CAAF decided the Army case of United States v. Criswell, 78 M.J. 136, No. 18-0091/AR (CAAFlog case page) (link to slip op.), on November 16, 2018. The court granted review to determine whether the military judge erred by allowing the alleged victim to identify Criswell as her assailant during her testimony, even though she did not know him before the alleged assault and she was shown a picture of him (and only him) before trial in a way that was found to be unnecessarily suggestive. CAAF narrowly affirms the military judge ruling and Criswell’s convictions, with the majority applying a highly-deferential standard of review that focuses on Criswell’s appellate-stage objections to the military judge’s ruling. The dissenters, however, conduct a broader review, find numerous flaws in the military judge’s ruling, and would reverse the findings and authorize a rehearing.

Judge Maggs writes for the court, joined by Chief Judge Stucky and Judge Ryan. Judge Ohlson dissents, joined by Judge Sparks.

CAAF granted review of a single issue:

Whether the military judge abused his discretion in denying a defense motion to suppress the accusing witness’s in-court identification of Appellant.

A general court-martial composed of a military judge alone convicted Specialist (E-4) Criswell, contrary to his pleas of not guilty, of one specification of making a false official statement, two specifications of abusive sexual contact, one specification of assault consummated by battery, and one specification of indecent language in violation of Articles 107, 120, 128, and 134. Criswell was sentenced to confinement for two years, reduction to E-1, and a dishonorable discharge.

Criswell’s convictions relate to an allegation by Specialist (SPC) AM that she was verbally harassed and sexually touched on two separate occasions during a large party at a convention center. SPC AM is a white female; her assailant was a black man. The lighting was poor during both encounters and SPC AM did not know her assailant, but she saw his face well enough to provide a general description to another soldier at the party and based on that description the other soldier suspected Criswell. Military criminal investigators showed SPC AM a single picture of Criswell the next day, and she identified him at that time as her assailant “immediately and without hesitation.” Slip op. at 7 (quoting military judge’s findings of fact). Testifying in court, SPC AM again identified Criswell as her assailant. The military judge did not allow the prosecution to introduce SPC AM’s out-of-court identification of Criswell, but he allowed the in-court identification over a defense objection that challenged it as the unreliable product of the actions of the military investigators.

Considering those facts, all of CAAF’s judge’s agree that eyewitness identifications can lead to injustice. Judge Maggs observes for the majority that “eyewitness identifications are problematic in any criminal justice system.” Slip op. at 2. Judge Ohlson observes for the dissenters that “mistaken eyewitness identifications are responsible for more wrongful convictions than all other causes combined.” Diss. op. at 11 (quoting United States v. Brownlee, 454 F.3d 131, 142 (3d Cir. 2006)). But CAAF splits over how to review the eyewitness identification allowed in this case. While the ordinary standard of review for rulings admitting evidence is abuse of discretion – where findings of fact are reviewed for clear error and conclusions of law are reviewed de novo – the majority narrows that approach, with Judge Maggs explaining:

Applying this standard in this case, we do not undertake a de novo analysis of whether the in-court identification should have been admitted. Instead, we focus on Appellant’s objections to the military judge’s findings of fact, view of the law, and conclusions in applying the law to the facts.

Slip op. at 7 (emphasis added). The majority then addresses the challenges raised in Criswell’s “brief and reply brief,” slip 8, and during oral argument, slip op. at 8-9 (citing oral argument recording). See also slip op. at 11 (excusing the military judge’s failure to specifically address a factor because “Appellant does not contend, either in his briefs or his oral argument[,] that the military judge misunderstood this aspect [of the law]”). The dissenters, however, make no reference to Criswell’s briefs or to the oral argument while concluding that “the military judge committed three errors in this case.” Diss. op. at 4.

Mil. R. Evid. 321 governs admission of an eyewitness identification. Such an identification is inadmissible if it was “the result of an unlawful lineup or other unlawful identification process,” Mil. R. Evid. 321(b)(1). An unlawful identification process includes one that is unreliable because it was “so suggestive as to create a substantial likelihood of misidentification.” Mil. R. Evid. 321(c)(1). An eyewitness identification is also inadmissible if its exclusion “is required by the Due Process Clause of the Fifth Amendment.” Mil. R. Evid. 321(b)(2).

