, Silva does not ask us to overturn the BIA‘s ruling only by asking us to endorse the tighter nexus that the Third and Ninth Circuits described. He also contends, as a fallback argument, that even if we were to accept that the BIA has correctly described the nexus requirement to be looser than the Third and Ninth Circuits held it to be, the Massachusetts accessory-after-the-fact offense of which he was convicted would not be an “offense relating to obstruction of justice” under
As I will explain, Silva‘s fallback argument for granting his petition for review is a winning one. For that reason, the majority‘s holding that the BIA‘s construction of the nexus requirement for “an offense relating to obstruction of justice” is a reasonable one under Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984), provides no basis for me to conclude otherwise. In that holding, the majority just endorses -- after drawing on the approach for interpreting
But, I emphasize, I also can see no reason to reject Silva‘s fallback argument insofar as the majority independently means to hold -- again pursuant to Esquivel-Quintana but this time without purporting to afford any deference under Chevron -- that the generic, federal “obstruction of justice“-related offense to which
II.
Before turning to the majority‘s independent holding endorsing a nexus-less construction of the offense referred to in
A.
In Valenzuela Gallardo III, the BIA was attempting to address a Ninth Circuit ruling that had rejected a prior BIA construction of
To address that concern, the BIA explained in Valenzuela Gallardo III that, in accord with Esquivel-Quintana, it would draw its revised understanding of the offense referred to in
In canvassing those materials, the BIA in Valenzuela Gallardo III acknowledged that most of the offenses in Chapter 73 contained a quite tight nexus requirement; the provisions in that chapter that set forth those offenses made clear that they could be committed only if the defendant obstructed an investigation or proceeding that was in fact pending. Id. at 454. But, the BIA also noted that the “Obstruction of Justice” offense set forth in
At the time that
The BIA explained that, in setting forth the synthesis of the case law that concerned
It is thus clear that Valenzuela Gallardo III was relying on Marinello‘s understanding of the “reasonably foreseeable” standard for defining the required nexus to address the vagueness concern that had led the Ninth Circuit to reject the BIA‘s prior nexus-less construction of
Specifically, the Court in Marinello made a point of explaining that to demonstrate that a “proceeding” or “investigation” was “reasonably foreseeable” at the time of the obstructive conduct, “[i]t is not enough for the Government to claim that the defendant knew [law enforcement] may catch on to his unlawful scheme eventually. To use a maritime analogy, the proceeding must at least be in the offing.” Marinello, 138 S. Ct. at 1110. Thus, Marinello clearly rejected the notion that evidence that a defendant had engaged in a tax scheme that was unlawful could suffice in and of itself to prove that the “reasonably foreseeable“-based nexus requirement had been met under the federal tax crime there at issue for “obstruct[ing] or imped[ing] . . . the due administration” of Title 26. See id. (discussing
Indeed, were that not so, the Court could have just said that proof of the defendant‘s predicate unlawful tax scheme would suffice to satisfy that nexus requirement for the offense defined in
There is, of course, a separate question regarding the element or elements of the “obstruction of justice“-related offense referred to in
The BIA, after having canvassed Chapter 73, Marinello, and other aspects of federal and state law, as well as other materials, expressly laid out the elements of the “obstruction of justice“-related offense in
As a result, I understand the BIA in Valenzuela Gallardo III to describe the offense referred to in
That said, I am aware that the BIA in Valenzuela Gallardo III did assert that Congress intended
Nonetheless, as I have explained, the BIA, after completing its analysis of
B.
