Vigile v. Sava

535 F. Supp. 1002, 1982 U.S. Dist. LEXIS 11003

• Court: District Court, S.D. New York
• Decided: March 5, 1982
• Docket: 81 Civ. 7372 (RLC), 81 Civ. 7371 (RLC)
• Status: Published

Opinion

OPINION

ROBERT L. CARTER, District Judge.

These consolidated petitions for writ of habeas corpus were filed by eight Haitians who arrived in Florida between June and early July 1981. Along with 78 of their countrymen and women, petitioners were transferred from the Immigration and Naturalization Service (“INS”) Krome Avenue facility in Miami to the Service Processing Center (“SPC”), an immigration detention facility in Brooklyn, New York. They have been incarcerated at the SPC to the present date.

All eight petitioners have applied for political asylum in this country. All have been subject to exclusion proceedings since early August. Between August 12 and October 9, requests for parole pending final adjudication of the asylum claims were made on behalf of all petitioners. Defendant Charles C. Sava, INS District Director for the New York District, has the authority to grant or deny those requests. See 8 U.S.C. § 1182(d)(5)(A); 8 C.F.R. § 212.5. In all eight cases, Sava refused to exercise his discretion in favor of parole. 1

The stated reason for the two earliest parole denials was that the “information presented in [the] requests] ... are [sic] insufficient to warrant a change in the alien’s custody status.” Joint Exhibit (“JE”) 2 17, 24. The six early December denials restated the above rationale and added one of two additional justifications. Two responses noted that the Department of State had yet to reply to the asylum applications and that a final decision on custody status would not be made until Sava received such information. JE 19, 23. The remaining four decisions expressly took into consideration Department of State recommendations “in which they stated that the subject has failed to establish a well-founded fear of persecution upon return to Haiti.” JE 20, 21, 22, 25.

The habeas corpus petitions allege that Sava has abused his discretionary release authority by acting arbitrarily, capriciously and in sharp contrast to established parole policy. In addition, petitioners contend that Sava’s treatment of Haitian “boat people” has been discriminatory in violation of the due process clause of the fifth amendment and the United Nations Convention and Protocol Relating to the Status of Refugees (“Protocol”), a treaty to which the United States is a party. See 19 U.S.T. 6223, T.I.A.S. No. 6557. The government opposes the petitions as both substantively and procedurally infirm. It argues that the court lacks jurisdiction to review the District Director’s parole decision, that petitioners have presented the wrong standard for parole determinations, that defendant Sava properly exercised his discretion, that these aliens have no rights under the Constitution or the Protocol and that, in any event, neither the Protocol nor the Constitution has been violated.

JURISDICTION

Petitioners assert that jurisdiction is grounded in 28 U.S.C. § 2241, insofar as they are in custody pursuant to the Immigration and Nationality Act, 8 U.S.C. § 1101 et seq., and such custody allegedly is in violation of the Constitution, the Immigration and Nationality Act, the regulations of the INS, 8 C.P.R. § 1.1 et seq., and the Protocol. Respondent asserts that jurisdiction under the habeas statute is limited to review of the basis of petitioners’ incarceration and does not permit, in the absence of a more specific jurisdictional grant, review of parole determinations involving unadmitted aliens.

The habeas writ “lies to enforce the right of personal liberty; when that right is denied and a person confined, the federal court has the power to release him.” Fay v. Noia, 372 U.S. 391, 430-1, 83 S.Ct. 822, 844 5, 9 L.Ed.2d 837 (1963). Habeas relief is available to aliens detained on our shores after being found excludable by immigration authorities. Shaughnessy v. United States ex rel. Mezei, 345 U.S. 206, 213, 73 S.Ct. 625, 629, 97 L.Ed. 956 (1953) (conceding that an excluded alien housed on Ellis Island pending admission to another country “may by habeas corpus test the validity of his exclusion”); see also Rodriguez-Fernandez v. Wilkinson, 654 F.2d 1382, 1390 (10th Cir. 1981). Habeas jurisdiction is properly exercised to stay exclusion orders when the relief sought “inhere[s] in the question of custodial restraint upon liberty.” Pierre v. United States, 525 F.2d 933, 936 (5th Cir. 1976). Petitioners, clearly meeting the criteria of the habeas statute, cannot be denied review of the propriety of their detention on the basis of their immigration status.

Respondent’s argument confuses jurisdictional issues with those relevant to whether a complaint states a claim upon which relief can be granted. If the statutes and regulations relied upon by petitioners permit the actions complained of, the petition must be denied on the merits, not for want of jurisdiction. See Fogel v. Chestnutt, 668 F.2d 100 at 105-07 (2d Cir. 1981) (discussing the tendency to confuse jurisdictional and pleading requirements). The precise, restricted nature of jurisdictional concerns, see id., is illustrated by the cases cited by respondent. No explicit mention of the term “jurisdiction” can be found in any decision declining review of exclusion determinations. See, e.g., United States ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 543, 70 S.Ct. 309, 312, 94 L.Ed. 317 (1950); Leng May Ma v. Barber, 357 U.S. 185, 78 S.Ct. 1072, 2 L.Ed.2d 1246 (1958); Petition of Cahill, 447 F.2d 1343 (2d Cir. 1971). Habeas petitions in these cases were denied as “without merit.” Petition of Cahill, supra at 1344; see also Pierre v. United States, 547 F.2d 1281, 1290 (5th Cir.), vacated 434 U.S. 962, 98 S.Ct. 498, 54 L.Ed.2d 447 (1977), vacated and remanded, 570 F.2d 95 (5th Cir. 1978).

The arguments that parole denials are unreviewable, that aliens are unprotected by either the Constitution or the Protocol and that Sava’s conduct conformed with all applicable standards all go to the merits of this habeas action. To those contentions, the court must now turn its attention.

STANDARD OF REVIEW

The Attorney General may in his discretion parole into the United States “for emergent reasons or for reasons deemed strictly in the public interest” any alien applying for admission. 8 U.S.C. § 1182(d)(5)(A). This authority has been delegated by regulation to INS district directors. 8 C.F.R. § 212.5. The government contends that courts in the Second Circuit are without the power to review the discretionary denial of parole. This position derives from Petition of Cahill, supra, in which an alien was denied parole pending a five day adjournment of his exclusion hearing. 447 F.2d at 1343. Despite allegations that such denial was arbitrary and capricious, the court found review barred “as long as [the Attorney General] has exercised discretion under §§ 1182(d)(5) and (6) to deny parole.” Id. at 1344; see also Man Chung Lam v. Immigration and Naturalization Service, 79 Civ. 181 (S.D.N.Y.1979) (Lasker, J.) (no judicial power to review revocation of parole of alien physically in the country for thirteen years).

Nonreviewability of parole decisions stems from the long accepted doctrine that the determination to exclude aliens is the province of the political branch of the government. See United States ex rel. Knauff v. Shaughnessy, supra 338 U.S. at 543, 70 S.Ct. at 312. Since neither detention on United States territory nor parole alters an alien’s excludable status, Leng May Ma v. Barber, supra 357 U.S. at 188, 78 S.Ct. at 1074, it is not illogical to extend this judicial hands-off policy to parole decisions made in the course of proceedings testing admissability.

