SONG v. FITZ (2025)

MEMORANDUM OPINION

Young Song, whose first language is Korean, was convicted in state court of rape and sexual battery of the underage daughters of his on-again-off-again girlfriend. Before trial he requested and was denied the services of a court-appointed interpreter. After the Court denied Song's request, his trial counsel did not retain an interpreter.

Song proceeded to testify at trial in (broken) English. His testimony did not go well for him. He was convicted and sentenced to sixty-five years imprisonment. After exhausting his direct appeal and seeking state post-conviction relief, he now seeks habeas relief in this Court.¹

Song's § 2254 motion was referred to Magistrate Judge Holmes, who recommends that it be denied. (Doc. 128). Song filed objections to Judge Holmes’ Report and Recommendation (R&R). He seeks relief on the following grounds: (1) the state court's order denying his motion for an interpreter violated clearly established federal law, namely, the right to testify in his own defense; (2) the state court's factual finding that he spoke English well was an unreasonable determination of fact; and (3) his trial counsel provided ineffective assistance by failing to assess Song's English-language skills ahead of trial and by failing to hire an interpreter for the proceedings and, and as a result, Song demonstrated “cause” and “prejudice” sufficient to overcome his procedural default of that claim. (Doc. No. 130).

This Court has conducted a de novo review. Although the Court agrees with Song that his English was not up to the task of testifying without an interpreter, that observation, standing alone, is not legally sufficient to grant habeas relief. The standards set forth in § 2254(d)(1) and (2) and § 2254(e)(1) are designed to stop just “short of imposing a complete bar on federal court re-litigation of claims already rejected in state proceedings.” Harrington v. Richter, 562 U.S. 86, 102 (2011) (“If [§ 2254(d)’s] standard is difficult to meet, that is because it was meant to be.”). Under this restrictive framework, which greatly limits the remedial power of federal judges to second guess state judges, Song's objections must be overruled, the Magistrate Judge's R&R will be adopted, and Song's § 2254(d) claims must be denied.

I. BACKGROUND

The factual and procedural histories of this case are fully and accurately recited in the R&R. (Doc. No. 128 at 1-18). In short, Song was charged with rape and sexual battery based on the accusations of two minor children—the daughters of his former romantic partner. Before trial, Song's attorney filed a motion asking the state court to appoint a Korean-language interpreter to assist Song at trial. At the hearing on that motion, Song's attorney argued as follows:

Judge, my concern is, as I've put in the motion, my client is a Korean National. He speaks very good English, and I can sit down with him and understand him. But the problem is that numerous times, because of a heavy accent and he doesn't quite understand everything, not a hundred percent, I'll have to ask him to rephrase or vice-versa. And so it's difficult. But I can communicate with him, so I'm not asking for an interpreter for me to assist in my communication with him. My concern is in front of the jury. And my concerns are that, because Mr. Song does not understand all words and phrases, if there's any misinterpretation, you know, maybe his testimony might not be one hundred percent correct. My other concern is that if he does at times hesitate, trying to, it appears to me, translate in his own mind and then respond appropriately, concerned about the impression of the Jury that, well why is he hesitating, is he trying to be evasive, that type of thing.

(Doc. No. 27-1 at 54-55). Song's attorney did not cite a single case in support of his argument for a court-appointed translator. Neither did he offer his client to the trial court to assess Song's language skills. After listening to Song's attorney's argument, the trial judge concluded: “based on the arguments here today, I'm going deny the request [for an interpreter], absent some further showing though proof as to the need.” (Id. at 57). Song's attorney offered no “further showing of proof as to the need.”

Song proceeded to trial without an interpreter. His testimony was, at times, long-winded, irrelevant, non-sensical, and damaging. (See generally Doc. No. 26-7; see also Doc. No. 130 at 6-10). Song offers five examples of where his testimony went particularly wrong:

Example 1:

Q: Mr. Song –

A: Yes, sir.

Q: -- no matter what questions I ask you today –

A: Yes, sir.

Q: -- you are not going to ever tell us that you did these things to [J.L.] or to [S.L.], are you?

A: Whatever you say –

Q: Yeah.

A: -- I'm not talk to [J.L.], [S.L.].

Q: No. Lemme ask it a different way. I could ask you questions, from here till next year, you will never tell this Jury that you touched these girls in a sexual way, will you?

A: I can't follow you, what's the point is, sir.

Q: All right. I'm not going to get you to admit on the stand that you touched these girls, am I?

A: Can you ask me some easy (sic), please?

Example 2:

Q: What reason would [J.L.] have for you to make – for her to make up these terrible things about you?

A: I think many things. And then one of them is, maybe, [S.L.] ask her, you know, “Somebody is going some” – “that place? I don't wanna go by myself. Hey, come” – “hey, go with me.” You know, I had brother, sisters; and sometimes, when I was young, I could do that. So, that's why I think what [S.L.], [J.L.] think, you know – she – she could, maybe, I raped her. She could, maybe.

