913 F.3d 807

RAWLINSON, Circuit Judge, dissenting:
I respectfully dissent from the majority's decision to vacate and remand for a new trial. Although I absolutely support the use of expert testimony on Battered Woman Syndrome in appropriate circumstances, I cannot say that the able district court judge in this case abused his discretion in determining that the testimony was not admissible in the context of establishing a duress defense, as opposed to the usual context of its admission-to establish self-defense.
We start from the premise that we owe deference to the district court's evidentiary rulings, particularly in the realm of expert testimony. See Saravia v. Sessions , 905 F.3d 1137, 1141 (9th Cir. 2018) (" 'Abuse-of-discretion review is highly deferential to the district court. ...' ") (citation omitted); see also Skidmore v. Led Zeppelin , 905 F.3d 1116, 1136 (9th Cir. 2018) ("District courts have broad discretion in making evidentiary rulings, including whether to allow expert testimony....") (citation and internal quotation marks omitted).
A district court abuses its discretion if it commits an error of law or "reaches a result that is illogical, implausible, or without support in the inferences that may be drawn from the record." Kode v. Carlson , 596 F.3d 608, 612 (9th Cir. 2010) (citation omitted); see also Mujica v. AirScan, Inc. , 771 F.3d 580, 589 (9th Cir. 2014) (same). If the district court commits no error of law "[t]he abuse of discretion standard requires us to uphold a district court determination that falls within a broad range of permissible conclusions." Kode , 596 F.3d at 612 (citation omitted). In my view, the district court's decision to exclude the evidence was based on such a "permissible conclusion" that the evidence was inadmissible. Id.
The proposed testimony from Dr. Cheryl Karp focused on the defendant's history of having been "sexually abused by her step-father and the police never protecting her." However, our precedent has not characterized evidence of this nature as relevant to a duress defense.
In United States v. Homick , 964 F.2d 899, 905 (9th Cir. 1992), we recognized that the battered woman defense "is a species of the defense of duress." In that case, the defendant sought to have expert testimony admitted to establish that her ex-husband, who was charged with the murder of the victims, coerced her into participating in the charged offenses through "his complete domination over her." Id. at 902, 905. We ultimately concluded that any error in excluding the proffered expert testimony was harmless because the two recorded telephone conversations between the defendant and her ex-husband reflected ready acquiescence, with "nothing implicitly or explicitly threatening about either conversation." Id. at 906.
Notably, in Homick we addressed battered woman syndrome in the context of a defense tethered to the co-perpetrator, who was alleged to be the duressor. See id. at 902, 905. Similarly, in United States v. Johnson , 956 F.2d 894 (9th Cir. 1992), superseded on other grounds in Martinez v. Martinez , 369 F.3d 1076, 1089 (9th Cir. 2004), we considered the duress defense in the context of women convicted of drug offenses, who asserted that the drug kingpin for whom they worked abused and "psychologically threatened" them. Id. at 901-02. We phrased the issue in the following manner: "The question, relevant to the defense of duress in the cases before us, is whether a special vulnerability to fear-a vulnerability not produced by those persons causing the defendant's criminal action-may be taken into account." Id . at 898. We resolved the issue by answering the question in the negative: "As a defense to a charge of criminal conduct, such subjective vulnerability has not been admitted ." Id. (emphasis added).
In other words, we held in Johnson that evidence of subjective vulnerability "not produced by" the named duressor is not admissible to establish the affirmative defense of duress. Id . Rather, such evidence may be used at sentencing. See id . ("[A] purely subjective element that cannot be taken into account in determining criminal liability may be taken into account in sentencing. ...") (citations omitted).
The real takeaway from our holding in Johnson is that the expert testimony addressing Battered Woman Syndrome must address vulnerability "produced by" the named duressor. Id. And that holding, in my view, solidifies why the district court in this case acted within its discretion by excluding the proffered expert testimony. As discussed, the record reflects only that the testimony would focus on the defendant's history of having been "sexually abused by her step-father and the police never protecting her." Because her step-father was not the named duressor for the crime for which she was on trial, our reasoning in Johnson rendered the district court's exclusion of the proffered evidence a "permissible conclusion" under our precedent. Kode , 596 F.3d at 612.
