By Deirdre Dlu­goles­ki1

It almost goes with­out say­ing that in the con­text of asy­lum law, a judge’s deci­sion about whether or not to believe an applicant’s tes­ti­mo­ny can have life or death con­se­quences. But the statu­to­ry require­ments of the Refugee Act make sup­port­ing evi­dence, in addi­tion to the applicant’s tes­ti­mo­ny, par­tic­u­lar­ly impor­tant in most cas­es. Immi­gra­tion Judges’ (IJs’) deci­sions on cred­i­bil­i­ty and/or statu­to­ry eli­gi­bil­i­ty for asy­lum often turn on a deci­sion about which sources of evi­dence to believe – weigh­ing, for exam­ple, a news report from the asy­lum seeker’s home coun­try against a coun­try con­di­tion report from the State Depart­ment that the Immi­gra­tion and Nat­u­ral­iza­tion Ser­vice (INS) offi­cer has intro­duced. Although there are some guide­lines for how the IJ and, on appeal, the Board of Immi­gra­tion Appeals (BIA) must con­sid­er dif­fer­ent sources of evi­dence in rela­tion to each oth­er, IJs and the BIA have wide dis­cre­tion in these deci­sions, and courts are expect­ed to defer unless the facts would com­pel a rea­son­able fact find­er to con­clude oth­er­wise. The issue, at its core, is this: if an IJ reviews an asy­lum seeker’s sup­port­ing doc­u­ments and decides they’re unre­li­able, what can the fed­er­al courts of appeals do about it?

This Con­tri­bu­tion will argue that the scope of sub­stan­tial evi­dence review of sup­port­ing doc­u­men­ta­tion should be broad, and that courts play an impor­tant role in hold­ing the BIA account­able for bas­ing its deci­sions on reli­able infor­ma­tion. Because courts review the IJ’s deci­sion as if it were the BIA’s when the BIA sum­mar­i­ly affirms,2 the Con­tri­bu­tion will for simplicity’s sake refer to judi­cial review of the BIA.

* * * * *

Under the Refugee Act of 1980, asy­lum appli­cants must show they are unable or unwill­ing to return to their home coun­try because of “per­se­cu­tion or a well-found­ed fear of per­se­cu­tion on account of race, reli­gion, nation­al­i­ty, mem­ber­ship in a par­tic­u­lar social group, or polit­i­cal opin­ion.”3 Asy­lum appli­cants can meet this require­ment in either or both of two ways. First, they can show that they have expe­ri­enced per­se­cu­tion in the past based on one of the pro­tect­ed grounds (which cre­ates a pre­sump­tion that the appli­cant has a well-found­ed fear of future per­se­cu­tion).4 Sec­ond, they can pro­vide evi­dence of the con­text in their home coun­try that shows an objec­tive­ly rea­son­able fear of future per­se­cu­tion, should they return.5 In either case, the REAL ID Act of 2005 requires appli­cants to show that the per­se­cu­tion was based on one of the pro­tect­ed grounds as the “cen­tral motive,”6 and that they could not avoid tar­get­ed per­se­cu­tion by mov­ing to anoth­er part of the coun­try – if they had the option of rea­son­able relo­ca­tion, their eli­gi­bil­i­ty for asy­lum in the Unit­ed States is pre­clud­ed.7 All told, the evi­den­tiary bur­dens of the asy­lum process are high, and vir­tu­al­ly neces­si­tate that appli­cants find sec­ondary sources of infor­ma­tion about the con­di­tions in their home coun­try, such as news reports and human rights fact sheets.