Parole adjudications, however, are not completely akin to final exclusion orders. Qn the most obvious level, the former merely determine the setting and character of an alien’s existence until such time as the latter permanent decisions can be reached. Judicial review of the parole process, therefore, does not impinge upon the political judgment to exclude or accept nor interfere with the executive and legislative power to control our borders. It only insures that parole status, which Congress has determined does not necessarily interfere with such control, is conferred by district directors within the bounds anticipated by the delegation of discretion. A determination which governs the treatment of aliens while they await the possibility of admission cannot be left completely to unelected officials. Congress’ unlimited authority to exclude does not necessarily imply the district directors’ absolute, unreviewable discretion to decide what aliens may and what aliens may not be accorded parole status.

Most of the decisions found establish that parole in the exclusion context will be reviewed, even in this Circuit, in the proper circumstances. Petition of Cahill itself implies a scope of review sufficient to insure that the statutory discretion was in fact exercised. 447 F.2d at 1344. In United States v. Murff, 260 F.2d 610 (2d Cir. 1958), the Court of Appeals took a more active role in dealing with a Hungarian refugee paroled into this country from Austria. Despite the broad discretion granted the Attorney General to revoke parole, 8 U.S.C. § 1182(d)(5), the court ruled that due process could not be satisfied without a hearing to “give assurance that the discretion of the Attorney General shall be exercised against a background of facts fairly contested in the open.” Id. at 615. In light of the unique facts underlying the alien’s parole, the court was willing to depart from a broad reading of the parole authority, id. at 613, even though its ruling would lead to the temporary, and perhaps permanent, admission of an alien found inadmissible in an exclusion proceeding. See id. at 612.

In Conceiro v. Marks, 360 F.Supp. 454 (S.D.N.Y.1973) (Wyatt, J.), habeas corpus relief was denied because the court could find no abuse of discretion in the district director’s denial of parole to an excludable political asylum applicant. Id. at 457. Abuse of discretion was defined as a decision made “without a rational explanation,” in marked, unexplained contradiction of established policy or resting “on an impermissible basis such as an invidious discrimination against a particular race or group.” Id.; see also Wong Wing Hang v. Immigration and Naturalization Service, 360 F.2d 715, 719 (2d Cir. 1966) (same standard applied to discretionary denial of application to suspend deportation); Soroa-Gonzales v. Civiletti, 515 F.Supp. 1049, 1058 (N.D.Ga.1981) (same standard applied in habeas proceeding challenging a district director’s revocation of parole); Pierre v. United States, supra, 547 F.2d at 1289. Similarly, in Massoud v. Attorney General of the United States, 459 F.Supp. 672, 677 (W.D.Mo.1978), respondent’s denial of advance parole was approved because “a rational basis existed . . . for the denial.”

These decisions indicate that review for abuse of parole discretion is the norm. While displaying total restraint on issues of exclusion, federal courts have examined the treatment of aliens awaiting that final order. The summary disposition in Man Chung Lam cannot support nonreviewability in light of this series of rulings.

The three-pronged abuse of discretion standard set forth in Wong Wing Hang, supra, is appropriate in the case at bar. Although Petition of Cahill argues for an inquiry limited to whether discretion was exercised, its facts posed no serious problem of discrimination and no special circumstances. The alien seeking parole merely had his hearing delayed five days on his own request and sought parole during that time. Where petitioners faced long-term detention, Soroa-Gonzales, supra, possible discrimination by national origin, Conceiro, supra, or other problems unique to their immigration status, Murff, supra, parole decisions were reviewed under the abuse of discretion standard.

These Haitian petitioners have been incarcerated in substandard facilities for approximately eight months, allegedly pursuant to a program applicable to them alone. Their claims merit use of the stricter analysis. Therefore, defendant Sava will be found to have abused his authority if he failed to exercise his discretion, did so without a rational explanation, inexplicably departed from prior practice or discriminated against Haitians qua Haitians.

FACTS

a) Parole policy. The indicia of abuse of discretion cannot be identified without a clear understanding of the manner in which such discretion is usually exercised. Since Sava is vested with full authority over parole decisions, his description of the relevant criteria is highly probative. Of course, Sava’s explanation must be tested against, and interpreted in light of, his parole determinations and any official policies which suggest a different formula. 3

We are told that parole applications are given individual attention and are resolved on a case-by-case basis. Sava affidavit at ¶ 6. In each case, the application is reviewed by an immigration case officer, a recommendation is made by that officer and final action is taken by Sava or a delegate. Id. Three major themes pervade Sava’s lengthy explanation of the factor analysis performed at the final action stage. Sava determines whether the applicant poses a risk to the community if released, id. at ¶ 7, is likely to abscond, or presents a particularly compelling case for release. Id.

The perceived threat that petitioners would not return for exclusion hearings if paroled is the focus of the instant action. Sava conceded that petitioners pose no threat to the community, Transcript (“Tr.”) at 458, 4 and concluded that none possess “humanitarian factors” sufficient to warrant parole. Id. at 406. 5 Finally, he stated that “getting the people to those hearings ... is really the judgment that is being made along the line.” Id. at 408. Of the many factors mentioned by Sava as relevant to the parole process, most advance the inquiry into an alien’s likelihood of absconding. 6 Id.

An alien’s documentation is important to estimate the absconding threat insofar as documented aliens have been through a screening process at a United States consular office abroad and have had their identities and histories examined and verified. Tr. at 402. Extensive screening is performed on aliens applying for both immigrant and non-immigrant visas, see 8 U.S.C. §§ 1201, 1202, but is not required of aliens being transported through the United States as transients without visa (“TWOV” or “TROV”). See 8 C.F.R. § 212.1(e). Thus, “documented” aliens have passports and immigrant or non-immigrant visas while “undocumented” aliens have no documents, false documents, passports without visas or passports with TROV status. See Tr. at 385 (describing as “undocumented” Afghans travelling with “a passport and no visa or some counterfeit documents”). 7 In some respects, Sava’s lack of confidence in the identities of undocumented aliens has led him to consider them greater absconding risks. See Tr. at 459.

Sava believes that a paroled alien’s future cooperation with INS is determined in part by the conditions of release. Release can be secured through the posting of a bond, an offer of sponsorship or on one’s own recognizance. Sava prefers the former approach because it gives greater “leverage” over the parolee. Tr. at 408. He also claims that relatives are preferred over organizations as sponsoring entities. Tr. at 407. The advantage that relatives enjoy over church and civic groups is difficult to explain or to credit. They do not differ in their ability to provide food, shelter and clothing, id., but, according to Sava, are more reliable in their “tracking” capabilities because organizations are responsible for many people. Id. This distinction demonstrates a misunderstanding or misrepresentation of organizational sponsors. Such sponsors, used for many years by INS to help settle refugees, see Pszyk deposition at 8, place each alien in a separate home environment much the same as direct placement with family. See Tr. at 72, 89 (testimony of Livingston Chrichlow); Pszyk deposition at 23-6 (large organization only does initial processing, then local churches place persons and arrange for their sustenance). Thus, rather than risking diffusion of responsibility and rendering tracking less likely, these organizations appear better structured than individual families to provide information needed by INS and to act as liaison between INS and the parolee. 8

Other factors relevant to parole decisions are an alien’s financial condition and job skills, prior immigration history, criminal record and the likelihood that his or her immigration applications will meet with success. Financial condition and past employment are relevant to the extent they vitiate the need for a sponsor to feed, clothe and house the alien. Tr. at 455. Prior immigration history is taken into account when the alien has made a prior entry and complied with all applicable immigration regulations. Id. at 459. A criminal record obviously reduces drastically the Director’s confidence in the alien’s future appearances and may indicate dangerousness. Sava’s review of political asylum documents is limited to determining whether they are frivolous, for inadequate or non-serious papers might indicate a state of mind conducive to absconding. Id. at 414-5.

