Appellant, Vytautas Gecas, is a Lithuanian national who has lived in the United States as a resident alien for thirty-four years. The United States subpoenaed oral and written testimony from Gecas concerning allegations that he participated in the persecution of persons because of their race, religion, or political opinion during World War II. Such conduct would render Gecas deportable. See 8 U.S.C. § 1251(a)(4)(D) (1994). Gecas refused to testify and instead invoked the Fifth Amendment's privilege against self-incrimination. See U.S. Const. amend. V, cl. 2.

The United States petitioned the district court for an order to enforce its subpoena, and the district court granted the Government's petition. See United States v. Gecas, 830 F.Supp. 1403, 1423 (N.D.Fla.1993). Gecas appealed, claiming that his silence is protected by the constitutional privilege against self-incrimination. A divided three-judge panel reversed the district court's order, see United States v. Gecas, 50 F.3d 1549, 1567 (11th Cir.1995), and held that Gecas was entitled to the equitable relief provided by the privilege because he had a real and substantial fear of conviction under foreign law.1 We granted rehearing en banc,2 and now affirm the district court's order compelling Gecas to testify.

We examine the underlying facts in part I. We determine in part II that the conduct with which Gecas is charged could form the basis of a criminal conviction under foreign law. We also find that Gecas faces a real and substantial risk of actually being convicted under foreign law based on the alleged conduct. Therefore, in part III we turn to the issue of whether a real and substantial fear of foreign conviction is a valid basis for the invocation of the privilege against self-incrimination.

We start with the language of the Fifth Amendment's Self-Incrimination Clause and find it ambiguous. We then turn to precedent, and discover that the Clause, as currently interpreted, does not apply to Gecas. Because Gecas argues that the original intent of the Self-Incrimination Clause authorizes his invocation of the privilege, we discuss in part IV the history of the privilege against self-incrimination. We conclude in part V that the privilege against self-incrimination was not intended to apply here.

According to Gecas' immigration file, he was born on September 25, 1922, in Naumiestis, a town in the Taurage district of Lithuania. On August 2, 1962, Gecas applied at the American Consulate in Liverpool, England, for a visa to enter the United States. Gecas stated in his application that from 1938 to 1944 he lived in Naumiestis as a "pupil," and that from 1944 to 1947, he lived in Lubeck, Germany, in a camp for "displaced persons"--a refugee camp. The application also stated that he moved to England in 1947, where he obtained employment first as a farm worker and then as a miner.

Gecas obtained a four-month visa from the Consulate on August 2, 1962. On October 23, 1962, at the age of thirty-nine, Gecas arrived in New York City on the Queen Elizabeth and was admitted to the United States. Upon entry, Gecas apparently went to Chicago, Illinois, to live with his cousin, a naturalized American citizen. In 1964, Gecas moved to a separate residence in Chicago, where he lived until October 1989. Gecas then moved to Sunny Hills, Florida, where he has since resided.

On July 25, 1991, the Office of Special Investigations (the "OSI")3 of the United States Department of Justice issued to Gecas an administrative subpoena.4 See generally 8 U.S.C. § 1225(a) (1994) (authorizing the Attorney General to subpoena the testimony of witnesses and the production of documents "relating to the privilege of any person to ... reside in ... the United States"). The subpoena directed Gecas to testify "relating to [his] residence and activities in Europe (including but not limited to the years 1940-1945)" and to produce documents and photographs "which concern [his] date and place of birth, [his] whereabouts and activities in Europe (including but not limited to 1940-1945), [and his] immigration to and residence in the United States." The OSI claimed to have evidence that Gecas served in the 12th Lithuanian Schutzmannschaft Battalion between 1940 and 1945. According to the OSI, this armed police unit participated in the persecution of persons because of their race, religion, national origin, or political opinion under the direction of the Nazi forces then occupying Lithuania. If Gecas engaged in this prohibited conduct, he would become deportable under 8 U.S.C. § 1251(a)(4)(D).

A United States Marshal served the OSI's subpoena on Gecas on July 30, 1991. On September 12, 1991, OSI investigators interviewed Gecas in the presence of his attorney. Gecas was placed under oath, and provided his name, his current address in Florida, and his previous address in Illinois. The OSI then asked Gecas some 162 other questions. These questions generally concerned Gecas' biographical data, such as alleged alternate spellings of his name, and Gecas' activities during World War II, such as his alleged participation in wartime atrocities against Jewish persons. To each question, Gecas responded as follows: "I decline to answer on the ground that the answer might tend to incriminate me." Gecas produced no documents or photographs other than his alien registration card.

On February 19, 1992, the United States petitioned the district court for an order to enforce the OSI's administrative subpoena. On August 11, 1993, the district court granted the Government's petition. See Gecas, 830 F.Supp. at 1423. The district court found that Gecas faced a real and substantial fear of criminal conviction in Lithuania, Germany, or Israel. See id. at 1411. Each of these states, the court noted, had enacted a law under which Gecas could be convicted. The court found that the OSI had previously provided information to foreign prosecutors of deported war criminals. Therefore, the court concluded that Gecas faced a real and substantial fear of foreign conviction.

Despite the legitimacy of Gecas' fear, however, the district court granted the Government's petition. The district court held that the privilege against self-incrimination contained in the Fifth Amendment to the United States Constitution does not apply to foreign prosecutions. See id. at 1423. The district court examined at length Supreme Court precedent and concluded that the privilege is not a "personal 'right' conferred upon persons within the protection of American law." Id. at 1421. Rather, according to the district court, the privilege was designed to protect the individual from prosecutorial overreaching and inhumane treatment at the hands of government. See id. at 1421-22. The court reasoned that neither purpose would be served by allowing the invocation of the privilege here because "[t]he privilege is a limitation only upon the power of the United States and its component states," not upon foreign governments. Id. at 1422. The court also noted that extending the privilege to foreign prosecutions would severely weaken domestic law enforcement. Therefore, the district court granted the Government's petition to enforce the OSI's administrative subpoena.

Gecas appeals, arguing that the privilege against self-incrimination is a fundamental right of the individual rather than a mere limitation on the activities of government. The United States cross-appeals the district court's holding that Gecas' fear of foreign prosecution was substantial enough to invoke the equitable relief provided by the privilege.

The Self-Incrimination Clause of the Fifth Amendment to the United States Constitution provides that "[n]o person ... shall be compelled in any criminal case to be a witness against himself." U.S. Const. amend. V, cl. 2. To assert this privilege against self-incrimination, as an initial matter, a witness' fear of conviction on the basis of his testimony must be reasonable, real, and appreciable. See Brown v. Walker, 161 U.S. 591, 599, 16 S.Ct. 644, 648, 40 L.Ed. 819 (1896). The witness must face a "real danger" of conviction to invoke the privilege because the privilege does not protect against "remote and speculative possibilities." Zicarelli v. New Jersey State Comm'n of Investigation, 406 U.S. 472, 478, 92 S.Ct. 1670, 1675, 32 L.Ed.2d 234 (1972). If the privilege applies to foreign proceedings, Gecas must demonstrate, first, that the information to be disclosed through his testimony would incriminate him under foreign law and, second, that his fear of foreign conviction is real and substantial rather than merely speculative. See In re Application of the President's Comm'n on Organized Crime, 763 F.2d 1191, 1198 (11th Cir.1985) [hereinafter In re President's Comm'n ].

