Cookie cutter credibility findings are the antithesis of the individualized determination required in asylum cases. The integrity of the adjudicative process depends on judges reviewing each case on its merits. That integrity is called into question when boilerplate findings masquerade as individualized credibility determinations. Regrettably we have such a case here.

Uthayarosa Paramasamy, a young Tamil woman from Sri Lanka, petitions for review of a decision by the Board of Immigration Appeals ("Board"). The Board denied Paramasamy's application for asylum and withholding of removal, relying on the Immigration Judge's ("IJ's") adverse credibility determination. The IJ's adverse credibility finding focused on demeanor observations that were worded identically to findings in two other opinions that the judge issued in the same week. Many of the remaining findings are not supported by the record, or are an improper basis for making a credibility determination. Taken as a whole, these deficiencies are so significant that we conclude that the Board's decision is not supported by substantial evidence.

Paramasamy testified that she fled persecution by the Sri Lankan Army which detained and sexually assaulted her because she was a Tamil and consequently suspected of being a member of the rebel Liberation Tigers of Tamil Elam ("Tigers," or sometimes referred to in the record as "LTTE"). In fact, she claims that she was not a member and that she endured significant hardship resisting the Tigers because of her opposition to their terrorist tactics.

Paramasamy's account is consistent with reports in the record that document the situation in Sri Lanka at the time of her asylum application. The Sri Lankan Army had been engaged in an armed struggle with the Tigers for many years. The Tigers controlled portions of the northern and eastern areas of the country, with the ultimate objective of establishing an independent Tamil state. The Army controlled the south. According to the State Department country report, "both sides mistreat prisoners and arrest suspected opponents on an arbitrary basis." In the Army-controlled areas, young Tamils were at particular risk of persecution as suspected Tigers. Indeed, Paramasamy stated that her older brother was arrested by the Army and disappeared in 1990.

Paramasamy originally lived in Jaffna, a northern city under Tiger control. The Tigers attempted to recruit her forcibly as a young girl, but she refused because she did not agree with their terrorist methods. Her family paid a substantial sum to buy her freedom. This testimony is consistent with reports that the Tigers used forced conscription and a taxing system to support their organization.

Independent accounts in the record corroborate that, in 1995, the Tigers "emptied the city [Jaffna] by force" due to the Army's advance. Paramasamy stated that as she was fleeing, the Army arrested her and detained her in an army camp for about a month, although there was conflicting testimony about the time period that she was held. She was threatened and sexually assaulted by soldiers who removed her clothes and behaved "in an indecent manner." She testified to the assault only at the hearing, not during her previous interviews. Paramasamy explained that she could not tell the asylum officer about these sexual details because he was a man and relating the sexual assault was embarrassing. She later told her attorney and the IJ because they are women.

After her release, Paramasamy rejoined some of her family in a northern town under Army control. As a Tamil, she was required to register with the government every month and was frequently detained in the street and harassed. She was able to obtain permission to go to the national capital approximately 100 miles south, but only for two months. There, she said, the harassment continued, including frequent invasions of her home by government officials who conducted body searches. Finally, she fled to the United States, apparently attempting to rejoin her family members who, because of the harassment, have fled one by one to Canada. She does not speak English.

In an opinion that relied heavily on demeanor findings identical to those contained in two other cases before her, Immigration Judge Ho determined that Paramasamy was not credible. The IJ also highlighted perceived inconsistencies and failures of proof. She speculated that "[g]eneral fear of unsettled conditions in Sri Lanka and the anticipation of better job opportunities," as well as "a desire to leave the country to get married" were more likely reasons for Paramasamy's flight. Finally, the IJ cited Paramasamy's use of false documents to travel as "background" for the credibility determination.

The following paragraphs represent the IJ's findings on Paramasamy's demeanor. With the exception of the erratically altered pronouns, this passage is word-for-word identical to the credibility findings from this IJ's decision in Somasundaram, INS File No. Avr-guq-syr, and, save for the bracketed italicized detail, from her decision in Karunakaran, INS File No. Aqx-szv-tkn.1

With regard to the applicant's demeanor and her manner of recounting her alleged experiences, in the Court's judgment, there was an unnatural manner in her delivery of her testimony without the occasional pauses one would expect while she stopped to remember the details of terrible experiences. There was no visible change in her countenance or signs of emotional upheaval except at one point later in the proceedings when she was not describing the dark room [the lack of food, the dark room], but rather, her wish not to return to Sri Lanka in the end of her testimony.