The IJ’s role of assess­ing the cred­i­bil­i­ty of this evi­dence is much greater than that of a judge in a nor­mal civ­il tri­al because the asy­lum process involves a low­er stan­dard of proof. The appli­cant only has to pro­vide “reli­able, spe­cif­ic, objec­tive sup­port­ing evi­dence,” which can be either direct or cir­cum­stan­tial.8 For asy­lum appli­cants, who are not infre­quent­ly flee­ing for their lives, a low­er stan­dard of proof makes sense. Both courts and the BIA acknowl­edge that, in some cas­es, it is sim­ply impos­si­ble for appli­cants to pro­duce doc­u­men­tary evi­dence; appli­cants’ tes­ti­mo­ny in these cir­cum­stances can be ade­quate for eli­gi­bil­i­ty in and of itself.9

But it’s some­what hard­er to jus­ti­fy a sim­i­lar­ly low stan­dard of proof for the gov­ern­ment, which faces none of the chal­lenges asy­lum seek­ers have to over­come. In spite of a com­par­a­tive sur­plus of resources with which to pro­duce sound evi­dence, the stan­dard of proof for the INS is as low as it is for the appli­cant – in asy­lum pro­ceed­ings, the gov­ern­ment isn’t bound by the Fed­er­al Rules of Evi­dence, and, because it is a civ­il case, none of the high­er pro­ce­dur­al pro­tec­tions that a crim­i­nal tri­al would entail apply.10 As long as the evi­dence doesn’t vio­late the applicant’s due process rights, it’s admis­si­ble against the appli­cant.11 And the INS gives broad lee­way in the choice of sources to con­sult – the Basic Law Man­u­al of the INS spec­i­fies that, in addi­tion to State Depart­ment reports, the offi­cer can draw on coun­try pro­files devel­oped by the INS Resource Infor­ma­tion Cen­ter, and “reports from Amnesty Inter­na­tion­al and oth­er rep­utable orga­ni­za­tions, includ­ing aca­d­e­m­ic insti­tu­tions.”12 In prac­ti­cal­i­ty, the require­ment of obtain­ing infor­ma­tion from a “rep­utable” source is for­giv­ing – the agency has used, for exam­ple, let­ters from con­sular staff and air­port inter­views in the past.13 More­over, asy­lum appli­cants often have to over­come not only the evi­dence of the oppos­ing coun­sel, but also of the IJ her­self. Asy­lum tri­als are unique – the IJ both assess­es and intro­duces doc­u­men­tary evi­dence (with the same broad dis­cre­tion as the rest of the INS).14

 In sum­ma­ry, either side can intro­duce any evi­dence that’s suf­fi­cient­ly pro­ba­tive and “fun­da­men­tal­ly fair” – includ­ing hearsay evi­dence.15 But the par­ties’ respec­tive capa­bil­i­ties of pro­duc­ing that evi­dence could gen­er­ous­ly be called asym­met­ri­cal.  Essen­tial­ly, the gov­ern­ment, be it through the INS or the IJ, can intro­duce almost any evi­dence it wants into the record to refute an asy­lum seeker’s claims and send her back to a war zone.

* * * * *

So what can fed­er­al courts of appeals do to hold the agency account­able for bas­ing its deci­sions on reli­able information?

Facial­ly, it seems, not much. The same sub­stan­tial evi­dence stan­dard that applies to all of the BIA’s find­ings of fact applies to the ques­tion of doc­u­men­tary evi­dence as well, and the Supreme Court has made clear that the stan­dard is high­ly def­er­en­tial to the BIA.16 In INS v. Elias-Zacarias, the Court held that the review­ing court must affirm a deter­mi­na­tion of statu­to­ry inel­i­gi­bil­i­ty by the BIA unless the “evi­dence pre­sent­ed was so com­pelling that no rea­son­able fact find­er could fail to find” eli­gi­bil­i­ty for asy­lum.17 To reverse the BIA, “we must find that the evi­dence not only sup­ports that con­clu­sion, but com­pels it.”18 More­over, specif­i­cal­ly in rela­tion to doc­u­men­tary evi­dence pre­pared by the U.S. gov­ern­ment (for exam­ple, State Depart­ment reports and con­sular cor­re­spon­dence), courts have estab­lished a pre­sump­tion of reli­a­bil­i­ty – that is, that gov­ern­ment employ­ees can be pre­sumed to accu­rate­ly pre­pare their reports with­out bias.19 Grant­ed, the agency’s reliance on these doc­u­ments is cab­ined – the IJ must review all the evi­dence under the total­i­ty of the cir­cum­stances approach and can­not rely on “triv­ial incon­sis­ten­cies” that don’t impli­cate the petitioner’s verac­i­ty or the heart of the claim for an adverse cred­i­bil­i­ty deter­mi­na­tion or denial.20 In gen­er­al, though, as long as the agency has ade­quate­ly explained the basis for its deci­sion on the record (i.e., enough expla­na­tion to be under­stand­able and demon­strate the basis for its deci­sion), courts are expect­ed to defer.21