The parties disagree as to how these various factors and themes coalesce as Sava’s parole policy. Petitioners view the record as indicating that “[pjhysical detention of aliens is . . . the exception, not the rule, and is employed only as to security risks or those likely to abscond.” Leng May Ma , supra 357 U.S. at 190, 78 S.Ct. at 1075. Respondent contends that policy never was formulated in those terms, 9 but puts forth no coherent alternative. Sava’s testimony implies that he doesn’t stray far from the Leng May Ma standard. He indicates more of a focus on the statutory language in his search for “emergent reasons” and compelling humanitarian factors favoring parole. He manifests, however, a willingness to grant parole in the absence of such public policy justifications if no threat to the community or risk of absconding would result. When an alien’s file is such that Sava does not fear these negative consequences, parole is granted.

b) Parole Determinations. In order to determine whether Sava employed the above standards abusively with respect to petitioners, his resolution of comparable parole requests must be studied. If those granted parole do not possess more of the determinative factors than petitioners, abuse of discretion might be implied.

Petitioners based their fact presentation on file sheets for 183 aliens who arrived in the New York district in 1981 and were subject to exclusion proceedings. They compared the treatment of those 183 aliens with that of the 86 Haitians who were transferred from Miami in July. Sava based his explanations of parole policy and past results on this same sample. However, both sides err in their reliance on .the 183 files.

Sava assumed his present position on July 6, 1981. Tr. at 378. Although he analyzed parole determinations made by his predecessors during 1981, he could testify from first-hand knowledge about only those applications on which he acted. Id. at 401. Post-hoc rationalizations of decisions to which he was not a party cannot be accepted as evidence of Sava’s parole policy or the even-handedness of its application. Only testimony as to the reasons for Sava’s parole choices can be considered “ ‘supplementary articulations of the reasoning behind the original decision’.” Massoud, supra at 675.

Ninety-nine of the 183 sheets constituting JE 68 represent aliens whose date of arrival or date of parole indicates that Sava was responsible for the final action taken. This group is still not fairly comparable to petitioners and the other detained Haitians because it includes nine individuals who never applied for political asylum and/or parole or who withdrew their applications and acceded to return to their country of origin. 10 Thus, Sava acted upon 90 parole requests other than those of petitioners and their peers.

All of the 90 non-Haitian parole applicants were successful in avoiding detention pending final resolution of their exclusion hearings. Yet, many were paroled for reasons clearly not applicable to petitioners. For example, 78 of the 90 excludable aliens had characteristics which qualified them as compelling cases according to Sava’s guidelines. Current State Department policy, as interpreted by Sava, foreclosed exclusion hearings for the 65 Afghans in the group for at least one year after their arrival. See JE 66; Tr. at 385-8. These aliens, facing indeterminate periods of detention if not paroled, were released. Tr. at 387 (Sava deemed their situation a “compelling emergency”). 11

Eight humanitarian parolees were aliens arriving in nuclear family groups that included minor children. INS does not detain children, Tr. at 391, and, given his reluctance to fracture families, Sava considers such units “compelling cases for parole.” Id. at 398. As cases involving emergent reasons for parole, these did not require the normal inquiry into security or risk of absconding. Thus Sava’s conduct with respect to these applications has limited relevance to the instant lawsuit.

By the same token, five Iraqi nationals are not comparable to petitioners. These 5 “had approved applications for service benefits pending" as a result of petitions by relatives in this country able to pass derivative immigration benefits to aliens. Tr. at 396-7. 12 ' Sava foresaw difficulties in sustaining the government’s burden of proof in exclusion hearings concerning these aliens, especially those who arrived from third countries. 13 Id. Both potential immigration benefits and anticipated delays in exclusion proceedings are compelling reasons for parole. Thus, these five aliens were not subject to the risk of absconding factor analysis.

Twelve of the 90 parolees were released pursuant to Sava’s factor analysis and, thus, are particularly instructive in examining petitioners’ claims. Of these, six were Iraqis, 14 three Iranians, one Dominican, one Lebanese and one Pole. The vital information from their file sheets will be summarized to aid the comparative analysis to follow.

None of the six Iraqis listed any employment status (with the exception of one student), none posted any bond and none were formally sponsored. All six did state specific United States addresses on their papers, and all had some relatives in this country. They were undocumented aliens, possessing valid passports and transients without visa status. 15 Only one had a prior immigration history, and that was without incident. Three of the six received negative State Department responses to their political asylum applications (after they were paroled), and the others were not responded to.

The status of the three Iranians is less clear from the file sheets. All had passports, and two had temporary non-immigrant visas while the third was travelling without a visa. Although none of the three formally requested parole, 16 it appears that all were released on their own recognizance within a day of their arrival. All three had close relatives in this country, none of whom had any permanent immigration status. Each claimed political asylum, but the files were silent as to the State Department determination on these claims. Two had no employment listed and no prior immigration history while the third was a business trader who had entered this country once before.

As for the remaining three aliens, the Dominican had valid non-immigrant documents, made applications for both asylum and parole, received no State Department reply, posted a $2,000 bond, had an uncle and aunt living in New York (United States citizens) and showed no employment or pri- or immigration experience. The Pole had similar documentation and pending applications, but had received a negative response from the State Department, posted only a $500 bond, had no relatives here or prior immigration status and was a motor mechanic. Finally, the Lebanese had fraudulent papers, both parole and political asylum applications, no State Department reply, a $1,000 bond, a brother with permanent status in New York, skill as a handbag maker and no prior entries.

Sava viewed petitioners as without sufficient ties to the community to assure their appearance at hearings, unlikely to succeed on their immigration applications, undocumented, lacking relatives able to pass benefits to them, illiterate, indigent, farmers without prior entries. Tr. at 406, 527. The immigration files reveal more individual detail. While all but one lived on a farm in Haiti, five claim to have been involved in some degree in a trade other than farming. 17 Only three are illiterate, the others having between three and eight years of schooling in Creole and some French. Six claim to have some relatives in this country — one has a citizen uncle in New York, two have refugee relatives at various locations known and unknown, two have relatives of unknown status and address and one has a paroled brother at a known address in New York. None have relatives able to petition for special status for the alien. Sava’s other descriptions seem to be uncontradicted by the files.