Example 3:

Q: Okay. What reason does [S.L.] have to make up all of these terrible things about you?

A: My thinking – this is not a fact, it's my thinking – but my thinking is always our situation. We live together and we had some – some thing, what do you want, something, what do you want (sic). I think she got really scared. When I go through all the allegations, just like Big Foot, she talk to some people, she already knows I'm coming. How is she – she could know I'm coming? I only talk to the mama, and then we talk in Korean language. You – as you know, she – she – she say she could understand the Korean; but, you know, it's our private talking. Now, you – she could understand Korean languages, she could understand pretty well; but some reason she already knew that I'm coming. But actually I'm not coming. I said no, but she – some reason, she knew that. So, that's why I think, maybe, their mama talk to her, “You don't listen to me.” I just – “he's coming.” And then, “I'm gonna tell you, and then you just behave yourself.” She did all the times. So, that's why she make some frighten (sic). Now, I imagine – I didn't talk to the – the Suk Pak, but I imagine that's what all the times and then what is it – it – she doing – used to be, sir.

Example 4:

Q: The question is, did you ever see [the girls] watch a movie, in which oral sex was displayed on the movie.

A: After they watching the “Police Academy,” they are talking about what the woman was doing behind the pulpit; and I didn't answer. And they just talking, they do this, they do this, they do kiss, or something like that; but I don't allow even they talking like that. And then that's their business.

Q: You know, Mr. Song, your story just keeps getting bigger and bigger –

A: Okay.

Q: for some reason.

A: Why is that? I need – I have a lot of thing to do, because I was at the jail coupla months. I didn't do nothing. You know, I'm the – one of the men who․

Example 5:

Q: Okay. Would you describe the relationship you began to have with Pak?

A: I met one of the Korean gentlemen. His name was Kagya (phonetic) Pak, who is the younger sister (sic) Chong Suk Pak. And he and me, we talking; and we went to the restaurant, and then she ask me, “Come to my sister's house,” and then please show her Korean attitude. That means I train kind of the prince peerman (phonetic) from South Korea ․

Q: All right. I just – I – I did not understand that one. You tried what?

A: I would train, like the prince peerman – I mean, the –

Q: Gentlemen?

A: No, sir. That's some high, but I'm talking about some low. They're not gentlemen. Some people who – who respect older people and show every time – even he is one years older than me, show respect. So, he likes me; so, he introduce me to his older sister, because the older sister had some problem with –

Q: Okay. Let's – we're running a little –

A: Yes, sir.

Q: – far afield of where I was going. And I forgot something. Let's back up.

(Doc. No. 130 at 5-8). There are others. (See generally Doc. No. 26-7). Moreover, on numerous occasions the court reporter could not discern what Song was saying. (E.g., id. at 18, 27, 68). Presumably the jury could not either.

The jury convicted Song of seven counts of rape of a child and four counts of aggravated sexual battery. (Doc. No. 26-8 at 98). Song appealed his convictions. However, the Tennessee Court of Criminal Appeals affirmed on all grounds, including that the trial court did not err in denying Song's motion for an interpreter. See State v. Song, 2005 WL 2978972 (Tenn. Crim. App. Nov. 4, 2005). The appellate court highlighted “defense counsel's admission,” that Song “spoke ‘very good English’ and did not need ‘an interpreter to assist counsel.” Id. at *6. It also independently reviewed the trial transcript and held that Song “speaks English quite well, and ․ was able to effectively communicate to the jury.” Id. at *7. The appellate court agreed with the trial judge that:

[I]t's not unusual for defendants, whether they're Korean-speaking or American-born, English-speaking citizens, to want to elaborate on their answers, maybe get far afield in their answers, in what is pertinent to a particular trial.

And, oftentimes, even those American-born, English-speaking witnesses have to be brought back, either with-from their attorney, as was done in this case several times, or from the State, in requesting that [the defendant] answer the question. And he did that.

Id. The state appellate court concluded that nothing “preponderate[s] against” the trial judge's finding (in denying Song's motion for new trial) that “nothing transpired during the trial would lead [the trial court] to believe [Song] needed an interpreter.” Id.

After his direct appeal was exhausted, Song sought post-conviction relief in the state courts, but these efforts were also rebuffed. (See Doc. No. 26-20 at 93-105; Song v. State, 2008 WL 624926 (Tenn. Crim. App. Mar. 4, 2008); Song v. State, 2010 WL 4296673 (Tenn. Crim. App., Oct. 29, 2010); Song v. State, 2011 WL 2713738 (Tenn. Crim. App., July 13, 2011); Song v. Carlton, 2011 WL 900059 (Tenn. Crim. App. Mar. 16, 2011)).