The district court held that the information sought from Gecas would incriminate him under the laws of Israel, Germany, and Lithuania. See Gecas, 830 F.Supp. at 1408. The district court's interpretation of Israeli, German, and Lithuanian law is a question of law subject to review de novo. See Fed.R.Civ.P. 44.1. Israel punishes Nazi war criminals under the Nazis and Nazi Collaborators (Punishment) Law, 1950, S.H. 57.5 This act purports to apply extraterritorially, see Nazis and Nazi Collaborators (Punishment) Law, 1950, S.H. 57(2), and, in fact, it has been imposed for war crimes committed outside Israel. See generally Yoram Kessel, Israel Split on Extradition of Ex-Nazis from U.S., Wall Street Journal, June 2, 1983, at 34 (describing the imposition of this law on Adolph Eichmann "for his major part in implementing the Nazis' 'final solution' against European Jewry"). The questions posed by the OSI could lead to answers that would incriminate Gecas under the Israeli law. For example, the OSI asked Gecas, "[w]ere you involved in any operations [in a Minsk concentration camp] where pits or ditches were excavated?" and "[w]ere you involved in any executions [at the camp]?" Affirmative responses to these questions could certainly lead to Gecas' prosecution and conviction under the Nazis and Nazi Collaborators (Punishment) Law.

Germany does not specifically punish crimes against Jewish persons, crimes against humanity, or war crimes. Germany instead prosecutes Nazi war criminals under its murder statute, § 211, Nr. 1 StGB.6 Unlike the Israeli statute, section 211 on its face does not apply extraterritorially, but Germany has apparently used the law to prosecute former German citizens who committed war crimes outside the territorial boundaries of Germany. See, e.g., Douglas Martin, Canada Orders Extradition of Accused Nazi, N.Y. Times, Nov. 5, 1982, at A3 (describing the extradition of a former Gestapo officer to Germany to stand trial for the murder of 11,584 Jews in Lithuania). Gecas, however, is not a former German citizen, and German law appears unsettled as to whether section 211 applies to non-Germans who commit murder outside Germany. See generally Letter from Jurgen Schmude, Minister of Justice, Federal Republic of Germany, to William F. Smith, United States Attorney General 1 (Feb. 12, 1982) ("Just as in American law 'jurisdiction' basically exists only for criminal acts perpetrated in one's own territory, so there is also no German jurisdiction applicable to those punishable acts, under consideration here [war crimes], committed by foreign nationals in occupied territory without the addition of further exceptional circumstances." (emphasis added)). The issue is close, but we find that the information sought from Gecas could potentially incriminate him under section 211.

According to Gecas, Lithuania punishes Nazi war crimes under the Law Concerning Responsibility for Genocide of the People of Lithuania, Art. 1 (1992) [hereinafter Lithuanian Genocide Law].7 This law applies to "[t]he murder or torture of people in Lithuania [or] the deportation of her people ... during the period of occupation by Nazi Germany." Lithuanian Genocide Law, Art. 2 (1992). We hold that affirmative responses to the OSI's questions would incriminate Gecas under the Lithuanian Genocide Law. We therefore affirm the district court's holdings that the information sought by the OSI through their questions would incriminate Gecas under the laws of Israel, Germany, and Lithuania.

The district court also found that Gecas faced a real and substantial likelihood of prosecution and conviction in Israel, Germany, and Lithuania. See Gecas, 830 F.Supp. at 1411. The district court's determination of the likelihood of conviction is a question of fact subject to review for clear error. See Fed.R.Civ.P. 52(a). We have identified several factors which inform our inquiry into the probability of foreign conviction: whether there is a likelihood that his testimony would be disclosed to a foreign government, whether there is an existing or potential foreign prosecution of the claimant, and whether any of the charges would entitle the foreign jurisdiction to have him extradited. See In re President's Comm'n, 763 F.2d at 1198. The circuit courts that have addressed this issue, including this court, have narrowly construed these factors and have rarely found that a witness faces a real and substantial danger of foreign conviction. See, e.g., id. at 1199; Environmental Tectonics v. W.S. Kirkpatrick, Inc., 847 F.2d 1052, 1064-65 (3rd Cir.1988), aff'd on other grounds, 493 U.S. 400, 110 S.Ct. 701, 107 L.Ed.2d 816 (1990); In re Gilboe, 699 F.2d 71, 76 (2nd Cir.1983); In re Baird, 668 F.2d 432, 433 (8th Cir.), cert. denied, 456 U.S. 982, 102 S.Ct. 2255, 72 L.Ed.2d 860 (1982); In re Tierney, 465 F.2d 806, 811 (5th Cir.1972), cert. denied, 410 U.S. 914, 93 S.Ct. 959, 35 L.Ed.2d 276 (1973).8 Because these courts found that the witness faced no meaningful danger of foreign conviction, they did not reach the constitutional question we decide today.

Nevertheless, few witnesses who invoke the privilege based on a fear of foreign conviction face imminent expulsion of the sort anticipated by Gecas. The OSI exists to expel war criminals from the United States, see 28 C.F.R. § 0.55(f) (1995), and the OSI has been successful in its mission, see, e.g., Kalejs v. INS, 10 F.3d 441, 448 (7th Cir.1993) (affirming the deportation order sought by the OSI against a Latvian war criminal), cert. denied, 510 U.S. 1196, 114 S.Ct. 1305, 127 L.Ed.2d 656 (1994). As part of its mandate, the OSI is required to "[m]aintain liaison[s] with foreign prosecution, investigation and intelligence offices." Id. The OSI maintains liaisons with foreign prosecutors to cooperate in the indictment of expelled war criminals. See generally U.S. Gives Nazi Data to Israelis, Wash. Post, May 16, 1984, at A10 (describing the ceremonial transfer of information on the "investigation, interrogation and completed or pending prosecution of war criminals found living in the United States"). Gecas faces a real danger that the information obtained through his interrogation would be transferred to any country interested in obtaining his conviction.9

Once that information is transferred, Gecas faces potential conviction in Israel, Germany, and quite possibly Lithuania.10 As discussed above, both Israel and Germany have prosecuted alleged war criminals like Gecas. Gecas' ability to select his destination upon deportation offers little protection. Under 8 U.S.C. § 1253(a), the Attorney General may nullify Gecas' selection if transporting Gecas to that country "would be prejudicial to the interests of the United States." 8 U.S.C. § 1253(a) (1994). Accordingly, there is no guarantee that Gecas would be deported to the country of his choice.

Even if Gecas could avoid deportation, he could be extradited under the treaties the United States maintains with Israel, see Convention on Extradition, Dec. 10, 1962, U.S.-Isr., art. 1, 14 U.S.T. 1707, 1707, Germany, see Treaty Concerning Extradition, June 20, 1978, U.S.-F.R.G., art. 1(2)(a), 32 U.S.T. 1485, 1485, and Lithuania, see Treaty on the Extradition of Fugitives from Justice, Apr. 9, 1924, U.S.-Lith., art. 1, 43 Stat. 1835, 1835. Foreign governments have pursued the extradition of those who murdered civilians at the direction of the Germans during World War II. See, e.g., Demjanjuk v. Petrovsky, 10 F.3d 338, 340 (6th Cir.1993) (noting that Israel sought and obtained the extradition of an alleged Nazi prison guard), cert. denied, 513 U.S. 914, 115 S.Ct. 295, 130 L.Ed.2d 205 (1994). If Gecas' compelled testimony substantiates the OSI's allegations, it is distinctly possible that Gecas would face extradition for prosecution to Israel, Germany, or Lithuania. Therefore, we hold that the district court did not err when it concluded Gecas faces a real, substantial, reasonable, and appreciable fear of foreign conviction.