These descriptions make clear that, with respect to the factors identified by Sava as relevant to the parole inquiry, petitioners are indistinguishable from aliens paroled by Sava. Like eight of the twelve most similar applicants, petitioners are undocumented. As for the conditions of release, nine of the twelve were paroled on their own recognizance (without bond or formal sponsors), while only three posted bond. Petitioners all have prospective sponsors and none offered bond because Sava never indicated that bond was desired. 18 Unlike the Haitians who all listed job skills, only three of the twelve similar applicants had such qualities. None of these aliens, Haitian or non-Haitian, exhibited compelling reasons for parole — none travelled with minor children, none had family here able to confer immigration benefits on them 19 and none faced indeterminate detention due to foreign policy considerations. 20 The only constant differences in this group of 20 aliens are their race, their national origin and Sava’s action on their parole application.

DETERMINATION

I

As a preliminary matter, the court must decide whether Sava has indeed exercised his discretion in denying petitioners’ parole applications. See Petition of Cahill, supra at 1344. The record renders implausible Sava’s representation that the requests of the Haitian petitioners were considered on a case-by-case basis. Rather, the testimony and exhibits indicate that the Haitians at the SPC have always been treated as a group without individual characteristics.

Sava’s testimony was riddled with specific recollections of the particular circumstances of many of the aliens he paroled. He was able to recall the religious affiliation and travel history of the Iranians, the Iraqis’ country of origin and other facts about individuals and groups of aliens. Many of these facts were not recorded in the exhibits accepted as evidence.

Yet, Sava never spoke about the individual characteristics of any Haitian. It is true that some descriptions are valid for the entire group — undocumented and indigent, for example. A cursory review of their parole and asylum applications reveals, however, that petitioners are not uniform in other ways relevant to Sava’s decisional matrix. Petitioners vary markedly in their literacy levels, their job skills and their ties to this country through relatives. Sava apparently is ignorant of these differences in qualities admittedly relevant to a parole application and of which he had specific knowledge with respect to other, non-Haitian aliens. 21 This lack of familiarity with the particulars is especially puzzling given the extensiveness of petitioners’ applications. Since many parole requests are made orally, see n.16 supra, and most were granted within several days, Sava probably had more information about petitioners than any other alien with whom he dealt.

The Haitian rejection letters themselves indicate a failure to give attention to each parole request as a distinct entity. Each denial, five of which were issued on the same day, contained the same language concerning the insufficiency of the information presented in the parole requests. Both the wording and the timing of these responses indicate that denial was a foregone conclusion, not a decision reached through independent review and analysis of each file.

While the last six denials also cited State Department inaction or negative action as reasons for denial, inclusion of this rationale was disingenuous in two ways. First, the timing and content of Sava’s letters indicate that he delayed action until hearing from the State Department. 22 This makes a mockery of his claim that processing time is an important consideration in determining whether to release an alien, and contradicts his testimony as to how political asylum applications affect parole determinations. His review of such applications is limited to whether the alien’s claim is frivolous, but does not anticipate, let alone depend upon, the State Department’s final decision as to the legitimacy of the alien’s fear of persecution if returned to his or her home nation. Tr. at 413-5. 23 Moreover, Sava admitted that petitioners have submitted serious, non-skeletal asylum applications. Id. at 440. Finally, a review of the 12 parole grants given to non-“humanitarian” aliens reveals that seven re ceived no State Department recommendation and five failed to demonstrate a well-founded fear of persecution. See n.20 supra. In none of these 12 cases did Sava deny parole or put off decision pending State Department action, nor did he revoke the parole of the five aliens whose requests were denied soon after their release. Sava’s improper consideration of State Department recommendations in denying petitioner’s parole applications may, in and of itself, make the determinations defective. See Siang Ken Wang v. Immigration and Naturalization Service, 413 F.2d 286, 287 (9th Cir. 1969); Soroa-Gonzales, supra at 1060 (reliance on one impermissible factor refutes the contention that discretion was exercised rationally).

Assuming arguendo that Sava actually exercised his discretion, petitioners must meet an extremely heavy burden to show abuse thereof or arbitrary or capricious behavior by the Director. Soroa-Gonzales, supra at 1058. They must prove that the parole decisions were made without reasonable foundation, that they were “without a rational explanation, inexplicably departed from established policies, or rested on an impermissible basis such as an invidious discrimination against a particular race or group.” Id.

Is there a rational explanation for Sava’s parole decisions concerning the eight petitioners? A two-fold justification was offered for their continued detention. First, and foremost, Sava implied that the Haitians were a risk to abscond. He reached this conclusion by observing that none of the parole requests offered to post a bond, and all sought release to organizational sponsors rather than family or friends. Tr. at 406. Furthermore, he stated that no petitioner had documents, and claimed that none had any education or job skills. Second, no Haitian presented a compelling case for release. Sava anticipated no delays in getting State Department recommendations or final exclusion decisions, and found no relatives able to pass derivative benefits to petitioners. Id. at 405-6.

Given Sava’s overriding concern about “getting the people to those [exclusion] hearings,” Tr. at 408, the rationality of his explanation is questionable. To the extent that he developed legitimate fears of absconding by petitioners, those fears arose by analyzing their applications differently from those of non-Haitians. This indicates discriminatory decision-making.

Documentation and release conditions are the two main indicia of absconding relevant to the instant case. 24 Lack of documentation alone cannot justify a parole denial. Tr. at 459. Similarly, Sava did not deny any parole application because he “wasn’t satisfied with the sponsorship.” Tr. at 531. Nonetheless, he made clear his preferences for bond over sponsorship, see Tr. at 407, and family sponsors over organizations. Id. at 529-30. The validity of these opinions is questionable. See pp. 12-13 & n.8, supra.

Neither factor being independently sufficient to justify parole denial, Sava’s determination must have been based on a combination thereof. Nothing in the record shows that Sava was more than slightly uneasy about granting parole to non-Haitian aliens not manifesting the preferred documentation and release condition traits. He made no effort to explain how these two noncontrolling factors became controlling in combination, and no rational explanation is evident. It is particularly unclear how a visa or a $500 bond would significantly change petitioners’ likelihood of reappearing. INS, through petitioners’ own applications, has detailed information about each petitioner, making reliance on lack of documentation even less rational than Sava thought. See Tr. at 460 (Q: “Do you know whether your predecessor . . . concurred in your view that undocumented persons are absconding risks ... ? A: .... I didn’t know until you mentioned it that I had that view.”). Furthermore, after Sava admitted that he never studied the proposed sponsors’ experience with undocumented aliens, he did not attempt to explain how a $500 bond would assure the aliens’ attendance at hearings if their sponsors could not. Sava’s stated fears that petitioners were absconding risks were not based in reasons communicated to the court.

Examination of Sava’s other parole decisions supports this conclusion and shows that his treatment of Haitians inexplicably departed from his own established policies. The record discloses no formal family sponsorship for any of the 12 non-Haitians paroled by Sava. The exhibits merely document the extent of an alien’s family here. Furthermore, nine of the 12 were released on their own recognizance. Only the Dominican, the Pole and the Lebanese were forced to post bond as a condition of release. These bonds ranged from $500 to $2,000 and, in the case of the Polish alien, seemed to result from a process in which INS requested the bond and the alien negotiated the amount down from the sum demanded. See n.18 supra.