After failing to obtain relief in the state courts, Song turned to this Court, filing a pro se habeas corpus petition on November 7, 2011. That petition raised sixteen grounds for relief, including trial court error and ineffective assistance of trial counsel. (Doc. No. 1). Respondent filed an answer that was accompanied by the state court record. (Doc. Nos. 26 and 27). The Court subsequently appointed counsel for Song and his habeas counsel filed a first amended petition. (Doc. No. 43). Because it was unclear whether his claims from his pro se petition were included in the amended petition, the Court directed Song to file a second amended petition, which he did on December 10, 2014. (Doc. No. 72). The second amended petition is the operative pleading.

After he filed his federal habeas petitions, Song again sought habeas relief in state court. This federal action was stayed while the state habeas proceedings were pending. The state courts again denied Song relief, see Song v. State, 2017 WL 2192083 (Tenn. Crim. App. May 17, 2017), and the stay was lifted in this case.

In May 2023, the Magistrate Judge entered her R&R recommending the denial of the seven claims for habeas relief Song outlined in the second amended petition.² (Doc. No. 128 at 10-14). The matter is now before the Court on Song's objections to the R&R. (Doc. Nos. 129 and 130).

II. AEDPA STANDARDS

Song's § 2254 petition is governed by the Antiterrorism and Effective Death Penalty Act (AEDPA). The AEDPA erects two significant hurdles for state court prisoners to challenge their convictions in federal court: (1) the exhaustion requirement; and (2) the AEDPA's standard of review.

1. Exhaustion

Generally, a federal habeas court may only consider a state prisoner's constitutional claim if he first presented it to the state court in accordance with state procedures. See Shinn v. Ramirez, 596 U.S. 366, 371 (2022); O'Sullivan v. Boerckel, 526 U.S. 838, 848 (1999). This is known as the exhaustion requirement. 28 U.S.C. § 2254(b)(1)(A).

State prisoners often fail to raise their constitutional claims in state courts, and even when they do, they often fail to comply with state court procedures. If a state court would dismiss the constitutional claims because of procedural failures, such claims are technically “exhausted.” See Woodford v. Ngo, 548 U.S. 81, 92–93 (2006) (in the habeas context, “state-court remedies are ․ ‘exhausted’ when they are no longer available, regardless of the reason for their unavailability.”). However, state prisoners cannot simply bypass state procedures on the way to raising their constitutional claims in federal court. See Coleman v. Thompson, 501 U.S. 722, 732 (1991). That would “defeat the [ ] goal of the exhaustion rule.” Shinn, 596 U.S. at 371. To prevent that end-run, federal habeas courts must apply the doctrine of procedural default. Id. Under that doctrine, federal courts generally decline to hear procedurally defaulted claims unless the prisoner can demonstrate “cause” to excuse the procedural defect and “actual prejudice.” Coleman, 501 U.S. at 750.

To establish cause, a prisoner must “show that some objective factor external to the defense impeded counsel's efforts to comply with the State's procedural rule.” Murray v. Carrier, 477 U.S. 478, 488 (1986). The Supreme Court has held that ineffective assistance of state postconviction counsel may constitute “cause” to forgive procedural default of a trial-ineffective-assistance claim, but only if the State requires prisoners to raise such claims for the first time during state collateral proceedings. Martinez v. Ryan, 566 U.S. 1, 9 (2012). Cause is established where the state court's procedures effectively foreclose direct review of trial-ineffective-assistance claims. Trevino v. Thaler, 569 U.S. 413 (2013).

To establish prejudice, the prisoner must show more than “errors” that “created a possibility of prejudice” at trial. United States v. Frady, 456 U.S. 152, 170 (1982). Instead, the constitutional violation must “work[ ] to [the prisoner's] actual and substantial disadvantage.’ ” Id. For example, with a defaulted ineffective assistance of counsel claim, actual prejudice to excuse the petitioner's default requires more than the prejudice prong under Strickland v. Washington, 466 U.S. 668 (1984). See Ambrose v. Booker, 684 F.3d 638, 649 (6th Cir. 2012). The “most important aspect to the inquiry is the strength of the case against the defendant” and whether a trial without errors would still have resulted in conviction. Id. at 652; Jones v. Bell, 801 F.3d 556, 563 (6th Cir. 2015).

2. AEDPA's Standard of Review

In addition to the exhaustion require, the AEDPA severely limits the kind of relief a federal court can grant to a state prisoner:

An application for a writ of habeas corpus on behalf of a person in custody pursuant to the judgment of a State court shall not be granted with respect to any claim that was adjudicated on the merits in State court proceedings unless the adjudication of the claim—

(1) resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States; or

(2) resulted in a decision that was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.

28 U.S.C. § 2254(d). In other words, a state prisoner cannot obtain relief unless the state court: (1) violated a clearly established holding of the Supreme Court; (2) unreasonably applied a clearly established holding of the Supreme Court; or (3) unreasonably determined facts.