The majority seeks to avoid our holding in Johnson by expressing its "unwilling[ness] to read Johnson as establishing a categorical bar on [Battered Woman Syndrome] evidence in support of a duress defense at trial." Majority Opinion , p. 818. But Johnson does not purport to establish a categorical bar on Battered Woman Syndrome evidence. Rather, it limits admissibility of that evidence to circumstances where the individual producing the vulnerability to duress is the same as the named duressor for the offense being defended. See Johnson , 956 F.2d at 898. And as a three-judge panel, we are bound to follow the precedent of our court, no matter how unwilling we may be to hew to our past decisions, or how unpalatable the result may be. See Lair v. Bullock , 798 F.3d 736, 747 (9th Cir. 2015), as amended ("[W]e are bound by a prior three-judge panel's published opinions ...") (citing Miller v. Gammie , 335 F.3d 889, 892-93 (9th Cir. 2003) (en banc) ).
The majority also relies on Homick to bolster its disregard of our analysis in Johnson . See Majority Opinion , pp. 818-19. But as the majority must acknowledge, " Homick ... adds little clarity to the proper role of [Battered Woman Syndrome] evidence at trial ..." Majority Opinion , p. 819. Indeed, other than a cf . citation to Johnson , the panel in Homick did not grapple at all with the analysis in Johnson . Rather, the Homick panel resolved the issue by noting that the facts in Homick did not "fall within the scope of any reasonable approach to the battered woman defense, no matter how we modify the traditional duress standards." 964 F.2d at 905-906 (emphasis added). Nevertheless, it is notable that the Homick case fit within the parameters of the Johnson analysis: a battered woman defense tethered to the alleged duressor. See Homick , 964 F.2d at 902, 905.
The only other circuit to substantively address this issue is the Fifth Circuit. In United States v. Willis , 38 F.3d 170, 176 (5th Cir. 1994), the Court relied on our Johnson decision to hold that subjective evidence of vulnerability is irrelevant to the duress defense in determining criminal liability. Admittedly, the Fifth Circuit did not acknowledge our explanation that the evidence attesting to subjective vulnerability is only subjective when the "vulnerability is not produced by those persons causing the defendant's criminal action." Johnson , 956 F.2d at 898. Nevertheless, the analysis is consistent with Johnson because the Fifth Circuit similarly focused on the irrelevance of subjective evidence in meeting the objective standard required to establish a duress defense. See Willis , 38 F.3d at 175.
The expert testimony proffered by the defendant in this case hinged on the childhood abuse suffered by the defendant at the hands of her step-father. In an affidavit, Dr. Karp described how she would have testified:
My testimony would have provided the jurors with information about the cycle of violence, how battered women behave when under duress, and what the dynamics of the "hyper-vigilant" behaviors experienced by abused women, as part of the dynamics of PTSD, are in deciding the dangerousness of the situation.
My testimony would also have given the jury an opportunity to hear from an expert on Trauma and how that may have influenced [defendant's] feelings when she did not seek the help of the police, given her own trauma history of being sexually abused by her stepfather and the police never protecting her. [Defendant] felt they would not protect her or believe her, given her prior childhood history of trauma and never being "heard" or "protected" by law enforcement. The jury should have been given an explanation by an expert to understand how [Defendant's] childhood abuse influenced her decision-making.
It is enlightening to consider what the expert did say in her affidavit and what she did not say. She did say that the source of Defendant's PTSD was her step-father. She did not say that the source of her PTSD was the named duressor. In fact, the named duressor was not mentioned once in the expert's affidavit. She did pinpoint the time of the trauma as Defendant's childhood. She did not point to any trauma during adulthood. Clearly, the focus of the expert was the defendant's childhood, specifically the sexual abuse inflicted upon the defendant when she was a child. This is precisely the type of "subjective vulnerability" evidence that Johnson held was not relevant to a duress offense. See 956 F.2d at 898.