The court, in its judgment, did not see any indication in the respondent's countenance or manner of speech indicating if he [sic] was reliving these terrible experiences. In fact, the applicant failed to mention that she was abused by the Sri Lankan Army prior to the hearing on the merits of the case. [In fact, the applicant failed to mention that he was burned by cigarettes by the soldiers in January, 2000. ...]. Since the abuse [the cigarette burn] is indicative of torture, and is material to her asylum and torture convention claim, the failure to mention it at the merits' hearing [sic] is material inconsistency between her testimony and her statement to the asylum officer, immigration officer and her own asylum application.

The Court has made allowance for the fact that she testified through an interpreter. This may be difficult for the applicant in recounting his [sic] experiences, but the court is not satisfied as the sincerity of the respondent based on her demeanor. Again, this determination is not dispositive of the claim, but provides a background for other observations.

The Board dismissed the appeal, holding that the copied passages did not provide a basis to disturb the demeanor findings. The Board relied on certain findings it believed were individualized and supportive of the adverse credibility determination. Finally, it reasoned that reliance on the use of false documents for travel, although improper, "was not a significant factor in the Immigration Judge's credibility analysis."

Congress has directed the Attorney General to consider asylum requests from refugees who arrive on our shores, 8 U.S.C. §§ 1158(b), 1225(b)(1), and has mandated that they not be removed to countries where their life or freedom would be threatened, 8 U.S.C. § 1231(b)(3). Paramasamy's extremely serious claims must be fairly measured against the standards established by Congress.

We review the Board's determination that Paramasamy was not eligible for asylum, 8 U.S.C. § 1252, and must uphold it if it is supported by substantial evidence. INS v. Elias-Zacarias, 502 U.S. 478, 481, 112 S. Ct. 812, 117 L. Ed. 2d 38 (1992); Rios v. Ashcroft, 287 F.3d 895, 899 (9th Cir. 2002). Because the Board adopted IJ Ho's decision, we must also review her reasoning. Lopez-Reyes v. INS, 79 F.3d 908, 911 (9th Cir.1996). Examining both opinions under this deferential standard, we nonetheless conclude that they are unsupported by the requisite substantial evidence.

I. BOILERPLATE DEMEANOR "FINDINGS"

We begin with the proposition that we accord credibility findings, and particularly the IJ's demeanor findings, substantial deference. Abovian v. INS, 219 F.3d 972, 977-78(9th Cir.2000) (as amended). This deference presupposes, however, that "[e]ach case is evaluated on its own merits." Platero-Cortez v. INS, 804 F.2d 1127, 1130 (9th Cir.1986). Because of the basic tenet that each petitioner is entitled to an individualized hearing, it is no surprise then that we have repeatedly insisted that we "will not tolerate boilerplate" decisions. Ghaly v. INS, 58 F.3d 1425, 1430 (9th Cir.1995). The rationale is obvious: "Boilerplate opinions ... neither afford the petitioner the BIA review to which he or she is entitled, nor do they provide an adequate basis for this court to conduct its review." Castillo v. INS, 951 F.2d 1117, 1121(9th Cir.1991).

Although not a high bar, the Board must "provide a comprehensible reason for its decision sufficient for us to conduct our review and to be assured that the petitioner's case received individualized attention." Ghaly, 58 F.3d at 1430. The Board and the IJ failed to meet even this lenient test: The IJ's demeanor findings were, for the most part, identically worded in three different cases, and contained disturbing inaccuracies. The Board compounded the problem by citing the boilerplate demeanor language as evidence that the findings were individualized.

We are particularly troubled by the identical description of the demeanor of three different witnesses. Of all matters passed upon by a hearing judge, conclusions about demeanor are the least susceptible to generalized observations. Rather, the very personal nature of demeanor demands individualized findings. It strains credulity that three different people would testify in the same "unnatural manner ... without ... occasional pauses." It is equally improbable that all three petitioners had identical stony countenances except at the same "point later in the proceedings" describing their "wish not to return to Sri Lanka." Could each of the three petitioners fail to show a "change in countenance or signs of emotional upheaval" at exactly the same time in the hearing sequence? We fear these passages demonstrate the IJ's "predisposition to discredit" the testimony, rather than any lack of credibility on the part of the witnesses. See Garrovillas v. INS, 156 F.3d 1010, 1015 (9th Cir.1998).

The demeanor findings are also inaccurate. The IJ twice refers to Paramasamy, a woman, using male pronouns. This error is particularly troubling in an account that purports to be a direct observation of a witness. The decision also incorrectly accuses Paramasamy of failing to mention "the abuse" at the merits hearing. In fact, she mentioned the sexual assault at the merits hearing but not during her previous interviews. It was another petitioner, Karunakaran, who allegedly failed to include details in his hearing testimony that he had mentioned previously.