In prac­tice, how­ev­er, courts vary wide­ly in how they apply the sub­stan­tial evi­dence stan­dard. The Ninth Cir­cuit, for exam­ple, not only hears the most asy­lum cas­es out of all the cir­cuits, but has earned a rep­u­ta­tion for lib­er­al­ly con­stru­ing what it means for the evi­dence to “com­pel” a find­ing of asy­lum eli­gi­bil­i­ty – fre­quent­ly putting them, as the gov­ern­ment has claimed, “in con­flict with oth­er courts of appeals, which gen­er­al­ly respect the BIA’s role as fact-find­er.”22 But the Ninth Cir­cuit isn’t alone in push­ing the bounds of def­er­ence under sub­stan­tial evi­dence review. Specif­i­cal­ly in rela­tion to doc­u­men­tary evi­dence, the Sec­ond, Third, Fourth, Sixth, and Eighth Cir­cuits have all held that the IJ can­not rely on unau­then­ti­cat­ed con­sular let­ters in mak­ing an adverse deci­sion.23 In Bal­a­sub­ra­man­rim v. INS and Sen­athi­ra­jah v. INS, the Third Cir­cuit pro­vid­ed stan­dards for the BIA to con­sid­er in eval­u­at­ing state­ments asy­lum appli­cants make dur­ing inter­views with INS offi­cers.24 In Zhen Nan Lin v. Unit­ed States DOJ, the Sec­ond Cir­cuit delin­eat­ed stan­dards for the BIA to fol­low in review­ing gov­ern­ment reports, cov­er­ing the iden­ti­ty of the inves­ti­ga­tors, the pur­pose of the inves­ti­ga­tion, and the meth­ods of ver­i­fi­ca­tion used.25

The wide­spread con­cern in the cir­cuit courts with the reli­a­bil­i­ty of evi­dence the gov­ern­ment intro­duces should be enough to indi­cate that the scope of judi­cial review should be broad. But this ques­tion can also be resolved by look­ing at the sub­stan­tial evi­dence stan­dard itself.

While the stan­dard func­tions as a ceil­ing on the court’s abil­i­ty to review the agency, it also serves as a floor for the infor­ma­tion the agency has to pro­vide. The Third Cir­cuit, among oth­ers, has clar­i­fied that the agency is required to sup­port its find­ings with sub­stan­tial evi­dence, as artic­u­lat­ed by the Supreme Court in Elias-Zacarias.26 It rea­son­ably fol­lows, then, that the BIA can’t pro­vide the court a sub­stan­tial enough record to review if its rea­son­ing depend­ed, in turn, on slen­der evi­dence or sources of ques­tion­able reli­a­bil­i­ty.27 This log­ic is con­sis­tent with SEC v. Chen­ery Corp., the Supreme Court prece­dent that requires the BIA (like all oth­er exec­u­tive agen­cies) to ade­quate­ly explain its deci­sions on the record in the first place.28 Chen­ery sought to ensure that courts could “test” the agency’s admin­is­tra­tive action through a mean­ing­ful review of the deci­sion.29 With this in mind, the approach of the cir­cuits men­tioned above makes sense. When it comes to doc­u­men­tary evi­dence, the scope of judi­cial review under the sub­stan­tial evi­dence stan­dard should be broad.