A decision of a state court is “contrary to” clearly established federal law if the state court arrives at a conclusion opposite that reached by the Supreme Court on a question of law or if the state court decides a case differently than the Supreme Court has on a set of materially indistinguishable facts. Williams v. Taylor, 529 U.S. 362, 405-06 (2000). An “unreasonable application” occurs when “a state-court decision unreasonably applies the law of [the Supreme Court] to the facts of a prisoner's case.” Id. at 409. “[A] federal habeas court may not “issue the writ simply because that court concludes in its independent judgment that the relevant state-court decision applied clearly established federal law erroneously or incorrectly.” Id. at 411.

Whether deciding a state court violated or misapplied federal law or unreasonably determined facts, “[a] federal court's collateral review of a state-court decision must be consistent with the respect due state courts in our federal system.” Miller–El v. Cockrell, 537 U.S. 322, 340 (2003). The “AEDPA recognizes a foundational principle of our federal system: State courts are adequate forums for the vindication of federal rights.”³ Burt v. Titlow, 571 U.S. 12, 19–20 (2013). “[T]he States possess sovereignty concurrent with that of the Federal Government, subject only to limitations imposed by the Supremacy Clause. Under this system of dual sovereignty, ․ state courts have inherent authority, and are thus presumptively competent, to adjudicate claims arising under the laws of the United States.” Tafflin v. Levitt, 493 U.S. 455, 458 (1990). The “state courts have the solemn responsibility equally with the federal courts to safeguard constitutional rights.” Id.

The “AEDPA thus imposes a ‘highly deferential standard for evaluating state-court rulings,’ and ‘demands that state-court decisions be given the benefit of the doubt.’ ” Renico v. Lett, 559 U.S. 766, 773 (2010) (quoting Lindh v. Murphy, 521 U.S. 320, 333, n. 7 (1997)). “A state court's determination that a claim lacks merit precludes federal habeas relief so long as ‘fair-minded jurists could disagree’ on the correctness of the state court's decision.” Harrington v. Richter, 562 U.S. 86, 101 (2011) (citing Yarborough v. Alvarado, 541 U.S. 652, 664 (2004)). The Supreme Court has emphasized “that even a strong case for relief does not mean the state court's contrary conclusion was unreasonable.” Id. (citing Lockyer v. Andrade, 538 U.S. 63, 75 (2003)); see also Cullen v. Pinholster, 563 U.S. 170, 180 (2011) (describing the AEDPA standard as “a difficult to meet and highly deferential standard for evaluating state-court rulings, which demands that state-court decisions be given the benefit of the doubt”).

“[R]eview under § 2254(d)(1) is limited to the record that was before the state court that adjudicated the claim on the merits.” Pinholster, 563 U.S. at 180. The state court's factual findings are presumed to be correct unless rebutted by the petitioner by clear and convincing evidence. 28 U.S.C. § 2254(e)(1); Schriro v. Landrigan, 550 U.S. 465, 473–74 (2007).

III. ANALYSIS

Courts review de novo any portions of a R&R to which a party objects. See Javaherpour v. United States, 315 F. App'x 505, 508 (6th Cir. 2009) (citing 28 U.S.C. § 636(b)(1)); United States v. Walters, 638 F.2d 947 (6th Cir. 1981). However, a party who fails to object to the R&R (or portions thereof) waives further district and appellate court review. See Miller v. Currie, 50 F.3d 373, 380 (6th Cir. 1995); United States v. Walters, 638 F.2d 947, 950 (6th Cir. 1981).

The Magistrate Judge recommends denying all of Song's claims. (Doc. No. 128). Song has only objected to the Magistrate Judge's proposed resolutions of Claim 1 and Claim 2 (subparts 10-11). (Doc. Nos. 129, 130). Accordingly, he has waived further review of Claim 2 (subparts 1-9 and 12) and Claims 3 through 7 and the Magistrate Judge's R&R will be adopted as to those claims without further discussion.

Song timely objected to the Magistrate Judge's recommendation that Claims 1 and 2 (subparts 10-11) be denied, arguing that the Magistrate Judge erred by finding that:

(1) (a) Rock v. Arkansas, 483 U.S. 44 (1987) and Perovich v. United States, 205 U.S. 86, 91 (1907) do not clearly establish the right of a non-native English speaker to testify with the aid of an interpreter; and

(b) that the state court's decision to deny Song's request for an interpret was not “contrary to clearly established federal law” or was not an “unreasonable application of clearly established federal law,” 28 U.S.C. § 2254(d)(1);

(2) The state court's factual finding that Song spoke English well was not based on “an unreasonable determination of the facts in light of the evidence presented in the State court proceeding,” 28 U.S.C. § 2254(d)(2);

(3) Song was not actually prejudiced by his attorney's failure to obtain an interpreter because the Tennessee Court of Criminal Appeals found that “[Song] spoke English quite well, that he was able to effectively communicate to the jury despite English not being his native language, and that nothing occurred during the course of the trial that indicated that an interpreter for Petitioner was necessary.”