There is one hanging thread. The majority asserts that it is wrong to read the BIA as “requir[ing] an identical nexus in both the statute at issue [in Marinello],
In support of this assertion, the majority first points to a statement by the BIA in Valenzuela Gallardo III in which it supposedly “expressly rejected that false equivalence.” Id. The statement from Valenzuela Gallardo III that the majority has in mind is the one in which the BIA observes that Congress intended
But, a closer examination of the BIA‘s decision reveals that the purported “express[] reject[ion]” that statement makes has nothing to do with the scope of the nexus requirement for “an offense relating to obstruction of justice” under
In performing that clarifying function, the footnoted statement in no way purports to be defining anew the scope of the nexus requirement for the generic, federal “obstruction of justice“-related offense or the meaning of the “reasonably foreseeable” standard that constrains the nexus requirement of that offense‘s actus reus. The BIA in the main body of its opinion relied heavily on Marinello, as I have described, and then expressly cited Marinello and United States v. Aguilar, 515 U.S. 593 (1995), which construed the nexus requirement in
The majority also points to a BIA case decided after Valenzuela Gallardo III that, the majority contends, demonstrates that the BIA did not intend to rely on Marinello to define the actus reus of “an offense relating to obstruction of justice.” Maj. Op. at 33-34 (discussing Matter of Cordero-Garcia, 27 I. & N. Dec. 652 (BIA 2019)). But, the BIA there did not purport to
In other words, I read Cordero-Garcia to be no different from Valenzuela Gallardo III or the BIA decision in Silva‘s case. It, too, embraces an understanding of the actus reus of “an offense relating to obstruction of justice” that is nexus-based and constrained by the reasonably foreseeable standard that Marinello sets forth. And, it, too, does so notwithstanding that it concludes that a state-law accessorial offense is such an “obstruction of justice“-related offense even though that state-law accessorial offense has no similarly constrained actus reus. I also emphasize that, in concluding that the BIA has not yet backed away from Valenzuela Gallardo III‘s quite clear description of the offense‘s actus reus, I am taking a position that the government itself appears to share, given the way that the government‘s briefing to us describes the understanding of actus reus of the offense that the BIA embraced in Valenzuela Gallardo III and re-embraced yet again in the case before us.
III.
With the BIA‘s operative understanding of the nexus requirement for “an offense relating to obstruction of justice” now in view, I am almost ready to address whether Silva‘s offense of conviction is such an offense. But, before taking up that question -- which, as I will explain, is the dispositive one in this case -- there remains one more task to complete. It concerns the majority‘s independent, Chevron-free holding that I mentioned at the outset.
In that holding, the majority appears to conclude that, insofar as we are not bound to defer under Chevron to the BIA‘s nexus-based construction of the offense referred to in
But, notably, the majority does not explain in so holding what basis there is for not applying Chevron here. It does not suggest, for example, that the BIA‘s nexus-based construction of the “obstruction of justice“-related offense referred to in
Nor does the majority suggest that Chevron deference is not available to the interpretive question at hand, even if the BIA‘s construction is a reasonable one. Rather, the majority rejects each of Silva‘s arguments to that effect as well as those of the amici and even explains why our precedents
There is, though, still one more problem with holding that Silva‘s petition must be rejected under Esquivel-Quintana based on a nexus-less construction of the “obstruction of justice“-related offense referred to in
The majority does, in applying those factors, advance reasons to reject the strict, pending-investigation-or-proceeding nexus requirement that the Third and Ninth Circuits have adopted.29
And, I suppose,
The majority also points to the fact that the Model Penal Code identifies an accessory after the fact offense as “an independent offense of obstruction of justice.” Maj. Op. at 26 (quoting Model Penal Code § 242.1 explanatory note (Am. L. Inst. 1985) [hereinafter MPC]). But, the majority does not hold that the only reasonable construction of
Finally, although the majority relies on some precedent to assert that the federal accessory-after-the-fact statute,
if we are not permitted to afford Chevron deference to the BIA‘s nexus-based construction despite its reasonableness, then we must adopt the nexus-less construction if it is superior to the BIA‘s. See H.J. Friendly, Chenery Revisited: Reflections on Reversal and Remand of Administrative Orders, 1969 Duke L.J. 199, 210.
But, in light of Chenery, to justify the rejection of Silva‘s petition based on a nexus-less construction of the “obstruction of justice“-related offense referred to in
Thus, because I see no basis under Chenery for attributing to the BIA the nexus-less understanding of
IV.
The majority draws in this portion of its opinion on the approach -- commonly referred to as the “categorical approach” -- that the Court relied on in Moncrieffe to construe
I have explained how the BIA defines the elements of the “offense relating to obstruction of justice” to which
That accessorial offense contains the following four elements: that the defendant (1) “harbors, conceals, maintains, or assists the principal felon or accessory before the fact, or gives such offender any other aid,” (2) does so “after the commission of a felony,” (3) does so “knowing that [the person being assisted] has committed a felony or has been accessory thereto before the fact,” and (4) does so with the specific intent that the defendant “avoid or escape arrest, detention, trial, or punishment.”
A.
The first element (though I suppose it could be understood to be a compendium of elements in its own right) defines part of the actus reus of the offense. It provides that the government must prove that the defendant engaged in conduct that could qualify as obstructive conduct of the kind that
But, nothing in the element requires there to be proof of a linkage of any kind between that conduct and an official investigation or proceeding into any crime. Thus, this element cannot ensure the required categorical match between the two offenses. Nor does the BIA -- or, for that matter, the majority -- suggest otherwise.
B.