The military rule generally codifies the Supreme Court’s decision in Neil v. Biggers, 409 U.S. 188 (1972), except that the military rule considers unreliability (as the result of suggestiveness) as a basis for exclusion separate from due process (Biggers, in contrast, addressed unreliability as a violation of due process). Nevertheless, Biggers and CAAF’s subsequent decisions “ask[] two questions. The first question [i]s whether the identification process at issue was unnecessarily suggestive. Assuming that it was, the second question [i]s whether the process was conducive to a likelihood of misidentification.” Slip op. at 2 (citing United States v. Baker, 70 M.J. 283, 288 (C.A.A.F. 2011)) (marks omitted). Furthermore, in Baker:

In answering the second question, [CAAF] considered five factors identified by the Supreme Court in Biggers and Manson v. Brathwaite, 432 U.S. 98 (1977). The five factors (commonly called the “Biggers factors”) are:

(1) the opportunity of the witness to view the perpetrator at the time of the crime;

(2) the witness’s degree of attention;

(3) the accuracy of the witness’s prior description of the perpetrator;

(4) the witness’s demonstrated level of certainty during the confrontation; and

(5) the elapsed time between the criminal act and the confrontation. Manson, 432 U.S. at 114 (citing Biggers, 409 U.S. at 199−200).

The Court in Baker then “weighed” against these factors “the corrupting effect of the suggestive identification itself.” Baker, 70 M.J. at 291 (internal quotation marks omitted) (quoting Manson, 432 U.S. at 114). The purpose of the weighing was to “determine whether under the totality of the circumstances the identification was reliable even though the confrontation procedure was suggestive.” Id. (citing Biggers, 490 U.S. at 199).

Slip op. at 3-4 (paragraphing added). Put differently, if there was an unnecessarily suggestive identification process (and both sides agree that there was in this case, when military investigators showed SPC AM a picture of only Criswell), the Biggers factors and Manson weighing are the legal tests that are applied to determine whether that suggestive process makes that (or a subsequent) identification unreliable.

Mil. R. Evid. 321 also makes it the prosecution’s burden to prove that an eyewitness identification is admissible in one of two ways. First, when the defense objects to any identification as unreliable, “the prosecution must prove by a preponderance of the evidence that the identification was reliable under the circumstances.” Mil. R. Evid. 321(d)(6)(B)(i). Second, if an initial identification is found to be unreliable then a later identification may only be admitted if “the prosecution proves by clear and convincing evidence that the later identification is not the result of the inadmissible identification.” Mil. R. Evid. 321(d)(6)(B)(ii).

In Criswell, the military judge suppressed SPC AM’s initial identification of Criswell (when she was shown the picture by military investigators) because it was unreliable, but he allowed SPC AM’s later, in-court identification. Slip op. at 7. Criswell challenged the ruling allowing the in-court identification, asserting that the in-court identification was tainted by the earlier, unreliable identification. CAAF’s review is highly fact-specific, and Judge Maggs’ majority opinion considers the military judge’s decision using a three-part analysis:

1. “Whether the military judge’s findings of fact were clearly erroneous.” Slip op. at 8.

2. “Whether the military judge’s view of the law was erroneous.” Slip op. at 9.

3. “Whether the military judge’s conclusions were outside the range of reasonable choices.” Slip op. at 12.

All of the Biggers factors and the Manson weighing are reviewed in the third part where the military judge gets an extra measure of deference from CAAF, with Judge Maggs’ explaining that:

Under the abuse of discretion standard—because the military judge did not make clearly erroneous findings of facts and did not exhibit a misunderstanding of the law—we are limited to considering whether the military judge’s conclusion with respect to any issue “is outside the range of choices reasonably arising from the applicable facts and the law.” Irizarry, 72 M.J. at 103 (citation omitted) (internal quotation marks omitted). Under this deferential standard, we see no error.

Slip op. at 12. Any question about whether the range of choices analysis really is a separate and distinct part of the test for abuse of discretion will have to wait for a different case, however, beacuse “the parties agree” on the standard, slip op. at 7, and the range of choices language was included in Criswell’s brief.