We now consider whether a real and substantial fear of foreign conviction is a valid basis for the invocation of the privilege against self-incrimination. The Fourth and Tenth Circuits have refused to extend the privilege to prevent this alleged harm. See United States v. (Under Seal), 794 F.2d 920, 926 (4th Cir.), cert. denied sub nom. Araneta v. United States, 479 U.S. 924, 107 S.Ct. 331, 93 L.Ed.2d 303 (1986) [hereinafter Araneta ]; In Re Parker, 411 F.2d 1067, 1070 (10th Cir.1969), vacated as moot sub nom. Parker v. United States, 397 U.S. 96, 90 S.Ct. 819, 25 L.Ed.2d 81 (1970). The North Dakota Supreme Court has similarly refused to expand the privilege's scope. See Phoenix Assurance Co. of Canada v. Runck, 317 N.W.2d 402, 413 (N.D.), cert. denied, 459 U.S. 862, 103 S.Ct. 137, 74 L.Ed.2d 117 (1982).

However, several district courts, including one district court in this circuit, have held that the privilege may apply in this situation. See Yves Farms, Inc. v. Rickett, 659 F.Supp. 932, 940 (M.D.Ga.1987); United States v. Lileikis, 899 F.Supp. 802, 809 (D.Mass.1995); Moses v. Allard, 779 F.Supp. 857, 882 (E.D.Mich.1991); Mishima v. United States, 507 F.Supp. 131, 135 (D.Alaska 1981); United States v. Trucis, 89 F.R.D. 671, 673 (E.D.Pa.1981); In re Cardassi, 351 F.Supp. 1080, 1086 (D.Conn.1972). In addition, the Second Circuit recently held that "the Fifth Amendment privilege against self-incrimination may be invoked by a witness who possesses a real and substantial fear of foreign prosecution." United States v. Balsys, 119 F.3d 122 (2nd Cir.1997). Because neither the Supreme Court nor the Eleventh Circuit has decided the issue, however, we treat it as a matter of first impression.

The plain language of the Self-Incrimination Clause provides no conclusive answer. The Clause protects a "person" from being "compelled in any criminal case to be a witness against himself." U.S. Const. amend. V, cl. 2. There is no question that resident aliens such as Gecas are "persons" within the meaning of the Fifth Amendment. See Kwong Hai Chew v. Colding, 344 U.S. 590, 596, 73 S.Ct. 472, 477, 97 L.Ed. 576 (1953). The parties do not dispute that the Government seeks to "compel" testimony from Gecas that will make him a "witness against himself." The question is, therefore, whether the current proceeding is a "criminal case." This term has been interpreted ambiguously.

A deportation hearing itself is a civil proceeding, not a criminal proceeding. See I.N.S. v. Lopez-Mendoza, 468 U.S. 1032, 1038-39, 104 S.Ct. 3479, 3483, 82 L.Ed.2d 778 (1984); Harisiades v. Shaughnessy, 342 U.S. 580, 594, 72 S.Ct. 512, 521, 96 L.Ed. 586 (1952); Fong Yue Ting v. United States, 149 U.S. 698, 730, 13 S.Ct. 1016, 1028-29, 37 L.Ed. 905 (1893). For this reason, the Government need not give to deportees all of the constitutional protections guaranteed to criminal defendants. See Lopez-Mendoza, 468 U.S. at 1038-39, 104 S.Ct. at 3483; see, e.g., United States ex rel. Bilokumsky v. Tod, 263 U.S. 149, 154-55, 44 S.Ct. 54, 56, 68 L.Ed. 221 (1923) (holding that a deportee's silence can be held against him where he refuses to testify concerning an issue that would not incriminate him under American law). Under a literal reading of the Self-Incrimination Clause, then, Gecas could be compelled to testify against himself because the deportation proceeding is not a "criminal case." In fact, the Supreme Court has never extended the Self-Incrimination Clause to deportation proceedings.

The Supreme Court, however, has held that a witness facing a legitimate possibility of conviction (in either the current or a subsequent proceeding) may invoke the privilege against self-incrimination "in any proceeding, civil or criminal, administrative or judicial, investigatory or adjudicatory." Kastigar v. United States, 406 U.S. 441, 444, 92 S.Ct. 1653, 1656, 32 L.Ed.2d 212 (1972). Thus, for example, the Court has held that a bankruptcy proceeding becomes a "criminal case," for purposes of the Self-Incrimination Clause, as long as the testimony sought from the witness "might tend to subject to criminal responsibility him who gives it." McCarthy v. Arndstein, 266 U.S. 34, 40, 45 S.Ct. 16, 17, 69 L.Ed. 158 (1924). This language suggests that, regardless of the noncriminal character of a deportation proceeding, it might become a "criminal case" under certain circumstances. We therefore turn to precedent to determine whether the current deportation proceeding involving Gecas is a "criminal case" within the meaning of the Fifth Amendment's Self-Incrimination Clause.

The Self-Incrimination Clause protects a witness against the infliction of criminal penalties on the basis of compelled, testimonial self-incrimination. See Kastigar, 406 U.S. at 453, 92 S.Ct. at 1661.11 The Clause does not protect a witness against the disclosure of facts which might "disgrace him or bring him into disrepute." Brown v. Walker, 161 U.S. 591, 598, 16 S.Ct. 644, 647, 40 L.Ed. 819 (1896). The Clause does not shield a witness from the "loss of [his] job, expulsion from labor unions ... [or] general public opprobrium" stemming from his incriminating testimony. Ullmann v. United States, 350 U.S. 422, 430, 76 S.Ct. 497, 502, 100 L.Ed. 511 (1956). The Clause does not even prevent the prosecution of a witness based on coerced admissions. See Kastigar, 406 U.S. at 453, 92 S.Ct. at 1661. The Clause only protects against the actual "infliction of criminal penalties on the witness"--a criminal conviction--based on self-incriminating testimony. Id. For this reason, the Supreme Court has stated that a violation of the Self-Incrimination Clause does not necessarily occur at the time the self-incriminating testimony is compelled. The actual violation, if any, occurs only at a witness's own criminal trial. See United States v. Verdugo-Urquidez, 494 U.S. 259, 264, 110 S.Ct. 1056, 1060, 108 L.Ed.2d 222 (1990).

Courts can prevent the occurrence of this harm in one of two ways. First and most basically, courts presiding over a criminal trial must exclude from evidence compelled, testimonial self-incrimination and its fruits. See Murphy v. Waterfront Comm'n of New York Harbor, 378 U.S. 52, 79, 84 S.Ct. 1594, 1609, 12 L.Ed.2d 678 (1964). For example, if overzealous law-enforcement officers unfairly coerce a damaging admission from the defendant, the court must exclude from evidence that admission and its fruits. See Bram v. United States, 168 U.S. 532, 542, 18 S.Ct. 183, 187, 42 L.Ed. 568 (1897). Similarly, if another court compelled the defendant's testimony12 at a prior proceeding--pursuant to a grant of immunity from prosecution, for example--a court presiding over the defendant's subsequent trial must also exclude from evidence that testimony and its fruits. See Murphy, 378 U.S. at 79 n. 18, 84 S.Ct. at 1610 n. 18. In either situation, a trial court must force the Government to "launder" its case and purge the taint of compulsion from its evidence. Forcing the Government to launder its case at a special hearing complies with the Self-Incrimination Clause by ensuring that the compelled testimony can in no way lead to the infliction of criminal penalties. See Kastigar, 406 U.S. at 461, 92 S.Ct. at 1665.

Second, courts can prevent the infliction of criminal penalties based on compelled, testimonial self-incrimination by refusing to force witnesses to testify against their own penal interest. See, e.g., United States v. Saline Bank of Virginia, 26 U.S. (1 Pet.) 100, 104, 7 L.Ed. 69 (1828) (affirming the denial of the Government's discovery motion because "a party is not bound to make any discovery which would expose him to penalties"). This prophylactic rule13 prevents the infliction of criminal penalties on the basis of self-incrimination by keeping the testimony hidden from the opposing party. What the movant does not know it cannot later use.