Far from correcting the problem, the Board recharacterized the boilerplate findings as "individualized" findings. It held that the IJ had given individualized attention to the cases in part because: "She additionally observed that the respondent showed no emotion when describing the time she spent confined in the dark room, yet she became very emotional at the end of the hearing about not wanting to return." Because this "finding" is verbatim identical to that of the other decisions, the Board apparently parroted, rather than evaluated, the IJ's identical demeanor findings.2

This is not the first time we have considered the Board's reliance on boilerplate language. In Castillo-Villagra v. INS, 972 F.2d 1017, 1023 (9th Cir.1992), we held that the Board's use of "language identical to that used in a large number of other cases before it involving anti-Sandinista Nicaraguans" failed to provide the requisite individualized evaluation. In fact, this case is even more compelling because it is difficult to discern a good faith explanation for the boilerplate demeanor "findings," unlike the administrative notice of changed country conditions at issue in Castillo. The boiler-plate findings here are not simply a question of typographical errors or a case of computer "copy and paste" on a common point of law. We do not suggest that a decision cannot, if based on individual evidence, include portions that are common to other decisions. Indeed, within our own decisions and those of the district courts, it is not uncommon to see what might be described as standard verbiage setting forth a well-accepted legal principle, such as a standard of review. But demeanor is different. Demeanor reflects "a way of acting," "behavior," "bearing," and "outward manner." New Shorter Oxford English Dictionary 628 (1973). The findings here detail demeanor and might well have been accurate for one of the petitioners — but which one? On this record we cannot tell. As in Castillo and Garrovillas, a remand for individualized, impartial evaluation is necessary.

The identically worded passages are sufficiently troubling to warrant a remand, but the remainder of the opinion reinforces, rather than assuages, our concerns with respect to the adverse credibility determination. Both the Board and the IJ relied heavily on errors and assumptions that do not form an adequate basis for an adverse credibility determination. We identify four areas of concern: the Board's characterization of Paramasamy's smile during the hearing; the use of a boiler-plate hypothesis to reject evidence; perceived inconsistencies not based on the evidence; and reliance on hypothetical corroborative evidence.

Emblematic of the substitution of assumption for fact, the Board inaccurately stated that Paramasamy smiled "when testifying about alleged sexual assaults." In fact, the record contains no such observation. Instead, according to the transcript, the IJ stated that Paramasamy smiled when explaining that she revealed the sexual assault at the hearing, but not during prior interviews, because the judge was a woman. The judge's question at the hearing confirms this timing: "Why are you smiling when you said, I'm telling you because you are a woman?" The difference is critical; the facts as they appear in the record do not support the Board's articulated basis for an adverse credibility determination. See Stoyanov v. INS, 149 F.3d 1226, 1227-28 (9th Cir.1998) (remanding where Board relied on mistaken interpretation of critical date).

Similarly, the IJ improperly substituted her own hypothesis for the evidence in the record. Acknowledging that Paramasamy was afraid to return to Sri Lanka, the judge then speculated about Paramasamy's "real" motives: "a desire to leave the country to come to get married in the United States or Canada," or "[g]eneral fear of unsettled conditions in Sri Lanka and the anticipation of better job opportunities." She hypothesized identical motives for Somasundaram. An immigration judge's personal conjecture "cannot be substituted for objective and substantial evidence." Bandari v. INS, 227 F.3d 1160, 1167 (9th Cir.2000) (holding immigration judge's belief that petitioner would have bled from described beating to be an improper basis for credibility determination). Here, boilerplate speculation is not substantial evidence.

A third assumption without support in the record is the assertion that, if true, Paramasamy would have detailed the Army sexual assault earlier in the asylum process. Paramasamy's failure to elaborate the previously asserted claim of Army abuse cannot reasonably be characterized as an inconsistency. See Akinmade v. INS, 196 F.3d 951, 956 (9th Cir.1999); Abovian, 219 F.3d at 979; Turcios v. INS, 821 F.2d 1396, 1400-01 (9th Cir.1987).