A broad stan­dard is espe­cial­ly appro­pri­ate because the BIA has thus far failed to accom­plish the main goal that courts have cit­ed as grounds for insu­la­tion from judi­cial review. Although the Supreme Court has repeat­ed­ly empha­sized the need for uni­for­mi­ty in the con­text of immi­gra­tion, IJs have var­ied wild­ly in their accep­tance rates for asy­lum appli­ca­tions. A study under­tak­en from 2000 to the first few months of 2005, for exam­ple, showed the medi­an rate of asy­lum denial at 65%, but indi­vid­ual denial rates rang­ing from 10% to 98%.30 And even if this weren’t the case, the BIA is bound to fol­low the cir­cuit court review­ing it. For exam­ple, an IJ request­ing coun­try con­di­tion infor­ma­tion from the State Depart­ment would be free, in the Fifth Cir­cuit, to eval­u­ate it how­ev­er she pleased, but an IJ in the Sec­ond Cir­cuit would have to review it in light of the fac­tors in Zhen Nan Lin.31 More­over, in spite of the pre­sump­tion of reli­a­bil­i­ty for gov­ern­ment-pro­duced doc­u­ments, some cir­cuits have expressed con­cerns with the exec­u­tive branch being in charge of this infor­ma­tion. The Sec­ond Cir­cuit, for exam­ple, not­ed that because the exec­u­tive branch (and, by exten­sion, the State Depart­ment) is unique­ly con­cerned with the Unit­ed States’ rela­tion­ship with oth­er coun­tries, it is like­ly that the State Department’s reports would be sub­ject to the administration’s bias.32

There are also strong pol­i­cy rea­sons for courts to hold the BIA account­able for pro­vid­ing enough sub­stance to review. In 2002, then-Attor­ney Gen­er­al Ashcroft intro­duced new reg­u­la­tions to stream­line the BIA, reduc­ing the num­ber of review­ing judges to one (from three) and expand­ing the cir­cum­stances in which the BIA could issue an affir­mance with­out opin­ion (AWO) of the IJ’s deci­sion. The 2002 reforms had dis­as­trous impli­ca­tions for the qual­i­ty of the BIA’s deci­sions. BIA appeals have not only tak­en up a dis­pro­por­tion­ate­ly large share of the case­load in fed­er­al courts of appeals, but have also been sub­ject to a high­er rever­sal rate. In 2001, BIA deci­sions con­sti­tut­ed only 3% of fed­er­al appeals cas­es. By 2004, this had jumped to ful­ly 25% – 50% in the Sec­ond Cir­cuit alone.33 The BIA’s rever­sal rate cor­re­spond­ing­ly spiked. By 2005, the Sev­enth Cir­cuit over­turned 40% of the opin­ions decid­ed on the mer­its by the IJ (com­pared to a rever­sal rate for oth­er civ­il cas­es that year of 18%).34 Cir­cuit court judges have open­ly expressed their lack of con­fi­dence in the agency – most notably, Judge Pos­ner, who wrote that the post-2002 admin­is­tra­tive adju­di­ca­tion of immi­gra­tion cas­es “has fall­en below the min­i­mum stan­dards of legal jus­tice.”35

The issue of account­abil­i­ty in ver­i­fy­ing evi­dence is only going to grow in impor­tance. With the rise of cit­i­zen jour­nal­ism, com­mu­ni­ty mon­i­tor­ing and fact-find­ing, and online plat­forms like YouTube and Face­book post­ing video footage of human rights vio­la­tions, it’s not unrea­son­able to imag­ine that asy­lum appli­cants may soon bring these sources to bear in their tes­ti­mo­ny.36 The BIA may soon have to assess a much wider vari­ety of sources than Con­gress ever imag­ined when it draft­ed the Refugee Act. Impor­tant­ly, as these kinds of non-tra­di­tion­al sources of infor­ma­tion have pro­lif­er­at­ed, ver­i­fi­ca­tion meth­ods have kept pace. Tools like the Elec­tron­ic Dis­cov­ery Ref­er­ence Mod­el (Duke Uni­ver­si­ty School of Law), the Ver­i­fi­ca­tion Hand­book (Euro­pean Jour­nal­ism Cen­tre), and Amnesty International’s Cit­i­zen Evi­dence Lab are only some of the increas­ing­ly wide array of user-friend­ly resources that can ver­i­fy video and social media con­tent.37 This means that, the­o­ret­i­cal­ly, for much of this kind of infor­ma­tion, the agency and the courts are equal­ly well sit­u­at­ed to ver­i­fy it. Oth­er courts have tak­en notice – most sig­nif­i­cant­ly, the Inter­na­tion­al Crim­i­nal Court issued its first arrest war­rant based entire­ly on social media posts in August 2017.38 It is crit­i­cal that this oppor­tu­ni­ty for review not be fore­closed in the Unit­ed States through stare deci­sis.