The second element requires proof that the recipient of the defendant‘s aid committed a felony or was an accessory before the fact to that felony. This element also defines part of the actus reus of the offense. But, it requires no more of a showing than that the defendant lent the requisite type of aid to the recipient of it after that recipient had in fact committed a felony or been an accessory before the fact to it. Thus, it would appear that, just like the first element, this element also does not on its face require proof of what the BIA by its own account understands the generic, federal “obstruction of justice“-related offense to demand: proof that an official investigation is reasonably foreseeable at the time of the obstructive conduct and thus that there is a nexus of at least that sort between the defendant‘s obstructive conduct and an official investigation.
In addition, the accessorial offense cannot be understood to be impliedly subject to a “reasonably foreseeable“-based nexus requirement as an element, because it applies to being an accessory after the fact to Massachusetts felonies that are in their nature no more likely to be detected than the offense at issue in Marinello. See, e.g.,
The only potential complication that I can see with this conclusion is that the underlying “felony” in Silva‘s case is murder, which as a general matter is a crime more likely to trigger an investigation upon its occurrence than a tax offense. But, Silva asserts, without dispute from the government or the expression of any contrary view by the BIA, that the Massachusetts accessory-after-the-fact statute is not divisible. Nor did the BIA suggest here that Silva‘s offense of conviction qualifies as “an offense relating to obstruction of justice” because of the nature of the predicate felony that the principal committed.
C.
The third element does require proof that the defendant accused of the accessorial conduct lent aid to the one who committed the felony or was an accessory before the fact to it “knowing that [the person being assisted] ha[d] committed a felony or has been accessory thereto before the fact.”
D.
That leaves the fourth element, which requires proof that the defendant engaged in the alleged accessorial conduct “with intent that [the principal or accessory before the fact] shall avoid or escape detention, arrest, trial or punishment,”
Nor does the element require proof -- as a means of proving the defendant‘s mental state -- that an investigation was in fact in the offing (or, for that matter, was even reasonably foreseeable under any understanding of that constraint). The element requires proof merely that the defendant lent certain kinds of aid to the one who committed a felony or was an accessory before the fact to that felony with the intent to help that recipient evade arrest, trial, punishment, or detention. See Watson, 165 N.E.3d at 1025.
In fact, it appears that under this Massachusetts offense the defendant need not even have been shown to have reasonably but mistakenly believed that an official investigation was in the offing when he
Proof of no more than that the defendant held a belief (and not necessarily even an objectively reasonable one) about the investigation being reasonably foreseeable as of the time of his accessorial conduct is not proof of the investigation being reasonably foreseeable in fact as of that time. As such, proof of that sort is not proof of what the BIA -- and thus the majority -- agrees the generic, federal offense in
To be sure, in finding here that the Massachusetts accessorial offense at issue is a categorical match for the generic, federal offense referred to in
One need only consider the case in which the felon tells the accessory that he, the felon, is sure to be arrested as a means of inducing the accessory to lend assistance. If, in fact, there is no evidence to show that it was reasonably foreseeable at the time the assistance was provided by the accessory that authorities would investigate the felon‘s unlawful conduct -- and there was instead at most evidence at that time that authorities “may” do so “eventually” -- the BIA‘s own articulated standard for proving the nexus would not be met even though the specific intent requirement of the Massachusetts accessory-after-the-fact offense would be satisfied. Cf., e.g., United States v. De Leon, 270 F.3d 90, 92 (1st Cir. 2001) (recognizing that an offense “is a specific intent crime” because
E.
In sum, careful attention to the elements of the Massachusetts accessorial offense of which Silva was convicted and the elements of the generic, federal “obstruction of justice“-related offense referred to in
V.
The majority contends that I am mistaken in reaching this conclusion. It does so in part by emphasizing that in applying the categorical approach we may not imagine “fanciful, hypothetical scenarios.” United States v. Fish, 758 F.3d 1, 6 (1st Cir. 2014) (citing Gonzales v. Duenas-Alvarez, 549 U.S. 183, 193 (2007)); see also United States v. Starks, 861 F.3d 306, 315 (1st Cir. 2017). The majority points in this regard to the absence of any reported cases in Massachusetts of prosecutions for the accessory after the fact offense that is at issue in circumstances in which the defendant‘s obstructive conduct did not have a nexus to a reasonably foreseeable investigation or proceeding.