The documentation factor is equally revealing. Of the 12 aliens most like petitioners, seven were TRO Vs, one had falsified documents, and four had visas inappropriate for immigration. 25 Of the eight undocumented persons in this group, only the Lebanese with falsified papers posted a bond. 26 While the seven others were released on their own recognizance, the eight petitioners in the same situation were all denied parole in large part for their failure to offer bond money. This discrepancy indicates, at the least, a clear, unexplained departure from prior parole policy.

This data suggests that Sava’s actions were motivated by a desire to discriminate against Haitian aliens. Petitioners can invoke no fifth amendment protection against exclusion from the United States on the basis of their race or national origin. Rodriguez-Fernandez, supra at 1386; see generally, United States ex rel. Knauff v. Shaughnessy, supra 338 U.S. at 542-4, 70 S.Ct. at 312 -3. Even though Congress may employ race or national origin as criteria in determining which aliens to exclude from the country, a district director may not apply neutral regulations to discriminate on such grounds. Wong Wing Hang, supra at 719; see also Washington v. Davis, 426 U.S. 229, 241, 96 S.Ct. 2040, 2048, 48 L.Ed.2d 597 (1976) (“A statute, otherwise neutral on its face, must not be applied so as invidiously to discriminate on the basis of race”). Such invidious racial or national origin based discrimination constitutes abuse of discretion when insinuated into a neutral grant of decision-making authority. Wong Wing Hang, supra at 719.

The parole results during Sava’s tenure as New York District Director demonstrate a gross maldistribution of releases. Of 86 Haitians whose applications he received, Sava paroled five, all pregnant women whose condition was an obvious compelling factor. Of 91 non-Haitian applicants, 90 were released and the other returned to his country of origin after withdrawing the parole request. These numbers imply more than coincidence, especially since they are paralleled in the population of 12 applicants most like petitioners. A prima facie case of discrimination may be made out by a showing of highly disproportionate impact, see Washington v. Davis, supra 426 U.S. at 242, 96 S.Ct. at 2049, especially when a “clear pattern, unexplainable on grounds other than race, emerges” from actions taken pursuant to facially neutral legislation. Village of Arlington Heights v. Metropolitan Housing Development Corporation, 429 U.S. 252, 266, 97 S.Ct. 555, 564, 50 L.Ed.2d 450 (1977). Impact alone is not determinative, however, except in rare cases exhibiting stark patterns of unequal impact.

Impact could not be more disproportionate than that involved here. Sava does not parole Haitians unless pregnant, but paroles all non-Haitians who press their application. This statistical pattern is so extremé as to meet the strictest impact-based standards and render unnecessary inquiry into whether ’discriminatory intent motivated Sava’s actions. See Village of Arlington Heights, supra at 266 & n.13, 97 S.Ct. at 564 & n.13. The Supreme Court, in Village of Arlington Heights, set forth Yick Wo v. Hopkins, 118 U.S. 356, 6 S.Ct. 1064, 30 L.Ed. 220 (1886), and Gomillion v. Light-foot, 364 U.S. 339, 81 S.Ct. 125, 5 L.Ed.2d 110 (1960) as examples of cases in which impact alone is determinative. In the former, the Court deemed unconstitutional discriminatory application of a zoning ordinance which resulted in denials of laundry licenses to 200 Chinese aliens while permitting 80 non-Chinese to “carry on the same business under similar circumstances.” 118 U.S. at 374, 6 S.Ct. at 1073. Gomillion involved a legislative attempt to redefine local boundaries in such a way that all but five of 400 black voters would be ineligible to vote in the city while all whites would remain residents. 364 U.S. at 341, 81 S.Ct. at 127. Sava’s parole determinations fall into a pattern indistinguishable from these two cases.

Thus, the burden of proof rests with respondent to rebut the presumption of invalidity by showing that the results were reached through racially neutral criteria. Washington v. Davis, supra 426 U.S. at 241, 96 S.Ct. at 2048. For the reasons detailed above, Sava has not met this burden. His purported rational explanations could not justify all the parole decisions, nor overcome the strong inference that Haitians did not merit his usual pro-parole point of view.

Where impact alone is insufficient to justify a finding of discrimination, other evidence such as the historical background of the decision and departures from normal procedures must be examined. Village of Arlington Heights, supra 429 U.S. at 267, 97 S.Ct. at 564. History, both related to these petitioners and to INS treatment of other Haitians in recent years, is instructive on the existence of discriminatory intent. Sava’s handling of these parole applications evidenced callousness towards the claims. Petitioners’ requests, for the most part, went unanswered until the State Department issued asylum applications, a procedure not used for any other alien, not relevant to Sava’s stated decision-making formula and resulting in delays in responding unmatched during his period as Director. Tr. at 439-40. After such inordinate delay in responding, Sava issued denial letters in bulk, using near identical language and completely devoid of rational, legitimate bases for his action. While there is no persuasive evidence that past or present INS policies led to Sava’s abuse of his discretion with respect to these aliens, 27 it is not irrelevant that Sava was involved in the formulation of the “Haitian program” found unconstitutionally discriminatory insofar as it prejudged all Haitian political asylum claims. See Haitian Refugee Center v. Civiletti, 503 F.Supp. 442, 511, 513-5 (S.D.Fla.1980) (a Sava authored memo on means of deterring Haitian immigration set forth policies which indicated “a predetermination that none of the Haitians could deserve asylum”) (emphasis in original). Such a finding echoes the finding here that Sava prejudged Haitian parole applications by determining that none could meet his release criteria.

Along with this suspect history, it is also clear that Sava departed both procedurally and substantively from his normal treatment of parole requests. He altered the sequence of events leading up to disposition of parole applications by waiting for State Department action on petitioners’ political asylum claims. The failure to give the same individual attention to Haitians resulted, moreover, in his overlooking of factors favoring release which led to release in other cases. See Village of Arlington Heights, supra 429 U.S. at 267, 97 S.Ct. at 564. The most poignant example of such inconsistency is the near identity of pertinent characteristics among some of the petitioners and some of the paroled Iraqi and Iranian nationals. The conclusion, therefore, is inescapable that Sava denied parole to petitioners because they were black and/or because they were Haitians.

II

Petitioners also claim that Sava’s actions violated the Protocol, a treaty adopted by the United States effective November 1, 1968. 19 U.S.T. 6257. The Protocol provides in Article 1 that except for certain technical alterations:

The States Parties to the present Protocol undertake to apply articles 2 to 34 inclusive of the Convention [relating to the Status of Refugees] to refugees as hereinafter defined.

Its provisions are designed to aid those persons who “owing to well-founded fear of being persecuted for reasons of race, religion, nationality, [etc.] ... [are] outside the country of [their] nationality and [are] . .. owing to such fear, ... unwilling to return to it.” Article 1, A(2) of the Convention. Petitioners seek to vindicate rights allegedly granted them by Articles 3 and 31 of the Convention. Article 31 protects unlawful aliens from the imposition of penalties on account of their illegal presence and from unnecessary restrictions on their movements. Article 3 provides that all aspects of the treaty are to be applied without discrimination as to race or national origin. Respondent contends that the Protocol does not create a private right of action and, if it does, those rights are incorporated by existing statutes and regulations governing immigration matters.