Paramasamy was not asked in her airport interview about the details or conditions of her Army confinement, so her failure to mention the sexual assault there cannot be considered an inconsistency. See Akinmade, 196 F.3d at 956. As we have previously observed, "an arriving alien who has suffered abuse during interrogation sessions by government officials may be reluctant to reveal such information during [her] first meeting with government officials in this country." Singh v. INS, 292 F.3d 1017, 1023 (9th Cir.2002) (quoting Balasubramanrim v. INS, 143 F.3d 157, 163 (9th Cir.1998)). That a woman who has suffered sexual abuse at the hands of male officials does not spontaneously reveal the details of that abuse to a male interviewer does not constitute an inconsistency from which it could reasonably be inferred that she is lying. Indeed, the assumption that the timing of a victim's disclosure of sexual assault is a bellwether of truth is belied by the reality that there is often delayed reporting of sexual abuse.3 Paramasamy provided a strong, unrebutted explanation for her reluctance to reveal details — her cultural reluctance to tell male interviewers that she had been violated. See Platero-Cortez, 804 F.2d at 1131 (overturning Board's determination that excuse for prior failure to apply for asylum was unconvincing); cf. Bryden & Lengnick, supra note 3, at 1229 (noting "national attitudinal changes" correlated to somewhat increased reporting of sexual assault in the U.S.). The perceived inconsistency, like the smile and the hypothetical motives, is without basis in the record.

Nor can the second interview be reasonably construed as inconsistent. There, Paramasamy stated that she had been detained and threatened by the Army and that she was afraid of sexual assault by the Army. Asked whether the soldiers had "physically harm[ed]" her, she replied in the negative:

Q. Did the soldiers physically harm you while they held you?

A. They threatened to kill me. They held me at gunpoint.

...

Q. Did the soldiers ever physically harm you? Did they beat you?

A. No, but they threatened me.

The Board and IJ attempted to characterize this negative reply as inconsistent with her later testimony that the soldiers touched her indecently. This characterization ignored Paramasamy's persistent assertions in the interview that she was at sexual risk from the Army were she to return. Further, it was far from obvious that "physically harm" would call to mind the sexual assault that Paramasamy later described as indecent touching. Nor does Paramasamy's response to the IJ's questioning at the hearing support this assertion:

Q. Ma'am, if a person touches you indecently, is that physical harm?

A. What do you mean by physical harm?

Q. Well, ma'am, if a person touches you when you don't want them to, is that harming you physically?

A. They were touching me and feeling me, and satisfying their desires.

Q. Ma'am please listen to my question, okay?

...

Q. When the men touch you to satisfy themselves, is that harming you physically? Just answer yes or no.

A. Yes, it is, uh, stress.

The final affirmative response cannot, in context, reasonably be construed to mean that the prior interview was inconsistent. The answer is qualified by an explanation that contradicts any effort to equate the assault with physical beating of the type described in the interview questions. And, the response comes at the close of over a transcript page of hostile cross-examination by the judge. In sum, nothing in the record supports the Board's and IJ's characterizations of Paramasamy's account of the sexual abuse as inconsistent.

Finally, the IJ concluded that Paramasamy should have corroborated her account by asking her mother, in Canada, to write a letter. The judge counted the lack of such corroboration against Paramasamy's credibility, and the Board agreed. However, the judge never questioned Paramasamy about this perceived deficiency, nor was Paramasamy given an opportunity to respond to this concern. See Diallo v. INS, 232 F.3d 279, 288-90 (2d Cir.2000) (holding Board failure to assess explanation for lack of evidence warranted reversal). It is also significant that although in her discussion the IJ called for additional detail about the assault, she twice cut off Paramasamy's attempts to provide further information at the hearing. The IJ's unilaterally imposed barrier in the form of hypothetical corroborative evidence is yet another factor that undermines the decision.

III. IMPROPER RELIANCE ON FALSE TRAVEL DOCUMENTS

As the Board correctly noted, the use of false documents for travel is not a proper basis for an adverse credibility determination. Akinmade, 196 F.3d at 955. Nonetheless, the IJ relied on this factor to "provide[] a background for other considerations." The Board reasoned that the fraudulent document concern "was not a significant factor in the Immigration Judge's credibility analysis, and that the decision would be supported without it." That rationale loses force once the other bases for an adverse credibility finding fall away. Under the circumstances here, the reliance on this improper factor further erodes our confidence in the existence of reasonable basis for the adverse credibility determination.

We cannot uphold the adverse credibility finding as supported by substantial evidence in the record. Although we note that Paramasamy's account appears consistent and corroborated, this is not a case where we can say with the requisite certainty that the Board would have been unreasonable in denying relief. Singh, 292 F.3d at 1025. We therefore remand for a new hearing, with an individualized determination of credibility.4 See Briones v. INS, 175 F.3d 727, 729-30 (9th Cir.1999) (en banc).

PETITION GRANTED and REMANDED for a new hearing.