When a court considers a motion to compel testimony, it acts as a court of equity, balancing the risk of a violation of the Self-Incrimination Clause against the Government's right to the evidence of every citizen. See Mason v. United States, 244 U.S. 362, 364, 37 S.Ct. 621, 622, 61 L.Ed. 1198 (1917). If the court concludes, based on the circumstances of the case, that disclosure of the testimony would not lead to a conviction that violates the Self-Incrimination Clause, then the court may force the witness to speak. See, e.g., Murphy, 378 U.S. at 79, 84 S.Ct. at 1609--10 (holding that the defendants could be compelled to answer the Government's questions because a grant of state-law immunity shielded them from the use of their testimony and its fruits in federal court).

If, on the other hand, the court finds that compelling the testimony will likely lead to a conviction that violates the Self-Incrimination Clause, then the court must refuse to compel the witness to testify. See Pillsbury Co. v. Conboy, 459 U.S. 248, 256-57, 103 S.Ct. 608, 614, 74 L.Ed.2d 430 (1983). Thus, courts use two basic mechanisms for enforcing the Clause: excluding coerced testimony at a criminal trial and denying a motion to compel disclosure. Both mechanisms are designed to prevent conviction of an offense revealed through compelled testimony--the only harm against which the Self-Incrimination Clause protects.142.

The Fifth Amendment does not apply to foreign court proceedings involving foreign citizens. See Verdugo-Urquidez, 494 U.S. at 269, 110 S.Ct. at 1063. The Supreme Court has extended at least some of the protections of the United States Constitution to foreign citizens residing in the United States. See, e.g., Yick Wo v. Hopkins, 118 U.S. 356, 369, 6 S.Ct. 1064, 1070, 30 L.Ed. 220 (1886) (applying the Due Process Clause of the Fourteenth Amendment to resident aliens). The Court has also applied the Bill of Rights to the criminal prosecution of American citizens conducted by the United States in foreign lands. See, e.g., Reid v. Covert, 354 U.S. 1, 5-6, 77 S.Ct. 1222, 1225, 1 L.Ed.2d 1148 (1957) (holding that the United States Government must abide by the Constitution when it prosecutes civilian dependents accompanying members of the armed forces overseas).

The Court, however, has refused to apply the procedural protections of the Constitution to our government's treatment of foreign citizens in foreign countries. See, e.g., Johnson v. Eisentrager, 339 U.S. 763, 771, 70 S.Ct. 936, 940, 94 L.Ed. 1255 (1950) ("[I]n extending constitutional protections beyond the citizenry, the Court has been at pains to point out that it was the alien's presence within its territorial jurisdiction that gave the Judiciary power to act."); United States v. Curtiss-Wright Export Corp., 299 U.S. 304, 318, 57 S.Ct. 216, 220, 81 L.Ed. 255 (1936) ("Neither the Constitution nor the laws passed in pursuance of it have any force in foreign territory unless in respect of our own citizens."). The Court has also refused to apply our notions of due process to foreign court proceedings against American citizens who have committed foreign crimes outside the United States. See, e.g., Neely v. Henkel, 180 U.S. 109, 123, 21 S.Ct. 302, 307, 45 L.Ed. 448 (1901) ("When an American citizen commits a crime in a foreign country, he cannot complain if required to submit to such modes of trial and to such punishment as the laws of that country may prescribe for its own people.").

By necessary implication, the United States Constitution places no restraints at all on a foreign government's treatment of its own citizens who have allegedly committed foreign crimes abroad. See Underhill v. Hernandez, 168 U.S. 250, 252, 18 S.Ct. 83, 84, 42 L.Ed. 456 (1897); see, e.g., United States ex rel. Steinvorth v. Watkins, 159 F.2d 50, 51 (2nd Cir.1947) (holding that a United States court will not review the Costa Rican government's cancellation of appellant's Costa Rican citizenship). As Justice Kennedy wrote in United States v. Verdugo-Urquidez, "[t]he Constitution does not create ... any juridical relation between our country and some undefined, limitless class of noncitizens who are beyond our territory." Verdugo-Urquidez, 494 U.S. at 275, 110 S.Ct. at 1066 (Kennedy, J., concurring). Therefore, a foreign court cannot violate the Fifth Amendment to the United States Constitution when it inflicts criminal penalties on a foreign national based on compelled, testimonial self-incrimination.

Applying these principles to the current appeal, we conclude that a proceeding becomes a "criminal case" only when a witness faces conviction on the basis of his testimony in a jurisdiction subject to the Fifth Amendment of the United States Constitution. Gecas faces no possibility of a criminal conviction in the United States. He only faces trial and conviction in Israel, Germany, or Lithuania. The United States Constitution does not prohibit foreign countries such as these from trying and convicting their own citizens on the basis of self-incrimination. Therefore, Gecas does not face the kind of conviction proscribed by the Self-Incrimination Clause, and his deportation proceeding is not a "criminal case." The district court did not err by granting the Government's motion to compel Gecas to respond to the OSI's questions.

Gecas cites Murphy v. Waterfront Comm'n of New York Harbor, 378 U.S. 52, 84 S.Ct. 1594, 12 L.Ed.2d 678 (1964), for the proposition that he should not be forced to testify regardless of which sovereign intends to use his testimony to obtain a conviction. In Murphy, the Waterfront Commission of New York Harbor subpoenaed the defendants to testify concerning a work stoppage at several New Jersey piers. See id. at 53, 84 S.Ct. at 1596. The defendants asserted the privilege against self-incrimination and refused to testify. The defendants were granted immunity from prosecution under the laws of New York and New Jersey, but they persisted in their refusal to testify and were cited for contempt. Their appeal reached the United States Supreme Court.

The defendants in Murphy argued that they could refuse to testify because the grant of state-law immunity did not prevent their prosecution under federal law. They contended that federal prosecutors could use their testimony to initiate a federal prosecution. The Supreme Court held that the defendants could be forced to testify before the Waterfront Commission because the Self-Incrimination Clause barred the federal government from using their state testimony or its fruits to obtain their conviction in federal court. See Murphy, 378 U.S. at 79, 84 S.Ct. at 1609-10.

Gecas contends that his silence is protected because Murphy applied the Self-Incrimination Clause to the jurisdiction compelling the testimony--the state tribunal--as well as to the jurisdiction using the testimony--the federal tribunal. According to Gecas, "[t]he law of the sovereign that ultimately [makes] use of the compelled testimony [is] not ... controlling. The controlling fact [is] whether the witness is compelled to testify by a government whose law grants to him the right not to incriminate himself." Brief for Appellant at 14. Hence, Gecas claims that the Clause prohibits an American court from compelling his testimony even when it will be used, if at all, only in a foreign proceeding. We disagree.

To understand the holding of Murphy, we must place it in context. In Twining v. New Jersey, 211 U.S. 78, 29 S.Ct. 14, 53 L.Ed. 97 (1908), the Supreme Court held that the Fifth Amendment's privilege against self-incrimination was not one of the rights made applicable to the states by the Fourteenth Amendment. Id. at 110, 29 S.Ct. at 24. State-court defendants therefore had to rely on state-law provisions securing the privilege against self-incrimination. The decision in Twining in part gave rise to a series of decisions limiting the scope of the privilege to the jurisdiction in which it was invoked. See, e.g., Feldman v. United States, 322 U.S. 487, 64 S.Ct. 1082, 88 L.Ed. 1408 (1944), overruled by Murphy v. Waterfront Comm'n of New York Harbor, 378 U.S. 52, 84 S.Ct. 1594, 12 L.Ed.2d 678 (1964); see also id. at 490, 64 S.Ct. at 1083 ("[O]ne of the settled principles of our Constitution has been that these Amendments[, including the Fifth Amendment's Self-Incrimination Clause,] protect only against invasion of civil liberties by the Government whose conduct they alone limit.").