“The history of the adoption of the Protocol by this country makes clear that all the individuals and institutions involved in that process had a continuing belief that the Convention would not alter or enlarge the effect of existing immigration laws, chiefly because it was felt that our immigration laws already embodied the principles of the Convention.” Ming v. Marks, 367 F.Supp. 673, 677 (S.D.N.Y.1973) (Carter, J.), aff’d, 505 F.2d 1170 (2d Cir. 1974), cert. denied, 421 U.S. 911, 95 S.Ct. 1564, 43 L.Ed.2d 776 (1975); see also Pierre v. United States, supra, 547 F.2d at, 1287-8 (discussing legislative history and reaching the same conclusion). 28 However, the Protocol and Convention have not been rendered meaningless by their infusion into pre-existing regulations. Several courts have construed vague or broad immigration statutes and regulations so as to incorporate rights declared in the treaty. See, e.g., Pierre v. United States, 525 F.2d 933, 935 (5th Cir. 1976) (statute providing for exclusion of certain aliens seeking to work in this country cannot apply to refugees because it would “render the Convention meaningless as a practical matter”); Sannon v. United States, 427 F.Supp. 1270, 1274-7 (S.D.Fla. 1977), vacated and remanded, 566 F.2d 104 (5th Cir. 1978) (regulation barring immigration judges from hearing asylum claims of excludable aliens invalid as inconsistent with the Protocol and the Immigration and Nationality Act). Thus, regulations promulgated pursuant to unspecific statutory provisions must be interpreted consistent with the Protocol as well as the overall intent of the Act. 29 The language of the Protocol can be viewed as creating rights separate from those embodied in the regulations. See, e.g., Kashani v. Immigration and Naturalization Service, 547 F.2d 376, 379 (7th Cir. 1977) (independent review of the two sources indicated that the Protocol and the Act required the same standard of proof of refugee status); Ming v. Marks, supra, 505 F.2d at 1172 (Article 31 of the Protocol directly benefits refugees).

An individual’s standing to invoke the Protocol has often depended upon his or her success in attaining refugee status. See, e.g., Pierre v. United States, supra, 547 F.2d at 1289; Sannon v. United States, supra at 1270. When parole could not be considered until INS and the State Department determined affirmatively an alien’s refugee status, failure to obtain such status foreclosed challenge of discretion regarding parole. Pierre, supra at 1289. 30 Parole may now be granted at any time after an alien applies for admission, see 8 U.S.C. § 1182(d)(5), so this procedural obstacle has been removed. “Refugee” is a self-imposed label, not one requiring the imprimatur of the INS. See U.S.C. §§ 1101(a)(42), 1157(c)(4) (Attorney General may remove one’s refugee status if it is determined that the provisions of § 1101(a)(42) are not met). Since parole is granted, in part, to avoid lengthy detention pending final exclusion determinations, and since Sava regularly paroles aliens before receiving the State Department’s recommendation concerning political asylum, it would be incongruous to hold petitioners powerless to invoke the Protocol in their challenge to parole denials.

To the extent that the Protocol vests petitioners with rights relevant to this case, such rights must emanate from specific, rather than general, treaty language. See Huynh Thi Anh v. Levi, 586 F.2d 625, 629 (6th Cir. 1978) (no right of action maintainable under Article 4 of the Protocol relating to religious freedoms). Petitioners may assert their rights to be free of penalties due to their illegal entry and unnecessary restrictions on their movements while here seeking asylum. These specific Article 31 rights are also violated, according to Article 3, if such penalties or restrictions were imposed based on race or national origin. All of these rights are guaranteed within the detention and parole regulations pertaining to excludable aliens. Therefore, if Sava has abused his discretion under these provisions by withholding from petitioners parole that would ordinarily be granted, he has imposed the type of hardships prohibited by the Protocol.

The Protocol may also establish independent rights for petitioners insofar as Sava’s parole policy departs from the Leng May Ma standard. By prohibiting all unnecessary restrictions on movement, the Protocol makes detention the “exception, not the rule, employed only as to security risks or those likely to abscond.” Id. 357 U.S. at 190, 78 S.Ct. at 1075. Review of Sava’s parole decisions has shown that detention was the rule for Haitians despite the absence of security risk and without serious, commonplace inquiry into likelihood of absconding. The Protocol’s emphasis on the necessity of confinement leaves no room for Sava to interpret the vague parole authority as placing the burden on aliens to prove the nonexistence of these dangers. Thus, the parole policy apparently reserved for Haitians might violate the Protocol even if applied in a race and national-origin blind, case-by-case manner. In any event, the Protocol was clearly violated by the abusive and discriminatory way in which these parole requests were handled.

RELIEF

The court having found that respondent Sava failed to exercise properly the discretion given him by the statutes and regulations concerning the parole of excludable aliens, and having discriminated against petitioners because of their race or national origin, the writ of habeas corpus shall issue under the following conditions. Petitioners shall be released on parole or reasonable conditions shall be set therefor 31 within 10 days from the date of this order unless respondent can show cause in writing, supported by affidavit, reasons for believing that any or all of them pose a risk of absconding. Such proof must contain individual appraisals of each petitioner and must respond to the considerations purportedly used by Sava to conclude that similar aliens did not pose sufficient risk to merit detention. An explanation will be deemed lacking if it does not discuss why bonds and/or sponsorships cannot overcome whatever risks are identified and provide INS with leverage equal to that which sufficed for non-Haitians heretofore paroled by Sava.

IT IS SO ORDERED.

1 . Notices of denial were sent to petitioners according to the following chronology; October 7, petitioner Vigile; October 7, petitioner Bertrand; December 1, petitioners Theodore, Mervil, Guerrier, Eril and Toussaint; and, December 4, petitioner Baptiste. Joint Exhibits (“JE”) 17, 19-25.

2 . At evidentiary hearings conducted on January 22 through 26, 1982, the parties stipulated to the admissibility of approximately 64 exhibits. These were accepted in evidence as JE 4 through 68 and will be referred to as such in this opinion.

3 . The need to compare policy pronouncements made from the witness stand with policy as inferred from practice is especially acute in the absence of any formal, written statements of Sava’s standards for parole. See Transcript (“Tr.”) at 480 1. Several written statements have been placed in the record, presumably to suggest what parole policy might have been at various points in time. For example, memoranda explaining detention policy to INS employees at John F. Kennedy Airport in New York indicate that on January 29, 1981, before Sava took charge of this district, excludable aliens were to be detained “only when it appear[ed] likely that they [would] abscond or ... would clearly represent a danger to public safety and/or security,” JE 7, while on November 2, 1981, all undocumented aliens were to be detained “unless release conditions [were] met as set forth by the District Director.” JE 63. On January 4, 1982, after the initiation of this lawsuit and after all the parole decisions studied herein were made, INS issued “Detention Policy Guidelines in Exclusion Cases” which announced that “the statutory rule is one of detention, and that the use of parole authority is an exception to that rule which should be carefully and narrowly exercised.” JE 65. Therefore, parole should be granted only to aliens with serious medical conditions (including pregnancy), those accompanied by minor children, those with parents, spouses, children or siblings who are United States citizens or lawful permanent aliens “who are eligible to file, and have filed, a visa petition on behalf of the detainee,” and “in cases where detention is impossible or impractical for the Service.” Id. As none of the Haitian petitioners and only one of the aliens studied infra applied for parole after November 2, and nothing in this entire case concerns decisions made after January 1982, these latter two statements are of limited relevance.