I am not persuaded by the cases cited by the majority, particularly in view of the binding precedent in this circuit that supports the decision of the district court.
In Dando v. Yukins , 461 F.3d 791 (6th Cir. 2006), the admissibility of similar evidence was decided in the context of a claim of ineffective assistance of counsel, see id . at 798. The Sixth Circuit concluded that defense counsel "failed ... to adequately investigate the availability of a duress defense and the related possibility that [the defendant] suffered from Battered Women's Syndrome." Id. However, this Sixth Circuit case actually fits within the holding of Johnson , inasmuch as the defendant in Dando asserted that her co-perpetrator and duressor were one and the same. See id . ; see also Johnson , 956 F.2d at 898 (explaining that the proffered expert testimony must address vulnerability "produced by" the duressor to be relevant). In any event, the Sixth Circuit expressly noted that the issue of allowing evidence of Battered Woman Syndrome to establish a duress defense had not been "addresse[d] either way" under Michigan law. See Dando , 461 F.3d at 801. This case absolutely does not support a conclusion that the district court in our case abused its discretion in disallowing the proffered testimony. Similarly, the non-binding district court case of United States v. Ramirez , No. 10-344 (PG), 2012 WL 733973 at *1 involved a co-perpetrator who was also the named duressor, as did the non-binding cases of United States v. Ceballos , 593 F.Supp.2d 1054, 1060-63 (S.D. Iowa 2009) ; and United States v. Marenghi , 893 F.Supp. 85, 97 (D. Maine 1995).
The non-binding state court cases are also singularly unpersuasive because they rely upon state statutes specifically addressing the admissibility of evidence establishing Battered Woman Syndrome. See Commonwealth v. Asenjo , 477 Mass. 599, 82 N.E.3d 966, 973 (2017) ("G.L.C. 233 § 23F, provides the defendant the statutory right to present [evidence of Battered Woman Syndrome]"); see also Wonnum v. State , 942 A.2d 569, 573 (Del. 2007) (referencing Del. Code §§ 303 and 304); State v. Williams , 132 Wash.2d 248, 937 P.2d 1052, 1058 (1997) (en banc) (involving a statutory duress defense and a co-perpetrator who was also the named duressor). Importantly, the Supreme Court of Arizona, the state where Lopez committed her crime, has recently ruled that similar expert testimony is not admissible under Arizona law. See State v. Richter , 245 Ariz. 1, 424 P.3d 402, 404 (2018).
In sum, none of the non-binding cases relied upon by the majority, singly or in combination, are sufficiently persuasive that the district court was compelled to admit the proffered expert testimony. This is especially true considering that the district court's decision fit squarely within our binding precedent, as the proffered testimony was not linked by the expert to the asserted duressor. See Johnson , 956 F.2d at 898 (explaining that the proffered expert testimony must address vulnerabilities that are "produced by" the duressor). And the district court's decision was consistent with precedent from the highest court in the state where the crime was committed. See Richter , 424 P.3d at 404.
Under the deferential abuse of discretion standard of review, it cannot be fairly said that the district court committed an error of law, because its decision did not run afoul of any binding precedent, and the non-binding precedent relied on by the majority was singularly unpersuasive. Indeed, the district court's decision was actually consistent with Ninth Circuit precedent and Arizona precedent. Largely for the same reason, the district court's decision was not "illogical, implausible, or without support in the inferences that may be drawn from the record." Kode , 596 F.3d at 612. Rather the district court's decision fell "within a broad range of permissible conclusions." Id. I would affirm the judgment of the district court.
The majority opinion cites to a case from the D.C. Circuit, United States v. Nwoye , 824 F.3d 1129 (D.C. Cir. 2016). However, that case addressed the issue in the context of a claim of ineffective assistance of counsel, and says nothing more than that Battered Woman Syndrome evidence "would have entitled [the defendant] to a jury instruction on duress." Id . at 1135. Nothing about that statement is inconsistent with our analysis in Johnson , especially considering that the batterer and the duressor were one and the same in Nwoye . See id . at 1131.

Passage view · GavelSight