(Doc. No. 130 at 13-23). The Court will engage in a de novo review of Claims 1 and 2 (subparts 10-11).

A. Did the trial court's denial of Song's motion for an interpreter violate his constitutional rights?

In Claim 1, Song argues that the state trial court's denial of his motion for an interpreter violated his constitutional rights “to confrontation, due process, equal protection, the assistance of counsel, a fair trial, and to testify and present a defense.” (Doc. No. 72 at 6-7) (citing § 2254(d)(1)). He argues that Rock v. Arkansas, 483 U.S. 44 (1987) and Perovich v. United States, 205 U.S. 86, 91 (1907) clearly establish the right of a non-native English speaker to testify with the aid of an interpreter. Although this Court believes such a right exists, neither Rock nor Perovich clearly establish that right.

“[C]learly established law as determined by th[e] [Supreme Court] ‘refers to the holdings, as opposed to the dicta, of [the Supreme] Court's decisions as of the time of the relevant state-court decision.’ ” Yarborough, 541 U.S. at 660–61. Clearly established law, within the meaning of § 2254(d), does not “include reasonable extensions of [Supreme Court] holdings in circuit precedent.” See Berkman v. Vanihel, 33 F.4th 937, 945 (7th Cir. 2022) (citing Glebe v. Frost, 574 U.S. 21, 24 (2014)). There is often a fine line between “extending existing law,” which is not permissible under § 2254(d)(1), and applying a well-established rule, which is. Yarborough, 541 U.S. at 666. But the line is not so fine here—Song urges an extension of Rock and Perovich, not an application of those decisions.

In Perovich, the Supreme Court noted in passing that whether to “appoint an interpreter when the defendant was testifying ․ is a matter largely resting in the discretion of the trial court.” 205 U.S. at 91. However, the Supreme Court did not hold that a defendant has a constitutional right to an interpreter. In fact, the Supreme Court did not resolve Mr. Perovich's assignment of error on the interpreter issue because he did not argue there was “any abuse of such discretion” in his briefs. Id. Because Mr. Perovich waived his right to appeal the trial court's decision to deny him an interpreter, there is simply no basis for this Court to find that Perovich held that a defendant has a constitutional right to an interpreter.

The same is true for Rock. In that case, the Supreme Court discussed the general right of a criminal defendant to testify on her own behalf and considered whether that right included using the defendant's “hypnotically refreshed testimony.” Rock, 483 U.S. at 45. Importantly, in Rock, the Supreme Court considered the validity of an Arkansas statute that excluded all posthypnosis testimony. Id. at 62. This rule had the practical effect of barring the criminal defendant in Rock from testifying because she underwent hypnosis before trial. Id. The Supreme Court held that Arkansas's “per se rule excluding all posthypnosis testimony infringe[d] impermissibly on the right of a defendant to testify on his own behalf.” However, the Supreme Court carefully noted that state courts retain the right to assess the reliability of evidence on a case-by-case basis, including posthypnosis testimony. Id. at 56.

Rock never mentions the rights of non-native English speakers or the constitutional necessity of interpreters. Instead, Rock addressed a narrow question—can a state categorically bar a defendant from offering posthypnosis testimony? Again, Rock's holding does not clearly establish a non-native English speaker's general or categorical right to an interpreter.

The Court's conclusion is consistent with the holdings of other courts that agree that neither Perovich nor Rock clearly establish a constitutional right to an interpreter. See, e.g., United States v. Si, 333 F.3d 1041, 1043 (9th Cir. 2003) (“The United States Supreme Court has not yet recognized a constitutional right to a court-appointed interpreter, but it has noted that the use of an interpreter is within the discretion of a trial court.”); Razo v. Bradshaw, 2006 WL 2056701, at *6 (N.D. Ohio July 21, 2006) (denying a § 2254 motion because the state court's decision that the defendant spoke English well enough to testify without an interpreter was not contrary to, or an unreasonable application of, clearly established federal law). The Court is unfamiliar with any cases that hold otherwise.⁴

Because the Supreme Court has not recognized a constitutional right to an interpreter, the state court's order denying Song's request for an interpreter cannot be contrary to or an unreasonable application of clearly established Supreme Court precedent. Thus, Song's § 2254(d)(1) claim fails.

B. Was the state court's finding that Song spoke English “well” an unreasonable determination of fact?

In Claim 1, Song also argues that the state court's factual finding that he spoke English “well” was an unreasonable determination of fact in violation of § 2254(d)(2). Although this Court agrees with Song that he did not speak English well enough to testify without the aid of an interpreter, that observation does not necessarily result in habeas relief.