In the inter­est of pro­vid­ing a fair chance to asy­lum seek­ers, courts should hold the BIA account­able for giv­ing ade­quate con­sid­er­a­tion to diverse sources of evi­dence, espe­cial­ly when, if truth­ful, it sheds light on the ground sit­u­a­tion with more nuance than a coun­try report could achieve. It’s dif­fi­cult enough for most appli­cants to leave their home coun­try with any sup­port­ing doc­u­men­ta­tion at all, but with­out a broad­er judi­cial review, the BIA will con­tin­ue to shut out deserv­ing asy­lum seekers.


1. Deirdre Dlu­goles­ki is a 3L at New York Uni­ver­si­ty School of Law. This piece is a com­men­tary on the 2018 prob­lem at the UC Davis Asy­lum & Refugee Nation­al Moot Court Com­pe­ti­tion held in Davis, Cal­i­for­nia. The issue in the prob­lem dealt with whether an asy­lum appli­cant who had refused to join an armed group in open con­flict with the gov­ern­ment of his home coun­try could be eli­gi­ble on the grounds of per­se­cu­tion for polit­i­cal opin­ion, and what the scope of judi­cial review was of the IJ’s ini­tial denial. The views expressed in this arti­cle do not nec­es­sar­i­ly rep­re­sent the views of the author on this point of law; rather, this arti­cle is a dis­til­la­tion of one side of the argu­ment that the team rep­re­sent­ed at the UC Davis Asy­lum & Refugee Nation­al Moot Court Competition.
2. Elzour v. Aschroft, 378 F.3d 1143, 1150 (10th Cir. 2004).
3. See 8 U.S.C. § 1101(a)(42)(A) (2012).
4. See Mel­gar de Tor­res v. Reno, 191 F.3d 307, 311 (2d Cir. 1999)
5. See Abankwah v. INS, 185 F.3d 18, 22 (2d Cir. 1999).
6. See 8 U.S.C. § 1158(b)(1)(B)(i) (2012); see also Parus­si­mo­va v. Mukasey, 555 F.3d 734, 742 (9th Cir. 2009).
7. See 8 U.S.C. § 1231(b)(3)(B)(ii) (2012).
8. Cordero-Tre­jo v. INS, 40 F.3d 482, 491 (1st Cir. 1994) (dis­cussing evi­dence); see also Ramirez Rivas v. INS, 899 F.2d 864 (9th Cir. 1990) (peti­tion­er can pro­vide either direct or cir­cum­stan­tial evi­dence of the persecutor’s motives), vacat­ed and remand­ed with­out opin­ion by 502 U.S. 1025 (1992).
9. See Mat­ter of Mogharra­bi, 19 I. & N. Dec. 439, 445, 1987 BIA LEXIS 5.
10. See Zhen Nan Lin v. Unit­ed States Dep’t of Jus­tice, 459 F.3d 255, 268 (2d Cir. 2006) (“The gov­ern­ment is not required to com­ply with either the require­ments of the Fed­er­al Rules of Evi­dence or the height­ened pro­ce­dur­al pro­tec­tions of a crim­i­nal tri­al when seek­ing to have doc­u­men­tary evi­dence, such as a con­sular report, admit­ted in a removal pro­ceed­ing. Evi­dence is admis­si­ble pro­vid­ed that it does not vio­late the alien’s right to due process of law.”).
11. See id.
12. Unit­ed States Dep’t. of Jus­tice, INS, The Basic Law Man­u­al, U.S. Law and INS Refugee/ Asy­lum Adju­di­ca­tions 100 (1994).