On June 15, 1964, the same day on which Murphy was decided, the Court overruled Twining in Malloy v. Hogan, 378 U.S. 1, 84 S.Ct. 1489, 12 L.Ed.2d 653 (1964). The Court held that the Fifth Amendment's Self-Incrimination Clause applies to the states through the Due Process Clause of the Fourteenth Amendment. Id. at 3, 84 S.Ct. at 1491. The application of the Fifth Amendment to both the federal and state governments raised the issue of whether the Fifth Amendment also applied between those governments. See Murphy, 378 U.S. at 53, 84 S.Ct. at 1595-96.

The Court chose Murphy as a vehicle for unifying the meaning of the privilege against self-incrimination within the United States. The Malloy Court cited Murphy and noted that "[i]t would be incongruous to have different standards determine the validity of a claim of privilege based on the same feared prosecution, depending on whether the claim was asserted in state or federal court." Malloy, 378 U.S. at 11, 84 S.Ct. at 1495. Likewise, the Murphy Court cited Malloy and reasoned that the policies behind the Self-Incrimination Clause would be defeated if a witness "could be whipsawed into incriminating himself under both state and federal law even though the constitutional privilege against self-incrimination is applicable to each." Murphy, 378 U.S. at 55, 84 S.Ct. at 1597 (quoting Knapp v. Schweitzer, 357 U.S. 371, 385, 78 S.Ct. 1302, 1310, 2 L.Ed.2d 1393 (1958) (Black, J., dissenting) (internal quotation marks omitted)). Thus, the Murphy Court's application of the Self-Incrimination Clause to both the compelling jurisdiction and the using jurisdiction was inextricably tied to a simultaneous decision applying the Self-Incrimination Clause to both jurisdictions. See (Under Seal), 794 F.2d at 926.

The Murphy Court noted this interrelationship:

In every "whipsaw" case, either the "compelling" government or the "using" government is a State, and, until today, the States were not deemed fully bound by the privilege against self-incrimination. Now that both governments are fully bound by the privilege, the conceptual difficulty of pinpointing the alleged violation of the privilege on "compulsion" or "use" need no longer concern us.

Id. at 57 n. 6, 84 S.Ct. at 1598 n. 6 (emphasis added). In this case, by contrast, there has been no recent decision applying the Clause to each of the governments involved. To the contrary, Supreme Court precedent indicates that Israel, Germany, and Lithuania are not bound by the privilege at all. They are not "States." We must therefore reach the question that the Murphy Court avoided: whether a violation of the privilege occurs upon compulsion of the testimony or upon its use.

Fortunately, in the years since Murphy, the Supreme Court has answered this question for us. The harm is the use of the testimony, not its compulsion. See Verdugo-Urquidez, 494 U.S. at 264, 110 S.Ct. at 1060 (citing Kastigar, 406 U.S. at 453, 92 S.Ct. at 1661). Gecas' testimony will be used, if at all, in a jurisdiction to which the Self-Incrimination Clause does not apply. Accordingly, we hold that Gecas cannot invoke the privilege before the compelling sovereign. We view this holding as consistent with the framework established in Murphy.

Gecas also claims support for his position from the Murphy Court's discussion of three old English cases: United States v. McRae, 3 L.R. 79 (Ch.App.1867); Brownsword v. Edwards, 28 Eng. Rep. 157 (Ex. 1750); and East India Co. v. Campbell, 27 Eng. Rep. 1010 (Ex. 1749). In analyzing the earliest of these cases, Campbell, the Murphy Court stated that "the privilege against self-incrimination protected a witness in an English court from being compelled to give testimony which could be used to convict him in the courts of another jurisdiction." Murphy, 378 U.S. at 58, 84 S.Ct. at 1598 (emphasis added). The Murphy Court wrote later in its opinion that "we now accept as correct the construction given the privilege by the English courts." Id. at 77, 84 S.Ct. at 1608. Gecas reads such language as an endorsement of his position. A conviction in Israel, Germany, or Lithuania would be a conviction in "another jurisdiction," so Gecas argues that Murphy proscribes court compulsion of his testimony in the United States.

Upon closer examination, however, Campbell, Brownsword, and McRae lend little support to Gecas' invocation of the privilege. In Campbell, the Court of the Exchequer considered a motion to compel the defendant to disclose how he obtained possession of the plaintiff's goods. See Campbell, 27 Eng. Rep. at 1010. The defendant answered that disclosure of this information would subject him to criminal penalties in India, which at that time was a colony of Britain. See id. The court denied the motion on the ground that "a jurisdiction is erected in Calcutta for criminal facts, where [the defendant] may be sent to government and tried." Id. at 1011. Thus, Campbell dealt with "disclosures which would have been incriminating under a separate system of laws operating within the same legislative sovereignty." Murphy, 378 U.S. at 82 n. 1, 84 S.Ct. at 1619 n. 1 (Harlan, J., concurring in the judgment).

Similarly, Brownsword concerned a discovery motion where the response would have subjected the defendant to prosecution for incest in an English ecclesiastical court. See Brownsword, 28 Eng. Rep. at 157. The court declared that "no one is bound to answer so as to subject himself to punishment, whether [sic] that punishment arises by the ecclesiastical law of the land." Id. at 158. Hence, Campbell and Brownsword stand for the proposition that different court systems operating under the same sovereign power must abide by the same procedural constraints.

So construed, Campbell and Brownsword directly support the holding of Murphy, which applied the privilege between jurisdictions within the United States. Campbell and Brownsword, however, did not consider the applicability of the privilege between separate, sovereign nations. Hence, they do not support Gecas' invocation of the privilege here.

The third case, McRae, did deal with separate nations. In that case, the United States itself sued McRae in England to recover funds McRae collected in England as an agent for the Confederacy during the Civil War. See McRae, 3 L.R. at 79. The United States moved to discover from McRae how much money he accepted on behalf of the Confederacy. McRae claimed that the United States had passed a law confiscating all of the property of the Confederacy. He argued that telling the United States how much he collected on behalf of the Confederacy would enable the United States to forfeit an equivalent amount of McRae's American property--even if McRae prevailed in the English court. Given this unusual scenario, the court denied the motion: "The United States coming into our Courts must be subject to every rule of evidence which prevails in them, and, amongst others, to that which protects a witness from exposing himself to penalties by his answer." Id. at 87.

Apart from the obvious factual distinctions between McRae and the present case, Parliament overruled McRae four years after Murphy was decided. In 1968, Parliament reviewed the English case law, including McRae, and limited the privilege against self-incrimination to potential criminal liability under the laws of the United Kingdom. See Civil Evidence Act, 1968, ch. 64, § 14(1) (Eng.); see also Law Reform Committee, Sixteenth Report, 1967, Cmnd. 3472, No. 113, at 7 (discussing the English cases). We decline to rely on a foreign case that has been overruled.

Even assuming that the Murphy Court's discussion of eighteenth- and nineteenth-century English case law supports Gecas' invocation of the privilege, we view this discussion as dicta. See Murphy, 378 U.S. at 81-82 & n. 1, 84 S.Ct. at 1619 & n. 1 (Harlan, J., concurring in the judgment); In Re Parker, 411 F.2d 1067, 1070 (10th Cir.1969), vacated as moot sub nom. Parker v. United States, 397 U.S. 96, 90 S.Ct. 819, 25 L.Ed.2d 81 (1970). As noted above, the defendants in Murphy did not face potential conviction under the laws of another sovereign nation. They faced potential conviction in a different court within our unified federal system. Therefore, any discussion in Murphy suggesting that the possibility of a foreign conviction permitted the invocation of the privilege could not have been necessary to the Court's decision. Murphy does not foreclose our own consideration of whether a fear of foreign conviction can justify invocation of the Self-Incrimination Clause.