The majority finds that all of the military judge’s findings of fact are supported by the record and that he had a correct view of the law. It then reviews each Biggers factor, in light of Criswell’s specific challenges, for a decision outside a range of reasonable choices. For the first factor (SPC AM’s opportunity to view the perpetrator at the time of the crime), Criswell argued that the military judge failed to consider the overall lack of light during the party. The majority acknowledges that the lack of light is a factor, but it concludes that:

In applying the abuse of discretion standard, the question is not whether conflicting evidence—such as the evidence about visibility at the party—might have been weighed differently. The question is only whether the military judge made a decision outside of the range of reasonable choices. Irizarry, 72 M.J. at 103. Viewing the relevant evidence in the light most favorable to the Government, we conclude he did not.

Slip op. at 13. For the second factor (SPC AM’s degree of attention at the time of the crime), Criswell highlighted some uncertainties in SPC AM’s descriptions of Criswell and also argued that the military judge should not have considered SPC AM’s descriptions of her attacker given after she saw Criswell’s photo. The majority rejects those arguments, concluding:

We cannot say that the military judge failed to make a reasonable conclusion when we view the evidence in the light most favorable to the Government.

Slip op. at 14. For the third factor (the accuracy of SPC AM’s prior description of the perpetrator), Criswell emphasized discrepancies between SPC AM’s description of her attacker and what Criswell was wearing. Noting that the military judge was presented evidence of the discrepancies, the majority concludes:

In these circumstances, the military judge’s conclusion that Specialist AM’s description was “accurate” was within his discretion to make.

Slip op. at 14. For the fourth factor (SPC AM’s level of certainty when she identified Criswell after seeing the photograph), Criswell challenged the military judge’s conclusion that SPC AM’s identification of him as the perpetrator “was immediate and certain,” on the basis that the military judge heard testimony that she identified him quickly, but not with certainty. Slip op. at 15 (quoting record). The majority rejects that challenge outright, observing that:

[A]n inference that she was certain from the speed of her identification is not unreasonable.

Slip op. at 15. For the fifth factor (the passage of time), Judge Maggs’ majority opinion notes that Criswell “does not contest the military judge’s conclusion with respect to the fifth Biggers factor.” Slip op. at 15. The majority concludes that the military judge did not abuse his discretion in applying that factor.

The majority then considers other concerns raised by Criswell, including “likelihood of other individuals in the area at the time of the offense matching the description given by the victim,” and “the fact that this case involved a highly stressful, violence-threatening, and cross-racial crime occurring in minimal lighting with a partially disguised assailant.” Slip op. at 16 (marks and citations omitted). Those concerns are rejected with the observation that:

we see no reason to believe that the military judge overlooked or gave insufficient weight to the additional factors that Appellant mentions. . . . in announcing his final conclusion, the
military judge said:

The Court considered the environmental factors and the description given by the alleged victim, in this case, and given all the surrounding circumstances the alleged victim’s description is not so far removed from . . . what one might expect, again, given the evidence known to the Court, at this time, that that in and of itself the discrepancies raised by the defense on cross-examination in and of themselves so concern the Court as to disallow the in court identification.

Given the military judge’s express statement that he had considered “all the surrounding circumstances,” and no reason for believing that he did not, we cannot conclude that the military judge abused his discretion.

Slip op. at 16-17 (emphasis added). Judge Maggs’ majority opinion does not separately consider the Manson weighing of the totality of the circumstances.

Having found that the military judge’s choices – as challenged by Criswell – were within a reasonable range, the majority affirms the military judge’s ruling and Criswell’s convictions.

The dissenters, however, take a remarkably different approach.

Judge Ohlson’s dissenting opinion (that is joined by Judge Sparks) begins with a less-deferential standard of review than that employed by Judge Maggs’ majority opinion:

it is apparent that in reaching his decision to admit this evidence, the military judge abused his discretion because: “(1) he predicate[d] his ruling on findings of fact that [were] not supported by the evidence of record; (2) he use[d] incorrect legal principles; [and] [(3)] he fail[ed] to consider important facts.” United States v. Commisso, 76 M.J. 315, 321 (C.A.A.F. 2017) (citation omitted).