13. See, e.g., Bal­a­sub­ra­man­rim v. INS, 143 F.3d 157, 162–63 (3d Cir. 1998); Sen­athi­ra­jah v. INS, 157 F.3d 210, 218–20 (3d Cir. 1998).
14. The IJ needs back­ground infor­ma­tion on the coun­try in ques­tion not only to make a cred­i­bil­i­ty deci­sion about the appli­cant, but also to deter­mine if asy­lum eli­gi­bil­i­ty is pre­clud­ed by the option of rea­son­able relo­ca­tion with­in the coun­try. See In re S‑M-J, 21 I. & N. Dec. 722, 728, 1997 WL 80984 (BIA 1997); 8 U.S.C. § 1252(b) (1994); see also Mat­ter of Dass, 20 I. & N. Dec. 120, 125, 1989 BIA LEXIS 27 (“Par­tic­u­lar­ly when the basis of an asy­lum claim becomes less focused on spe­cif­ic events involv­ing the respon­dent per­son­al­ly and instead is more direct­ed to broad alle­ga­tions regard­ing gen­er­al con­di­tions in the respon­den­t’s coun­try of ori­gin, cor­rob­o­ra­tive back­ground evi­dence that estab­lish­es a plau­si­ble con­text for the per­se­cu­tion claim (or an expla­na­tion for the absence of such evi­dence) may well be essen­tial. The more sweep­ing and gen­er­al a claim, the clear­er the need for an asy­lum appli­cant to intro­duce sup­port­ing evi­dence or to explain its absence.”) The BIA func­tions in the same man­ner – it has, on the record, not­ed that the UN’s Hand­book on Pro­ce­dures and Cri­te­ria for Deter­min­ing Refugee Sta­tus rec­om­mends the same pro­ce­dure. The Hand­book states that although the bur­den of proof rests on the asy­lum appli­cant, “the duty to ascer­tain and eval­u­ate all the rel­e­vant facts is shared between the appli­cant and the exam­in­er.” See In re S‑M-J, 21 I. & N. at 729; see also Hand­book on Pro­ce­dures and Cri­te­ria for Deter­min­ing Refugee Sta­tus under the 1951 Con­ven­tion and the 1967 Pro­to­col Relat­ing to the Sta­tus of Refugees ¶ 196 (U.N.H.C.R., Jan. 1992).
15. See Mar­tin-Men­doza v. INS, 499 F.2d 918, 921 (9th Cir. 1974) (deny­ing remand based on the admis­sion of state­ments the appli­cant made before flee­ing the coun­try, which made him unavail­able as a witness).
16. INS v. Elias-Zacarias, 502 U.S. 478, 483–84 (1992).
17. See id. see also Rusu v. INS, 296 F.3d 316, 325 n.14 (4th Cir. 2002)
18. Id. at 481 n.1.
19. See Espinoza v. INS, 45 F.3d 308, 310 (9th Cir. 1995). It’s impor­tant to note that this pre­sump­tion does not apply to doc­u­ments pro­duced by a par­ty oth­er than the U.S. gov­ern­ment or the sub­ject of the report. See Her­nan­dez-Guadar­ra­ma v. Ashcroft, 394 F.3d 674, 681 n.9 (9th Cir. 2005) (cit­ing Espinoza, 45 F.3d at 310).
20. See Yao­gang Ren v. Hold­er, 648 F.3d 1079, 1084 (9th Cir. 2011).
21. See, e.g., Kama­ra v. AG of the Unit­ed States, 420 F.3d 202, 211 (3d Cir. 2005); Dominguez v. Ashcroft, 336 F.3d 678, 680 (8th Cir. 2003); see also In re S‑M-J, 21 I. & N. at 728.
22. See, e.g., INS v. Ven­tu­ra, 123 U.S. 12, 16 (2002).