4 . Citations to Tr. refer, to the record of the evidentiary hearings held in January 1982.

5 . The factors recognized as “humanitarian” include travelling with minor children, presence of relatives in this country able and willing to confer immigration benefits on an alien and executive policy which will prevent processing of exclusion cases and cause possibly indeterminate periods of detention. See Tr. at 403, 405. No petitioner meets any of these tests.

6 . Some factors, defined in a broad sense, are relevant to more than one of Sava’s “themes.” Confusion is possible when factors are identified generally, such as “family units,” because such ambiguity muddies the clear division of themes. _ For example, an alien’s family is relevant to the absconding inquiry insofar as the family will provide sponsorship or food and shelter, but is a compelling factor for release if it means that infants are travelling with adult aliens or that the alien has family in this country from whom immigration benefits can be acquired. Similarly, an alien’s applications may indicate the projected length of the exclusion process and thus constitute a rationale for humanitarian release, or they may suggest, if in skeletal form, a serious danger of absconding.

7 . A review of the files, JE 67, 68, indicates that most Afghans were TROV’s and, therefore, that “a passport and no visa” must refer to such transients. In fact, at least two file sheets listed under “Visa” both “TROV” and “No visa.” JE 68, sheets 8, 16.

8 . It is also significant that Sava never investigated the organizational sponsors’ ability to maintain contact with aliens or their experience with tracking similar groups. Tr. at 468-9. Neither families nor organizations suffer any legal recourse if aliens they sponsor fail to appear for scheduled immigration proceedings. Id. at 97-8, 447.

9 . Respondent’s attack on the standard enunciated in Leng May Ma is twofold. First, the comments regarding detention and parole are shrugged aside as dicta. Next, respondent claims that the standard represents a misunderstanding of the INS and Attorney General’s reports from which it was taken. The report, in pertinent part, read as follows:

Detention and Parole of Applicants for Admission. Detentions of aliens were at the lowest figure in the history of the Service at the close of 1955. This was accomplished through a new detention policy . .. under which only those aliens likely to abscond and those whose release would be inimical to the national security are detained. Many aliens whose papers were not in order were previously detained .... Under the present policy, most aliens with purely technical difficulties are allowed to proceed to their destination under “parole.”

Within ten days of the change, the number of aliens in detention in New York City dropped to about 25, compared with a usual detention population of several hundred.... From the inception of the new program to the close of the fiscal year more than 200,000 aliens entered the United States through the, port of New York and 16 of these were detained.

As the fiscal year drew to a close the Service had put into effect, and found workable, a humane detention program while maintaining positive safeguards and security measures for protection of the Government and the public interest.

Annual Report of the Immigration and Naturalization Service at 6 (1955); see also Report of the Attorney General of the United States at 5-6 (1955). Respondent gives great weight to the last two sentences of the first paragraph and concludes that this “humane detention program” applied only to documented aliens with technical flaws in their papers. Such selective emphasis is undue. Both reports, when read in their entirety, present technical violators merely as one example of aliens no longer detained. The statements of policy are broad and general and not conducive to respondent’s strained construction. See Report of the Attorney General, supra at 5-6 (after discussing technical violators, noting that “[ajnother relaxation of the former stringent policy . .. was the decision to release those found to be unlawfully in the United States ... during the pendency of their deportation proceedings” unless they presented a danger to public safety). The numbers presented in the above extraction indicate that the policy could not apply in as limited a manner as respondent suggests as it is unlikely that only 16 of 200,000 aliens entering New York had non-technical problems with admissibility. Finally, both the substantive contention and the “dicta" argument are belied by the fact that several courts have accepted the Leng May Ma approach. Rodriguez-Fernandez v. Wilkinson, 654 F.2d 1382, 1389 (10th Cir. 1981) (noting that Justice Clark, author of Leng May Ma was Attorney General during the formation of the past and “presently effective immigration laws”); Soroa-Gonzales v. Civiletti, supra at 1060.

10 . Of these nine aliens, only one ever requested parole and two sought political asylum. Eight of the nine were excluded, including an individual who withdrew his asylum application, and the ninth, a Haitian requesting asylum but not parole, remains in custody at SPC. The one parole applicant withdrew his application for admission.

11 . Sava testified that family status and risk of absconding were also considered with respect to the Afghan nationals, Tr. at 385, 508-11, but since these factors apply to only some of the group, and their compelling nature applies to all, it is obvious that Sava would parole any Afghan regardless of these absconding factors.

12 . Sava claimed that the 11 Iraqis were “kind of unique in that they, for the most part, had approved applications for service benefits pending.” Tr. at 396. Only five of these aliens fit that description. To receive immigration benefits from relatives in this country, the family tie must be that of parent, spouse, child or sibling, the relative must be a citizen or lawful permanent resident and the relative must petition to pass on the benefit. See 8 U.S.C. §§ 1151(b), 1153(a)(1), (2), (4), (5); Tr. at 443-4 (aunts, uncles, cousins, and other extended family members cannot pass on immigration benefits). Thus, this humanitarian factor applies to only five of the eleven paroled Iraqis.

13 . The court cannot give much weight to these purported factors. The country of origin of an alien’s journey and the prospects for the government’s success in exclusion hearings were never before identified as factors relevant to the absconding or compelling reasons inquiry. Nor should they logically be so. Both are considerations relevant to the post-exclusion treatment of aliens. See generally Rodriguez-Fernandez, supra. Indeed, the difficulties of proof involved in returning an alien to a third country arise only when there is some official policy restraining return to the nation of citizenship. See Tr. at 198 (DiRaimondo explanation of reasons why Afghans are special). Finally, the file sheets indicate that only one Iraqi arrived from a third country, and he was a compelling case because of family here. JE 68, sheet 10.

14 . These six are the ones who did not fit the general description ascribed by Sava to all 11 Iraqis whom he paroled. They had no relatives in the United States able to confer derivative benefits on them, see n.12, supra, and did not exhibit the questionable factor of third country origin. See n.13, supra. Nor is there any indication that they arrived in family groups as Sava claimed. Tr. at 398. Family groups are irrelevant, in any event, except where they include minor children or lawful United States residents able to confer benefits. Neither of these necessary conditions were met by these six Iraqis. Thus, these six must be examined under the traditional indicia of absconding applied by Sava to all non-humanitarian aliens.

15 . Sava referred to all 11 Iraqis as “documented,” Tr. at 398, and transients without visa “documented to the extent that they had passports.” Id. at 396. Such characterizations are misleading and totally erroneous. As noted above, Afghans travelling TROV were officially considered “undocumented.” See supra at 8 & n.7. TROV’s do not go through a consular screening process, see 8 U.S.C. §§ 1201, 1202, so they are by definition not documented in any respect relevant to the amount of trust INS may have in their identity. See Tr. at 459. The most incredible aspect of Sava’s testimony on this matter is that transient without visa status is not available to citizens of Iraq. 8 C.F.R. § 212.1(e). In no sense can these Iraqis be deemed documented.