The Supreme Court has instructed that courts “may not characterize ․ state-court factual determinations as unreasonable ‘merely because [they] would have reached a different conclusion in the first instance.” Brumfield v. Cain, 576 U.S. 305, 313-14 (2015) (quoting Wood v. Allen, 558 U.S. 290, 301 (2010)). Section 2254(d)(2) “requires that [federal courts] accord the state trial court substantial deference.” 576 U.S. at 314. Therefore, if “ ‘[r]easonable minds reviewing the record might disagree’ about the factual finding in question, ‘on habeas review that does not suffice to supersede the trial court's [factual] determination.” Wood, 558 U.S. at 301 (quoting Rice v. Collins, 546 U.S. 333, 341-42 (2006)). Stated differently, a state court's findings of fact are “presumed to be correct” unless they are rebutted by “clear and convincing evidence.” See § 2254(e)(1); see also Benge v. Johnson, 474 F.3d 236, 241 (6th Cir. 2007). “The standard is demanding but not insatiable,” and the Supreme Court has noted that “deference does not by definition preclude relief.” Miller–El v. Dretke, 545 U.S. 231, 240 (2005)

After reviewing the trial transcript and listening to the audio-recording of Song's testimony, this Court disagrees with the state courts’ factual assessment that Song spoke English “well.” Song's English-language skills appear sufficient to navigate most ordinary tasks like ordering food or communicating simple messages to customers or co-workers. However, his English-language skills appear insufficient to, for example: lecture at a university; preach a sermon; or testify for his freedom. If this Court had presided over Song's trial, it would have: (1) personally assessed his language skills before ruling on his motion for an interpreter; (2) recessed the trial and appointed an interpreter after hearing Song's direct examination; or (3) granted Song's motion for new trial and then appointed an interpreter. Song's trial testimony shows that he struggled to understand and answer questions in English, often giving indiscernible or confusing responses. In sum, on the state court record, this Court would not have concluded that Song's English was up to the task of testifying without an interpreter.

Four state court judges (the trial judge and three appellate judges) disagree, though. Because the quality of Song's English is not a black-and-white inquiry, but instead a question of degrees,⁵ fair-minded jurists could find that Song's English was good enough to testify without an interpreter.⁶ This federal court cannot substitute its own determination of fact for the reasonable determination of a state court. Hodge v. Plappert, 136 F.4th 648, 661 (6th Cir. 2025) (“We do not conduct a de novo review and substitute our “own judgment for that of the state court”).

Because reasonable people reviewing Song's testimony could disagree on the quality of his English language skills, he has not rebutted the state court's factual finding by clear and convincing evidence. Therefore, is not entitled to relief under § 2254(d)(2). The Court reluctantly reaches this conclusion because of the substantial deference due to state court decisions under the AEDPA. However, justice was not well-served by: (1) determining that Song spoke English well; and (2) depriving Song of an interpreter at trial.

C. Was Song's trial and post-conviction counsel constitutionally ineffective, and if so, can the procedural default be excused?

In Claim 2 (subparts 10-11), Song alleges his trial counsel of ineffective by:

[F]ailing to adequately assess Mr. Song's English language proficiency, to make an informed determination of the need for an interpreter, to support a motion for an interpreter with argument and/or evidence of Song's actual level of proficiency, or to otherwise secure or retain an interpreter to assist in counsel's communication with Song before and during trial, to translate the entire trial proceedings for Song, and to provide translation during Song's own trial testimony.

(Id. at 9). When a state prisoner asks a federal court to vacate a conviction due to ineffective assistance of counsel, the federal court must use a “doubly deferential” standard of review because “both the state court and the defense attorney [deserve] the benefit of the doubt.” Burt v. Titlow, 571 U.S. 12, 15 (2013). If that “doubly deferential” standard did not impose high enough of a burden, Song's ineffective assistance claim is also procedurally defaulted, which adds an additional layer of difficulty.

1. Strickland claim.

To establish ineffective assistance of counsel “a defendant must show both deficient performance by counsel and prejudice.” Premo v. Moore, 562 U.S. 115, 121 (2011). “Surmounting Strickland's high bar is never an easy task.” Padilla v. Kentucky, 559 U.S. 356, 371 (2010).

a. Deficient Performance

An attorney should be “strongly presumed to have rendered adequate assistance and made all significant decisions in the exercise of reasonable professional judgment.” Strickland, 466 U.S. at 690. To establish deficient performance, a person challenging a conviction must show that “counsel's representation fell below an objective standard of reasonableness.” Id. at 688. The challenger's burden is to show “that counsel made errors so serious that counsel was not functioning as the counsel guaranteed the defendant by the Sixth Amendment.” Id. at 687.