Diss. op. at 1 (modifications in original). Judge Ohlson also emphasizes that the burden is on the Government:

When, as here, the military judge determines that an out-of-court identification “is the result of an unreliable identification,” an in-court “identification may be admitted if the prosecution proves by clear and convincing evidence that the later identification is not the result of the [prior] inadmissible identification.” M.R.E. 321(d)(6)(B)(ii). Placing this burden on the government is appropriate because, as the Supreme Court has noted, if an impermissibly suggestive out-of-court identification must be excluded at trial, then “a courtroom identification based on such a lineup logically ha[s] to be excluded as well.” Watkins, 449 U.S. at 349 (emphasis added) (citing United States v. Wade, 388 U.S. 218 (1967)).

Diss. op. at 1-2. Judge Ohlson then reviews the case without any caveat that the dissenters view the facts in the light most favorable to the prosecution, and he finds that the military judge made three errors:

First, he made clearly erroneous findings with respect to three of the five Biggers factors. Second, he failed to articulate whether he considered two applicable factors as part of a required totality of the circumstances analysis. And third, he did not weigh the Biggers factors against the corrupting effect of the suggestive CID pretrial identification procedure.

Diss. op. at 4. Each of those errors involves specific disagreement with the majority’s analysis.

The three Biggers factors errors identified by the dissenters are directly contrary to conclusions reached by the majority. First, the dissenters fault the military judge for relying on SPC AM’s “in-court description of her assailant’s features [that] was tainted by the previous highly suggestive out-of-court identification process.” Diss. op. at 4 (emphasis in original). The majority, however, concludes that “there is no rule prohibiting the military judge from considering subsequent descriptions of the accused.” Slip op. at 11. Second, the dissenters reject the military judge’s conclusion that SPC AM gave a detailed prior description of Criswell, with a footnote observing that:

Both the military judge and this Court’s majority rely on the victim’s in-court description of Appellant as evidence of the accuracy of her prior out-of-court description. However, it is unclear how a witness’s description of an accused after the unduly suggestive identification process is compelling evidence of the strength of her prior description. Similarly, it is unclear how a victim’s contemporaneous description of an accused as she looks at him sitting in the courtroom is particularly helpful on this point.

Diss. op. at 6 n.7 (emphasis in original). Third, the dissenters reject the majority’s inference that a fast identification is a certain identification, with Judge Ohlson writing that immediacy and certainty “are separate concepts.” Diss. op. at 6.

Judge Ohlson’s dissenting opinion next addresses the military judge’s failure to “specifically determine whether ‘under the totality of the circumstances’ the in-court identification was reliable despite the prior suggestive out-of-court identification.” Diss. op. at 6 (quoting Biggers, 409 U.S. at 199; Baker, 70 M.J. at 291). In this portion of the dissent, Judge Ohlson considers the totality of the circumstances separate from the Biggers factors and the Manson weighing; that’s something the majority does not do. Observing that the military judge (and the majority) invoked “all the surrounding circumstances” (and not the totality of the circumstances), Judge Ohlson focuses on the context in which the military judge used that phrase:

the military judge’s intent when he used the phrase “all the surrounding circumstances” is—at best—ambiguous. This is particularly true in light of the fact that the phrase “the totality of the circumstances” is so commonly used in the law, and yet the military judge did not employ it. Furthermore, the military judge did not place on the record any reasoning that would indicate that he had actually engaged in a totality of the circumstances analysis. . .

Diss. op. at 7. Accordingly, while the majority finds “no reason to believe that the military judge overlooked or gave insufficient weight” to the totality of the circumstances, the dissenters find “that there is an insufficient basis to conclude the military judge conducted the required totality of the circumstances analysis in this case.” Diss. op. at 8.

Finally, Judge Ohlson specifically addresses the Manson weighing of “the corrupting effect of the suggestive identification itself.” Diss. op. at 8 (quoting Manson, 432 U.S. 114). Rejecting the military judge’s general statement that he considered all the surrounding circumstances as insufficient to “demonstrate that he conducted the required weighing,” diss. op. at 8 n.9, Judge Ohlson concludes that:

 Nowhere in the military judge’s summary of the relevant law or his application of the law does he reference or apply this additional analytical step. Therefore, there is “clear evidence” in the record that overcomes the presumption that the military judge knew and properly applied the law. United States v. Rapert, 75 M.J. 164, 170 (C.A.A.F. 2016).

Diss. op. at 8-9.