But, the categorical approach has never been understood to make the test of a state law‘s legally operative scope wholly dependent on the presence of reported cases of actual prosecutions for such conduct, in part because reported cases may not be representative of charging practices. Da Graca v. Garland, 23 F.4th 106, 113-14 (1st Cir. 2022); Swaby v. Yates, 847 F.3d 62, 65-66 (1st Cir. 2017) (noting that the Duenas-Alvarez limitation “has no relevance” to a case where the text of a state law clearly prohibited certain conduct but the petitioner “had failed to show that there was a realistic probability that [the state] would actually prosecute offenses” derived from that conduct). Nor would such a test make sense.
The outcome under the categorical approach depends on what the statute setting forth the offense at issue is fairly read to require the government to prove. It does not depend on how aggressively prosecutors choose to enforce that offense or the actual conduct of defendants in committing it. See Swaby, 847 F.3d at 65-66.32
Thus, what matters here is what is clear from the text of the state-law statute that sets forth the accessorial offense at issue. And, what is clear from that text is both that the offense may be committed under Massachusetts law with respect to tax evasion and that the state may prove the commission of that accessorial offense in relation to that type of felony without proving that any investigation into that felony was “in the offing.” Thus, it is clear that there is no requirement to prove a nexus of the “in the offing“-constrained sort that the BIA has determined must be proved as part of the actus reus of an “offense relating to obstruction of justice.”
The majority does attempt to respond to this line of argument. It does so by asserting that the state law accessorial offense at issue here is a categorical match for the “obstruction of justice“-related offense referred to in
But, no party to the case or the BIA has indicated that the accessorial offense at issue either is divisible or is indivisible but, contrary to its plain text, has an implicit “seriousness” limitation that makes it just fine to be an accessory after the fact to a tax crime. Nor is the accessorial offense in Silva‘s case one for which the BIA‘s “reasonably foreseeable“-based nexus had to be proved as part of its actus reus. I thus do not understand how the basis on which the majority purports to be explaining away Marinello‘s significance to this case -- which, I note, is not the bass on which the BIA itself relied -- comports with the categorical approach to which the majority purports to be faithful.
I would add only in this connection that the “traditional concept” of accessory after the fact offenses is that “the accessory‘s liability derives from that of his principal.” Model Penal Code § 242.3 (Am. L. Inst. 1985); see also 5 Tucker‘s Blackstone *35 (defining an accessory as “he who is not the chief actor in the offense . . . but is someway concerned therein“). For that reason, there was no need at common law to require proof for such a crime that the aid to the felon that the accessory provided in fact interfered with an investigation that was “in the offing.” It was enough to require proof that the accessory shared some culpability for the principal‘s crime, which could be proved through evidence of his culpable mental state at the time that he lent the aid.
By contrast, offenses of the kind that Blackstone labeled as ones against “public
As a result, it is not “fanciful” to conclude that Silva‘s offense of conviction requires no proof of the kind of nexus between obstructive conduct and an investigation or proceeding that the majority and the BIA acknowledge Congress did require for the generic, federal “obstruction of justice“-related offense referred to in
The majority separately takes issue with my invocation of Marinello, to the extent that I take the view that the BIA relied on Marinello‘s understanding of the “reasonably foreseeable” standard to define the required nexus for an “offense relating to obstruction of justice” under
But, as I have already explained, proof of the commission of a predicate crime -- or the awareness of its commission by the one guilty of the accessorial offense -- is not proof that an investigation into that predicate crime was in fact reasonably foreseeable at the time of the accessorial conduct in the “in the offing” sense. It is at most proof that authorities “may” (or of the defendant‘s knowledge that the authorities may) “eventually” catch on to that predicate crime‘s commission. But, that is exactly the sort of proof that Marinello clearly deemed inadequate to satisfy the “reasonably foreseeable” standard that the Court described in that case.
Thus, because the state law accessorial offense of which Silva was convicted requires, as to its actus reus, no more than proof of the defendant‘s mere awareness of the predicate crime‘s commission -- which, as best I can tell, is also a point that the majority does not dispute -- that state law accessorial offense cannot be a categorical match for the “obstruction of justice“-related offense set forth in
VI.
I have explained before why I do not agree with the criticism that the categorical approach permits technical inquiries to
I do not mean to suggest that anything would prevent a state from defining the elements of an offense that it labels as an accessorial one in a way that would make that offense a match for “an offense relating to obstruction of justice,” as the BIA understands that offense. But, as I have explained, there is no indication that Massachusetts has done so here.
Accordingly, I conclude that although Silva was convicted of being an accessory after the fact to a felony, neither the BIA nor the majority has established that he has been convicted of an “offense relating to obstruction of justice,” such that he may be deemed to have committed an “aggravated felony” within the meaning of the INA. I therefore would grant Silva‘s petition for review and so, respectfully, dissent.