Here, it does not appear that Song's counsel developed any evidence to support his motion for an interpreter. For example, he did not present Song to the trial court, so that the trial judge could independently assess Song's language skills. Instead, counsel undermined his client's own motion by repeatedly saying that Song spoke English well. It is unclear how counsel's performance at the motions stage could meet an objective standard of reasonableness. Moreover, counsel compounded his errors at the motions stage by never subjecting Song to a moot direct examination or cross-examination with or without an interpreter. In other words, counsel failed to investigate how Song testified with an interpreter versus how he testified without an interpreter. Failures to investigate or inquire often point to deficient performance. See Smith v. Jenkins, 609 F. App'x 285, 292 (6th Cir. 2015). This is especially true where counsel offers no strategic reason for why he failed to assess Song's ability to testify in English before deciding to: (1) disregard the trial judge's invitation to renew his motion for a court-appointed interpreter and present new evidence, or (2) simply retain an interpreter for trial at Song's expense. Id.

Where, as here, Song had a reasonable argument that his trial counsel was ineffective based on counsel's handling of the interpreter issue, the Court assume, without deciding, that Song's post-conviction counsel was also deficient by failing to raise these arguments in state court.

b. Prejudice

With respect to Strickland prejudice, a challenger must demonstrate “a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different.” Strickland, 466 U.S. at 694; see also Jones v. Bell, 801 F.3d 556, 563 (6th Cir. 2015). “A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Id. In a classic “he said, she said,” case like this one, the jury's decision often turns on who it finds more believable—the victim or the alleged perpetrator. Cf. Vega v. Ryan, 757 F.3d 960, 974 (9th Cir. 2014). Any evidence that makes the alleged perpetrator look better or the purported victim look worse could affect the jury's decision, especially where there is no physical or other corroborating evidence. See Hawes v. Perry, 633 F. App'x 720, 725 (11th Cir. 2015). Where, as here, the conviction turned on the battling testimony of Song and his accusers, Song's inability to effectively communicate undermines confidence in the outcome. Accordingly, the Court will also assume Song has established Strickland prejudice based on the deficient performance of both his trial and post-conviction counsel.

2. Procedural Default

A finding of a substantial Strickland claim does not end the inquiry where, as here, that claim has been procedurally defaulted.⁷ That is only the first step. To ultimately prevail, a petitioner with a procedurally defaulted claim must show “cause” to excuse the procedural defect and “actual prejudice.” Coleman, 501 U.S. at 750. Song easily establishes the cause prong because Tennessee's procedural laws make it almost impossible for a defendant to present an ineffective assistance of trial counsel claim on direct appeal, relegating those claims to collateral review. Abdur'Rahman v. Carpenter, 805 F.3d 710, 713 (6th Cir. 2015). Where a state court's procedures effectively foreclose direct review of trial-ineffective-assistance claims, cause for the procedural default is established. Trevino, 569 U.S. 413 (2013). The question then is whether Song has established actual prejudice.

“Actual prejudice to excuse a habeas petitioner's procedural default requires more than the prejudice prong under Strickland.” Jones, 801 F.3d at 563 (citing Ambrose, 684 F.3d at 652). In cases where counsel was deficient and their error(s) resulted in Strickland prejudice, there is no presumption of actual prejudice to excuse a procedure default. Id. Instead, the question is whether “the outcome of the trial would have been different.” Id.; Lee v. Haas, 197 F. Supp. 3d 960, 974 (E.D. Mich. 2016) (“the Strickland prejudice prong required Petitioner to show a reasonable probability the result of the appeal would have been different, actual prejudice requires a showing the outcome of the trial would have been different.”). The “most important aspect to the [actual prejudice] inquiry is the strength of the case against the defendant” and whether a trial without errors would still have resulted in conviction. Ambrose, 684 F.3d at 652.

This is where Song's third claim falters. The prosecution offered the testimony of Song's accusers, the accusers’ mother, the accusers’ teacher and guidance counselor, a child psychologist, a nurse practitioner who examined one of the accusers, a social worker, and a detective with Nashville's sex crimes unit. (Doc. 128 at 6). Most of the witnesses simply repeated what they had been told by the accusers; however, the nurse practitioner testified that the hymen of one of the accusers had been injured in a way that was consistent with penetrative trauma. (Id.). Given the testimony of the accusers, which the jury clearly credited, and the testimony about physical injury to one of the accusers, Song cannot show that the outcome of the trial would have been different if he had been allowed to use an interpreter. Because the prosecution presented a credible case of rape and sexual battery, there is simply no telling whether Song's performance on the stand (assuming it had gone better with an interpreter) would have swayed the jury one way or the other. Therefore, he has not established actual prejudice, and his procedurally defaulted ineffective assistance claim is barred.