The dissenters conclude that because the military judge committed “significant errors,” his “decision is not entitled to deference.” Diss. op. at 9. They then determine that “the record before us does not provide a sufficient basis to conclude that the in-court identification was reliable,” diss. op. at 9, and that the in-court identification was not harmless, diss. op. at 10-11. Accordingly, the dissenters would reverse Criswell’s convictions and authorize a rehearing.

The divide between the majority and the dissenters is significant. They don’t just disagree on the result; they also disagree on the legal test and the standard of review. The only common ground seems to be their acknowledgements that eyewitness identifications are problematic in any case.

Case Links:
• ACCA decision
• Appellant’s brief
• Appellee’s (Army Gov’t App. Div.) brief
• Appellant’s reply brief
• Law student amicus in support of appellant
• Law student amicus in support of the appellee (Gov’t Div.)
• Blog post: Argument preview
Oral argument audio
CAAF opinion
Blog post: Opinion analysis

6 Responses to “Opinion Analysis: Admission of an in-court identification is (barely) affirmed by applying a highly-deferential standard of review, in United States v. Criswell”

  1. SgtDad says:

    it’s a little late, I know, but may I suggest Prof. Elizabeth Loftus, PhD, as an expert for the defense?  This is utterly insane.

  2. J.M. says:

     tainted by the previous highly suggestive out-of-court identification process

    No more tainted then CID taking an accuser to stakeout the accuseds apartment building and get a good look at him an hour before she ids him in a photo lineup.
    Perhaps there should be a focus on holding CID (and NCIS and OSI) to a higher standard when it comes to the basics, such as conducting a photo lineup properly.

  3. stewie says:

    So based on this, the only hindsight possible solution for the defense would be to present the evidence that the military judge suppressed (the initial show-up identification) and argue to the panel that this was highly suggestive and tainted her in-court identification.
    Of course, by doing so, likely killing future appellate paths for an uncertain result at best. Pretty Hobbesian choice there.
    Show-up identification should be banned. Period. We already know eyewitness identification is right about 50 percent of the time. That likely drops with a show-up, and if it’s cross-racial identification, it’s even worse.
    Yet panels consider it some of the strongest proof. Better start thinking about eyewitness experts to testify at trial for the defense about how unreliable it is.

  4. Alfonso Decimo says:

    In U.S. v. Abdirahman (NMCCA 2008), the court reasoned that while “none of the assignments of error raised by the appellant, or otherwise considered by this court, constitute reversible error individually, we find that the accumulation of errors . . . require us to evaluate the fairness of the appellant’s trial using the cumulative error doctrine.” Would that case have a different result ten years later?

  5. Charlie Gittins says:

    I had a cross-racial identification case in Korea some years ago and spoke to Elizabeth Loftus about being an expert for the reliability of cross-racial identification; got her fees requirement etc and made an expert request.  CA denied; took it to the MJ and said I would be happy with an equivalently qualified GOV expert in lieu of Dr. Loftus, but the expert testimony was important to the case.  Judge told the GOV to find one or pay Dr. Loftus.  Case settled favorable to the officer client.  Funny how that works.

  6. Fisch says:

    I’m not a big proponent of going judge alone.  There are some cases where I would be more inclined to go MJA, such as I know my client is not going to testify and the Judge has a good grasp of reasonable doubt.  But, in this case, had Criswell gone panel and defense counsel brought a timely written motion, then the M.J. would have ruled on the motion and fully developed a record on the motion to suppress the in court identification.  Instead, it appears to me that he possibly could have heard the entire case, deferred his ruling, determined the guy was guilty, then tailored his ruling on defense counsel’s last minute identification suppression motion to survive appellate scrutiny.
    With a panel, you set up the appellate issue with a timely motion to suppress and call in a cross racial identification expert to testify about how they are unreliable if you go with the SODDI defense.  But, the TC is going to get into evidence the admission to CID that Criswell spoke to her, and on cross the agent will admit that appellant denied saying and doing inappropriate things to her, so you would have to figure out a motive to fabricate if you went with the “she’s lying” defense.  I think it’s difficult for panel members to grasp the idea that something happened and she just misidentified him, or she’s lying and nothing actually happened to her.
    But, either way, what might have helped, depending on Criswell’s demeanor, is his testimony denying that he was wearing a bandana or saying or doing something inappropriate to her.  Then again, the wrong demeanor could have cemented his conviction.