Gecas argues in the alternative that the United States itself is both the compelling and the using sovereign. He claims that the OSI seeks to circumvent the Clause by orchestrating a foreign prosecution based on the testimony the OSI seeks here. Gecas contends that cooperation in this case among the OSI and the governments of Israel, Germany, and Lithuania transforms these governments into agents of the United States. As agents, these foreign governments are, according to Gecas, subject to the constraints of the Self-Incrimination Clause. See generally United States v. Heller, 625 F.2d 594, 599 (5th Cir.1980) (suggesting that the failure of foreign police to give Miranda warnings could violate the Fifth Amendment if the foreign police were acting as agents of the United States). He argues that the Clause therefore should prevent the compulsion of his testimony here. Again, we disagree.

The record suggests that the OSI may turn over its file on Gecas to whatever foreign government requests it. This type of intergovernmental cooperation, while it increases the chance Gecas will be convicted abroad, does not render the requesting government an agent of the United States. See, e.g., United States v. Behety, 32 F.3d 503, 511 (11th Cir.1994) (holding that Guatemalan officials did not become agents of the United States when they stopped and searched the defendants' boat on a tip from a DEA agent and then allowed other DEA agents to videotape them searching the defendants' boat), cert. denied, 515 U.S. 1137, 115 S.Ct. 2568, 132 L.Ed.2d 820 (1995); Birdsell v. United States, 346 F.2d 775, 782 (5th Cir.) (holding that Mexican police were not acting as agents of the United States when "American police officers gave information leading to the arrest and search" of the defendants in Mexico), cert. denied, 382 U.S. 963, 86 S.Ct. 449, 15 L.Ed.2d 366 (1965).

Extradition, another example of international cooperation, ordinarily does not subject the requesting sovereign's legal regime to constitutional scrutiny in American courts. See Glucksman v. Henkel, 221 U.S. 508, 512, 31 S.Ct. 704, 705, 55 L.Ed. 830 (1911); Garcia-Guillern v. United States, 450 F.2d 1189, 1192 (5th Cir.1971). Mere delivery of Gecas's file to a foreign government should produce no different result. Israel, Germany, and Lithuania are independent nations. They will decide for themselves whether or not to prosecute crimes falling within their jurisdictions. We will not assume that the Executive Branch is colluding with foreign powers to circumvent the United States Constitution whenever it shares information on the conduct of foreign nationals--particularly where the conduct in question is not a crime under American law. In short, Gecas has failed to show that the United States is effectively prosecuting him in a foreign country.

We do not base our decision on the negative repercussions of the rule urged by Gecas, but we recognize that those repercussions would be substantial. Extension of the Self-Incrimination Clause to reach foreign convictions would bind the conduct of American officials to the legislative acts of foreign nations. Assume, for example, that the sale and distribution of cocaine is illegal in the nation of Ames. A citizen of Ames visits the United States on a temporary visa. She is arrested in Miami International Airport after police find five kilograms of cocaine in her luggage. The prosecution offers her immunity in exchange for testimony about three leaders of a drug ring operating in Southern Florida. Under the rule proposed by Gecas, she can still refuse to testify because Ames may prosecute her for the same conduct. We agree with the Fourth Circuit that "our own national sovereignty would be compromised if our system of criminal justice were made to depend on the actions of foreign government[s] beyond our control." (Under Seal), 794 F.2d at 926.

Conversely, the rule urged by Gecas would constrain the law enforcement activities of foreign countries where the privilege against self-incrimination does not exist. Fugitives who flee to the United States would become entitled to a procedural right they lack in their own countries. "Comity among nations dictates that the United States not intrude into the law enforcement activities of other countries conducted abroad." Id. Stated differently, we leave to the Executive Branch, with the advice and consent of the Senate, the decision to negotiate the application of the privilege to both domestic and foreign court proceedings.15 See Phoenix Assurance Co. of Canada v. Runck, 317 N.W.2d 402, 413 (N.D.), cert. denied, 459 U.S. 862, 103 S.Ct. 137, 74 L.Ed.2d 117 (1982). In conclusion, we find that our precedent sensibly limits the application of the Fifth Amendment's Self-Incrimination Clause to domestic convictions.

We now turn to the broadest contention made by Gecas: he contends that, precedent notwithstanding, the Framers intended the privilege against self-incrimination to extend to the possibility of incrimination under the law of a foreign jurisdiction. According to Gecas, the history of the privilege against self-incrimination indicates that its "main purpose" is the protection of individual privacy and dignity. See Brief for Appellant at 5. He claims that the privilege creates in the individual a freedom to remain silent whenever his testimony may be adverse to his penal interests, regardless of where the infliction of criminal penalties will occur. Gecas contends that the Framers intended the Self-Incrimination Clause to bestow upon him this personal, individual right. We must therefore examine the history of the privilege against self-incrimination to determine whether the policies behind the Clause justify its application to Gecas.

Unfortunately, the cases do not provide us with a detailed exposition of the history of the privilege against self-incrimination. The Fifth Amendment's Self-Incrimination Clause itself has virtually no legislative history. Beyond the few ambiguous English cases cited by the Murphy Court, no court has examined the history of the privilege with regard to self-incrimination under foreign law. Therefore, we must pull together, largely from academic scholarship, an understanding of the history of the privilege as it developed in continental Europe, in England, and in America. This undertaking will fill a few more pages of the federal reporter than we might otherwise wish, but we consider a full understanding of the privilege essential to the resolution of the important issue presented in this case. The following is a summary of our discussion:

We begin in subpart A.1 with medieval ecclesiastical procedure. Ecclesiastical courts in continental Europe barred compelled, testimonial self-incrimination until an accuser stepped forward. Once accused, however, defendants were forced to prove their innocence through the performance of a ritualistic test. Because all convictions were based on the defendant's behavior in performing the test, all convictions were based on self-incrimination. Whether an individual would be forced to undergo a test depended on the legal sufficiency of the accusation.

We turn in subpart A.2 to medieval English courts, both ecclesiastical and secular. At Rome's direction, medieval English ecclesiastical courts relaxed the requirements for a legally sufficient accusation. In particular, judges themselves were allowed to accuse and force an incriminatory test. English common-law courts, by contrast, did not follow ecclesiastical law on this point. They required a personal accusation or the accusation of an inquest, the precursor to our modern grand jury.

In subpart B.1, we discuss the effect of the advent of the jury trial. The jury trial gradually replaced the ancient tests as the method of proof used in the common-law courts. Access to a neutral trier of fact meant that not every conviction would be a conviction based on self-incrimination. In subpart B.2, we explain why the jury trial did not cause the common-law courts to recognize a privilege against self-incrimination: when the jury began to hear evidence, defendants were disqualified from testifying under oath. They could not be called as witnesses for the prosecution, so there was no need for a privilege.

In subpart C, we address how the common-law courts came to recognize an implicit privilege against self-incrimination. By the mid-1500s, common-law judges themselves were taking testimony at preliminary hearings. Relying on their interpretation of medieval ecclesiastical law, they refused to interrogate suspects under oath at the preliminary hearing, although they did compel sworn testimony from nonparty witnesses. Abstention from sworn interrogation of defendants created an implied privilege against self-incrimination at common law.

In subpart D, we consider how that implied privilege eventually became explicit. During the 1600s, the British monarchy enlarged the jurisdiction of its prerogative tribunals,16 the Star Chamber and the Commission for Causes Ecclesiastical (the "Commission"),17 in an effort to suppress religious and political heterodoxy. Subpart D.1 shows how the common-law courts fought against the Crown's expansion of its prerogative jurisdiction. Subpart D.2 describes the procedures of the Star Chamber and the Commission. Subpart D.3 discusses how religious dissenters, particularly the Puritans, fought against the prosecutorial techniques of these two tribunals.