IV. CONCLUSION

Although Song's testimony certainly did not help his case, and this Court believes he would have benefited from a court-appointed or retained interpreter, Song has nonetheless failed to meet the high burden of overturning a state court conviction by means of a § 2254 motion. His motion will be denied.

An appropriate order will issue.

FOOTNOTES

1.   Song originally advanced a laundry list of reasons why he was entitled to habeas relief. However, with the advice of counsel, he has abandoned all but three claims for relief.

2.   Song's claim for ineffective assistance of counsel has twelve sub-claims outlining how his state attorneys were constitutionally ineffective. (Doc. No. 128 at 10-14).

3.   The AEDPA incorporates a longstanding belief that state courts are equally capable of protecting constitutional rights as federal courts. This Court must apply this view, because it is now enshrined in law. However, history belies this (seemingly well-accepted) premise. Because most state judges lack life tenure and are either directly elected or appointed by a political benefactor, state judges—who are generally more susceptible to political pressure—have been historically less willing to protect the constitutional rights of vulnerable minorities. England v. Louisiana State Bd. of Med. Examiners, 375 U.S. 411, 427 (1964) (Douglas, J., concurring) (“federal judges appointed for life are more likely to enforce the constitutional rights of unpopular minorities than elected state judges.”). The architect of our Constitution, James Madison, recognized the potential for this problem and advocated for a strong federal judiciary that could remedy constitutional violations committed by state tribunals:What was to be done after improper verdicts, in state tribunals, obtained under the biased directions of a dependent judge, or the local prejudices of an undirected jury? To remand the cause for a new trial would answer no purpose. To order a new trial at the supreme bar would oblige the parties to bring up their witnesses, though ever so distant from the seat of the court. An effective judiciary establishment, commensurate to the legislative authority, was essential. A government without a proper executive and judiciary would be the mere trunk of a body, without arms or legs to act or more.5 Elliot's Debates (Lipp. ed. 1941), p. 159. Simply put, current habeas law does not reflect our Founder's understanding of federalism or the now two and a half centuries of experience that show that some state courts have been less vigilant in protecting the rights of people of color, women, homosexuals, and immigrants. See, e.g., James L. Gibson and Michael J. Nelson, Judging Inequality: State Supreme Courts and the Equality Crisis (2021).

4.   Several circuits have held that, under certain circumstances, there is a constitutional right to an interpreter at trial. United States v. Mayans, 17 F.3d 1174, 1181 (9th Cir. 1994); United States ex rel. Negron v. New York, 434 F.2d 386, 389 (2d Cir. 1970); United States v. Carrion, 488 F.2d 12, 14 (1st Cir. 1973). This Court agrees that non-native English speakers have a right to an interpreter when language ability impairs their Fifth Amendment right to testify or their Sixth Amendment right to confrontation. This musing, though, offers Song little solace because he can only prevail here if the Supreme Court had clearly recognized a Fifth and Sixth Amendment right to an interpreter before Song's trial in 2004.

5.   Judging someone's ability to communicate is also difficult because the way someone speaks oftentimes has as much to do with their personality than their language skills. For example, some people are long-winded. Other people are direct. It is impossible to tell with any precision whether Song's testimony went horribly wrong because of his English skills or because of his personality.

6.   The Sixth Circuit has not addressed whether “unreasonable” has the same meaning for both §§ 2254(d)(1) and (d)(2). However, other judges have discussed this issue and concluded that the “fair-minded jurist” standard applies to the use of the word “unreasonable” in both §§ 2254(d)(1) and (d)(2). Smith v. Aldridge, 904 F.3d 874, 889 (10th Cir. 2018) (Tymkovich, J., concurring); Holsey v. Warden, Georgia Diagnostic Prison, 694 F.3d 1230, 1257 (11th Cir. 2012) (“A state court's ․ determination of the facts is unreasonable only if no ‘fair-minded jurist’ could agree with the state court's determination or conclusion.”); (“Ortiz v. Trimble, 2013 WL 943408, at *4 (E.D. Cal. Mar. 11, 2013)) (“It makes no sense to interpret ‘unreasonable’ in § 2254(d)(2) in a manner different from that same word as it appears in § 2254(d)(1)—i.e., the factual error must be so apparent that ‘fair-minded jurists’ examining the same record could not abide by the state court factual determination.”). Regardless of whether the Court applies the “fair-minded jurist” standard and/or the “clear and convincing evidence” standard, the Court finds that Song has not established a habeas violation under § 2254(d)(2).

7.   In his Objections, Song does not challenge the Magistrate Judge's conclusion that his Strickland claim was procedurally defaulted. (Doc. No. 130). In fact, his Objections do not meaningfully engage with the procedural default standard or the Magistrate Judge's conclusion that Song cannot demonstrate “actual prejudice” as opposed to Strickland prejudice.

WAVERLY D. CRENSHAW, JR., UNITED STATES DISTRICT JUDGE