23. See Banat v. Hold­er, 557 F.3d 886, 892–93 (8th Cir. 2009)Anim v. Mukasey, 535 F.3d 243, 256–58 (4th Cir. 2008)Alexan­drov v. Gon­za­les, 442 F.3d 395, 407 (6th Cir. 2006)Ezeag­wu­na v. Ashcroft, 325 F.3d 396, 405–08 (3d Cir. 2003)
24. See Bal­a­sub­ra­man­rim v. INS, 143 F.3d 157, 162–63 (3d Cir. 1998) (not­ing that records sum­ma­riz­ing the asy­lum seeker’s state­ments in an inter­view is inher­ent­ly less reli­able than a ver­ba­tim account or tran­script); see also Sen­athi­ra­jah v. INS, 157 F.3d 210, 218–20 (3d Cir. 1998); Ram­sameachire v. Ashcroft, 357 F.3d 169, 179 (2d Cir. 2004).
25. See Zhen Nan Lin v. Unit­ed States Dep’t of Jus­tice, 459 F.3d 255, 271 (2d Cir. 2006).
26. See Dia v. Ashcroft, 353 F.3d 228, 244–45 (3d Cir. 2003).
27. See, e.g., Hor v. Gon­za­les, 421 F.3d 497, 500 (7th Cir. 2005) (explain­ing that a cen­sored press in an applicant’s home coun­try could not rea­son­ably be expect­ed to report the harm suf­fered); Alexan­drov v. Gon­za­les, 442 F.3d at 404 (vacat­ing an IJ’s find­ing of friv­o­lous­ness because of the flim­si­ness of sup­port­ing mem­o­ran­da from the Unit­ed States embassy in Sofia, Bul­gar­ia report­ing that doc­u­ments sub­mit­ted by the appli­cant were fraud­u­lent.); cf. Boroviko­va v. Unit­ed States DOJ, 435 F.3d 151, 157–58 (2d Cir. 2006) (hold­ing that a sworn affi­davit from the INS employ­ee was suf­fi­cient to sup­port the IJ’s adverse cred­i­bil­i­ty find­ing). See also In re S‑M-J, 21 I. & N. at 730 (explain­ing that the BIA expects “gen­er­al cor­rob­o­rat­ing evi­dence, from a reli­able source, of per­se­cu­tion of per­sons in cir­cum­stances sim­i­lar to an appli­cant”) (empha­sis added).
28. See SEC v. Chen­ery Corp., 332 U.S. 194, 196 (1947).
29. See Dia, 353 F.3d at 241.
30. Lind­sey R. Vaala, Bias on the Bench: Rais­ing the Bar for U.S. Immi­gra­tion Judges to Ensure Equal­i­ty for Asy­lum Seek­ers, 49 William & Mary L. Rev. 1011, 1023 (2007),
31. See Abdu­lai v. Ashcroft, 239 F.3d 542, 554 (3d Cir. 2001) (“The BIA is required to fol­low court of appeals prece­dent with­in the geo­graph­i­cal con­fines of the rel­e­vant cir­cuit.” (cit­ing Mat­ter of Ansel­mo (Inter­im Deci­sion), 20 I. & N. Dec. 25, 30–31, 1989 BIA LEXIS 11 (May 11, 1989))).
32. Tam­badou v. Gon­za­les, 446 F.3d 298, 302–03 (2d Cir. 2006).
33. Vaala, supra note 30, at 1023.
34. Id.
35. See Bensli­mane v. Gon­za­les, 430 F.3d 828, 829–30 (7th Cir. 2005).
36. See Deirdre Dlu­goles­ki, ‘We are all jour­nal­ists now’: 140 Journos and Turkey’s “counter-media” move­ment, Colum. Jour­nal­ism Rev. (May 20, 2013),
37. See Enrique Piracés, Col­lect­ing, pre­serv­ing, and ver­i­fy­ing online evi­dence of human rights vio­la­tions, Open­Glob­al­Rights (Jan. 30, 2018),
38. Pros­e­cu­tor v. Mah­moud Mustafa Musayf al-Wer­fal­li, Case No. ICC-01/11–01/17, Pub­lic War­rant of Arrest, ¶ 11 (Aug. 15, 2017),