Subpart E deals with the actual recognition of the privilege against self-incrimination, first in ecclesiastical courts and then at common law. Together, the common-law courts and the religious dissenters of the seventeenth century achieved the abolition of the royal prerogative courts and caused the enactment of a privilege against self-incrimination in ecclesiastical courts. When the common-law courts finally gave substance to their own implied privilege against self-incrimination in the late-1600s, they viewed it as a limitation on the power of centralized government to punish crimes of conscience by diluting the procedural protections of the common law.

In subpart F we turn to the American colonies. Subpart F.1 discusses colonial criminal procedure and the recognition of the privilege in America. The colonists gradually came to view the repressive measures of the British authorities as a repeat of the actions of the prerogative courts of the seventeenth century--the institutions which had driven many of their ancestors out of England. We explain in subpart F.2 the various attempts to codify the privilege after the American Revolution. Eight of the thirteen newly-independent states enacted an explicit privilege against self-incrimination. Almost all early discussion of the privilege limited its application to criminal prosecutions. We conclude that the Framers adopted a privilege against self-incrimination, applicable in "any criminal case," as a limitation on the power of centralized government to prosecute crimes of conscience by weakening the procedural protections of the common law.

Early in the history of the Roman Catholic Church, confession of sin became an obligation of faith. The Church decided during the third century A.D. that parishioners would be more likely to confess privately to a priest than publicly before the congregation. See Albert W. Alschuler, A Peculiar Privilege in Historical Perspective: The Right to Remain Silent, 94 Mich. L.Rev. 2625, 2639-40 (1996). Parishioners naturally would have expected that a public confession of crime would result in punishment by ecclesiastical or secular authorities, and confidentiality helped to overcome their reluctance to expurgate their sins. Christians were free from the moral duty spontaneously to confess their sins in public. In the eyes of the Church, they had a right to remain silent about their own transgressions.

This privilege against public confessions, however, ceased to apply once an accuser stepped forward and informed the ecclesiastical authority that an individual had committed an offense. A formal accusation placed the burden of proof on the accused to exonerate himself. The accused discharged the burden of proof through the successful performance of a ritualistic test, the nature of which was decided by the ecclesiastical court based on the accusation. The primary methods of proof were ordeal,18 battle, and precise oaths accompanied by the supporting oaths of relatives. See generally 2 Frederick Pollock & Frederic W. Maitland, The History of English Law 598-602 (Legal Classics Library 1982) (1899) (explaining the different methods of proof sanctioned by the Ecclesiastical tribunals). The purpose of these rituals was to allow God through divine intervention to reveal which party was telling the truth.

Each of the methods of proof relied entirely on the behavior of the defendant, under the threat of a penalty, to determine guilt. In this sense, every conviction under the ancient methods of proof was a conviction based on self-incrimination.19 For example, if a party came forward with an accusation that adhered to the proper formulas, the court might force the defendant to pay a fine or, alternatively, to swear, along with twelve family members, to the falsity of the charge. If the defendant and these co-swearers precisely recited the appropriate set of oaths, the defendant would be acquitted. If not, the defendant would be convicted. All convictions under the ancient methods of proof resulted from the defendant's failure to perform the required test.

In early medieval ecclesiastical procedure, then, the accusation ended a defendant's right to remain silent. If properly accused, a defendant could be summoned and forced to prove his innocence: "Licet nemo tenetur seipsum prodere, tamen proditus per famam tenetur seipsum ostendere utrum possit suam innocentiam ostendere et seipsum purgare " ("Although no one is bound to produce himself, he who is singled out by rumor is bound to show whether he can prove his innocence and to purge himself."). See M.R.T. MacNair, The Early Development of the Privilege Against Self-Incrimination, 10 Oxford J. Legal Stud., Spring 1990, at 66, 70. This "nemo tenetur principle" marked the boundary between protected silence and compelled self-incrimination. There was no privilege against self-incrimination after an accusation was made.

Moreover, beginning in the ninth century, the Church relaxed the requirements for a legally sufficient accusation, which increased the use of self-incriminatory tests. Ecclesiastical courts began to prosecute individuals not just on the basis of a private accusation, but also on the basis of public rumors of wrongdoing. See Adhemar Esmein, A History of Continental Criminal Procedure with Special Reference to France 79 (1913). Church leaders sought this expansion of ecclesiastical jurisdiction because private accusers often refused to come forward and face the severe penalties for false accusation.

The Church borrowed one method of discovering rumors of wrongdoing from the royal courts of Charlemagne--a procedure later known as the "inquest." See generally Leonard W. Levy, Origins of the Fifth Amendment 22 (1st ed.1968) (explaining the ecclesiastical inquest). A bishop would visit a monastery and convene an inquest of several monks. The bishop would "deputize" this group to reveal those who were rumored to be guilty of certain crimes. See id. The bishop would then call the accused to exculpate themselves.

Defendants incriminated by popular report of an inquest could still obtain an acquittal through the successful performance of an ordeal or an oath, but, as discussed above, these methods of proof effectively compelled defendants to be witnesses against themselves by placing on them the burden of proving their innocence. On the other hand, those who refused the opportunity to exonerate themselves were condemned as if they had confessed. Thus, the implementation of the ecclesiastical inquest increased the use of compelled self-incrimination. Until 1198, however, an ecclesiastical court could not proceed against an individual without, at very least, a private accusation or a strong rumor of criminal activity, as indicated, for example, through an inquest.20

In 1198, Pope Innocent III began to issue a series of decretals authorizing a procedure that became infamous under the auspices of the Holy Inquisition, the processus per inquisitionem (the "inquisitional procedure"). See generally Esmein, supra, at 80. The inquisitional procedure was novel in two main respects. First, a priest was permitted to proceed, on the authority of his office, against individuals who had not been accused of any wrongdoing. A priest could initiate secret proceedings against any individual by gathering witnesses and interrogating them separately. If the witnesses examined by the priest incriminated the target of the investigation, the priest could summon the target.

Second, Innocent III authorized the administration of a different kind of oath, the oath de veritate dicenda (the "inquisitional oath") to witnesses. The inquisitional oath required witnesses, including defendants, to answer truthfully all the questions to be asked by the judge--often before they knew the nature of the questions to be asked.21 For defendants, silence in the face of an accusation was taken as a confession of guilt. Once again, the ecclesiastical courts had expanded the use of compelled self-incrimination.

The first defendants who were tried under the inquisitional procedure objected to the absence of any public accusation of wrongdoing against them. The church hierarchy responded by pointing to numerous biblical examples of God's intervention in the affairs of men. Just as God interceded to punish and reform sinners, so too could God's Church through its ministers. See Esmein, supra, at 81.

The first defendants also objected to the Church's refusal to allow them to exculpate themselves with a purgatory oath, a simple, blanket denial of wrongdoing. The church leaders responded that the inquisitional oath was actually more lenient than the purgatory oath. To use a purgatory oath, the defendant had to obtain supporting oaths from relatives and friends. Under the inquisitional oath, by contrast, the accused's promise to tell the truth sufficed without more.

Finally, these defendants argued that the longstanding privilege against public confessions, originally adopted to encourage confidential confessions, barred the administration of the inquisitional oath. The Church leaders rejected this argument too. The privilege against public confessions merely spared the faithful from the requirement of spontaneously producing evidence of their own guilt in public. That privilege did not entirely excuse the wayward from their duty to confess; once formally accused of a crime, they were required to confess their error or purge themselves. The Fourth Lateran Council upheld the inquisitional procedure in 1215 and effectively decided that the sanctity of the confessional provided no defense against the imposition of the inquisitional oath.