16 . The absence of parole request information on the file sheets in JE 68 may not indicate that no such request was made, but only that the request was oral and not recorded. See Tr. at 195. However, some aliens are released on their own recognizance without ever being detained. Id. at 195-6. DiRaimondo testified that the file suggests no detention for at least one of these Iranian aliens. Id. at 224-5. A review of the other two files reveals that file sheet number 101 was not detained and number 104 was in carrier custody (detained by the airline on which he arrived) for one day.

17 . Petitioner Guerrier was a cook/waiter in Haiti. The other petitioners were all involved in farming to some degree, but Eril, Mervil and Theodore were apprentice tailors, Vigile was an apprentice baker and Baptiste was a part-time jewelry maker. See Government Exhibits 76-85 (petitioners’ applications for political asylum).

18 . At a meeting on August 31, 1981, Sava discussed with representatives of the Haitians at the SPC their desires to have those persons released. Sava stated that he would consider affidavits in support of parole requests without the formality of political asylum applications. Tr. at 479. While Sava told those attending that release would be considered upon submission of a parole request, an informal asylum application and a sponsorship offer, id. at 480, he did not mention that release would not be considered without offers of bond. In fact, the “subject of bond never came up at the meeting.” Id. The record implies that the subject of bond is often opened by the District Director, see JE 68, file sheet 122 (indicating that the alien twice negotiated reductions in the bond sought by INS), and that many aliens are paroled on their own recognizance. Therefore, the absence of bond offers will not be taken as prejudicing petitioners’ claims or distinguishing them from the 12 most similar aliens discussed above.

19 . Some, both Haitian and non-Haitian, had family in this country unable to pass on such benefits. For example, Iranian number 101 had a mother and siblings here who were also asylum applicants and all six Iraqis had uncles and/or cousins with permanent resident status. JE 68, file sheet numbers 47, 138, 140, 141, 144 (brother with unknown immigration status), 145. Petitioners’ files demonstrate that Eril has a citizen uncle living in Brooklyn, New York, Baptiste has a paroled brother here and refugee cousins in Florida, and all but two of the others have siblings and/or uncles and cousins in some status (refugee or unknown) in the United States. Thus, six of them are indistinguishable from Iranian number 101 and Iraqi number 144, while Eril and Baptiste share stronger United States ties with others among the released non-Haitians.

20 . The government presented no proof of Afghan-type policies applicable to any of the other groups represented in the sample of 12. The State Department recommended against asylum with respect to five of the 12, finding that none had well-founded fears of persecution in their native countries. These responses came between one and five weeks after Sava paroled the aliens. The remaining seven had not received official recommendations at the time JE 68 was compiled.

21 . The blatant inadequacy of Sava’s review of petitioners’ detailed submissions is represented by the following colloquy:

THE COURT: Just generally, what kinds of things do you know about [petitioners]? Do you know about their education, things like that about them?

Sava: Well to the extent it is documented, I know that for the most part they are illiterate—

THE COURT: How do you know that?

Sava: I think most of them in their affidavits make that ... admission.

THE COURT: Do you know about their employment skills?

Sava: Most of them were farmers, working land in Haiti.

Tr. at 527. These “facts” supposedly were learned from petitioners own affidavits. Id. at 527-8. But those affidavits make clear that six of the petitioners had, or were in the process of acquiring, skills other than farming, see n.17 supra, and that only three admitted to being illiterate, the other five having between three and eight years formal education and literacy in Creole and/or French. See Government Exhibits 76 -85. Sava’s casual, inaccurate generalizations rebut his claim to have conducted individualized review.'

22 . Sava delayed these parole decisions longer than was his normal practice, see Tr. 439, and made specific reference in six of the denials to the State Department conclusion or the lack thereof necessitating keeping the application pending. See p. 1 & n.l, supra.

23 . In fact, the State Department recommendation is sent to Sava as District Director merely due to a procedural quirk. The recommendation is for the attention and use of the immigration court, not the District Director, and the latter serves only as a conduit in the bureaucratic scheme. Tr. at 451. This may help to explain why Sava’s past parole decisions were made before any action was taken by the State Department and why negative executive recommendations do not usually cause Sava to revoke parole. Tr. at 450 2.

24 . It has been demonstrated, either through Sava’s testimony or despite it, that the other absconding factors are not involved. Sava noted that petitioners’ asylum applications were not skeletal so as to be disregarded and looked upon with suspicion. Tr. at 440. Similarly, financial condition is not important due to the presence of prospective sponsors. And, none of the petitioners has a criminal record or prior immigration history.

25 . These four, file sheet numbers 78, 100, 104 and 122, possessed facially valid documents entitling them to entry into this country for a limited time or purpose. Tr. at 430. They were excludable aliens because the immigration officer who processed their cases concluded that the documents misrepresented the aliens’ intent, which was to settle here. Id. at 430-1. Thus, they each committed a fraud, either when they obtained the documents from a United States consul abroad or when they presented those documents upon entry to this country. Id. at 433. However, they are considered documented insofar as they were processed by a consular office.

26 . This Lebanese national was also the only one of the entire group of 12 who had a relative in this country able to pass on immigration benefits. It appears, however, that no petition to do so was filed. Thus, this alien belongs in a group of non-compelling cases.

27 . Petitioners attempted to show, through a speech by the Attorney General to a Congressional committee, JE 16, some newspaper reports concerning the administration’s policy towards Haiti, and some specific conduct of INS trial attorney Michael DiRaimondo, that there exists a national policy of detention for Haitians which applies to no other ethnic group. The court is not persuaded by the scanty evidence presented on this matter.

28 . In discussing petitioners’ rights under the Protocol the court will not engage in a discussion of whether the treaty is self-executing or executory. “Treaties effect the municipal law of the United States only when [they] are given effect by congressional legislation or are, by their nature, self-executing.” United States v. Postal, 589 F.2d 862, 875 (5th Cir.), cert. denied, 444 U.S. 832, 100 S.Ct. 61, 62 L.Ed.2d 40 (1979). As noted, several courts have found that Congress intended the Protocol to be given effect by pre-existing statutes concerning immigration. Thus, there is no need to engage in this most difficult task of treaty interpretation. Id. at 876. To the extent they are not inconsistent, both the treaty and the Immigration and Nationality Act determine the propriety of immigration regulations and their performance.

29 . See n.28 supra. “[T]he legislative history of the act reveals that Congress intended to rewrite the Immigration laws in a spirit of compassion and humanity toward excludable and deportable aliens alike.” Sannon, supra at 1275. It is in this spirit that the court has viewed Sava’s exercise of his parole authority and found it wanting.

30 . An alien could, of course, invoke the Protocol to challenge the procedures through which refugee status was distributed. See Sannon, supra at 1274 5.

31 . Such conditions may include sponsorships, reasonable bonds or some reasonable combination of these and other conditions sufficient to ensure INS of the degree of leverage typically required in similar cases.