Thus, the ecclesiastical courts of the early medieval period recognized no privilege against self-incrimination. The only limitation on prosecutorial discretion was the requirement of an accusation, as represented by the nemo tenetur principle. The first real debate about compelled self-incrimination centered on who could accuse--not whether the defendant should be exempt from all sworn interrogation. The inquisitional procedure simply allowed the judge to decide that a common report of criminal wrongdoing was sufficient to overcome the privilege against public confessions.

Before the Norman Conquest in 1066, the English common-law courts heard both secular and religious cases according to the ancient accusatorial system of ordeals, battles, and oaths. Religious and secular jurisdictions were unified. During this period of shared jurisdiction, the common-law courts effectively adopted the nemo tenetur principle: an individual could only be forced to face the ordeal, fight, or swear if properly accused. The nemo tenetur maxim was enshrined in chapter 28 of the Magna Carta, the first English bill of rights: "No Bailiff can put any one to his Law upon his single accusation, without sufficient witnesses." Magna Carta ch. 28 (1215).

In accordance with continental custom, however, William the Conqueror severed secular from ecclesiastical jurisdiction sometime in the latter 1160s. See 1 Pollack & Maitland, supra, at 97-98. In 1236, a papal emissary announced the Pope's promulgation of the inquisitional procedure at the Pan-Anglican Council in London. See John H. Wigmore, The Privilege Against Self-Incrimination; Its History, 15 Harv. L.Rev. 610, 610 (1902). Thereafter, administration of the inquisitional oath became routine in English ecclesiastical courts.

The English secular courts, by contrast, never adopted the inquisitional procedure. They had split off from the ecclesiastical courts some seventy years earlier. The most common form of secular criminal trial in twelfth-century England involved an accusation of wrongdoing by private citizens.22 Before the defendant was required to answer, the accuser would be required to make an offer of proof by obtaining the supporting oaths of so-called "suitors," who were usually the accuser's relatives or allies in the community.

Unlike the ecclesiastical courts, English common-law courts dispensed with this formal, private accusation of crime in only two situations. First, if someone was caught in the act of committing a crime or in flight from a crime, courts would usually impose a punishment without any further inquiry. Second, individuals could be tried without a private accusation if an inquest offered up their names as reputed felons.

From the time of Charlemagne, the Norman kings authorized their itinerant justices of the peace ("JPs") to summon groups of prominent local inhabitants--or "jurors"--to answer under oath about the affairs of the locale. See generally 2 Pollock & Maitland, supra, at 520-21. During the twelfth century, JPs began to use the inquest of jurors to inquire about reputed criminals. See generally 1 Pollock & Maitland, supra, at 140-141. If the inquest swore that a particular individual was rumored to be a thief, for example, that individual would be arrested and jailed.23 The court held no preliminary inquiry regarding the truth of the allegation. See 2 Pollock & Maitland, supra, at 582. At trial, the accused would be required to make an offer proof, just as if a private plaintiff had made the accusation.

In sum, like defendants in the ecclesiastical courts of the twelfth century, defendants at common law could only be summoned to answer for a crime if properly accused. See Esmein, supra, at 337. Once accused, defendants at common law, like defendants in church proceedings, were required to offer proof of their innocence in the form of an ordeal, battle, or oath. They too had no privilege against self-incrimination. Ecclesiastical and secular proceedings differed primarily in that, from the early thirteenth century, church leaders themselves were permitted to accuse on the authority of their office, while JPs had to wait for a private accusation or an indictment by inquest.

Beginning in the thirteenth century, trial by jury began to replace the ancient methods of proof in English common-law courts. This development removed from the defendant's shoulders the burden of proof associated with ordeals, battles, and oaths. By allowing defendants recourse to an ostensibly neutral determination of guilt or innocence, the jury trial ended the automatic self-incrimination inherent in the ancient methods of proof.

Ordeals, battles, and oaths fell into disfavor. In 1215, the Fourth Lateran Council forbade the clergy from participating in trial by ordeal, depriving these rituals of their religious sanction. Trial by battle was rarely available because of several exceptions, such as the exception for aged and infirm litigants, and the widespread use of hired champions undermined the perceived validity of battle. See 2 Pollock & Maitland, supra, at 633. Courts were reluctant to allow trial by oath in metropolitan jurisdictions, where professional swearers could earn easy acquittals for their clients. Defendants in rural common-law courts often found it difficult to gather sufficient relatives and neighbors willing to swear the required supporting oaths. Moreover, courts often refused to allow repeat offenders to take the purgatory oath. With these older forms of proof in decay, the common-law courts of the thirteenth century began to turn to a different method of proof--the ubiquitous inquest.

Trial by jury arose from the combination of two procedural reforms: the fact-finding inquest and the exception. As noted above, the Norman kings used inquests in the eleventh century to indict suspected criminals. Beginning in 1164, King Henry II also authorized private litigants to summon an inquest to resolve certain types of civil disputes. See 1 Pollock & Maitland, supra, at 145-46. Around the same time, the royal courts were beginning to recognize the defendant's ability to "take exception" to the plaintiff's allegations. See 2 Pollock & Maitland, supra, at 587--88. Instead of denying the allegations outright, the defendant could, for example, attack the plaintiff's pleadings on the ground that the complaint was motivated by "hatred and spite." During the mid-thirteenth century, defendants began to obtain writs of inquest to determine the validity of the exception. By taking exception and summoning an inquest, a civil litigant in the king's courts could choose to be "put upon his country"--to rely on a verdict of his neighborhood--instead of vindicating himself through ordeal, battle, or oath. The combination of inquest and fact-based exception created the jury trial.

Jury trials spread to criminal cases in the latter half of the 1200s. See id. at 618. At first, the inquest that was called to inform the king of suspected lawbreakers also decided the guilt of the accused at trial. In 1352, however, King Edward III established that the defendant could exclude from the trial jury any person who was on the inquest that accused him, a reform which effectively separated the grand jury from the petit jury. Although trial by battle was not abolished until 1819, and trial by oath was not abolished until 1833, the jury trial soon came to dominate the landscape of English criminal procedure. See Esmein, supra, at 334.

The jury trial was a necessary step towards the recognition of the privilege against self-incrimination. The arrival of the jury trial allowed defendants to discharge their burden of proof without themselves becoming involved in the adjudication. Instead of performing a religious ritual, engaging in battle, or taking an oath, defendants could rely on an impartial determination of the underlying facts. Unlike a conviction under the ancient methods of proof, conviction by a jury did not necessarily entail self-incrimination by the defendant. The rise of the jury trial effectively ended the use of the self-incriminatory methods of proof.

The advent of the jury trial, however, did not cause the common-law courts to recognize a privilege against self-incrimination. There was no immediate need for a privilege against compelled, testimonial self-incrimination because in early-fourteenth-century England, the jury heard no courtroom testimony at all--much less compelled, self-incriminatory testimony. See id. at 328. Jurors decided a defendant's guilt based on their own personal observations, their judgments about the defendant's character, and even hearsay and gossip within the community. Conscientious jurors investigated the crime on their own initiative by interviewing witnesses and visiting the crime scene. See id. at 327. Juries still rendered verdicts on the basis of their own private sleuthing well into the eighteenth century. See id. at 329.

Live witness testimony became a regular fixture in jury trials only in the sixteenth century. See 9 William Holdsworth, A History of English Law 178 (3rd ed. 2nd prtg.1976) (1926). Even then, the question of testimonial self-incrimination did not immediately arise, for two reasons.

First, defendants were barred from testifying under oath at trial. Defendants were initially disqualified because allowing them to give sworn testimony was thought to combine unfairly two methods of proof--trial by oath and trial by jury. See Wigmore, The Privilege Against Self-Incrimination, supra, at 628. The common-law courts allowed defendants only one method of proof--one "bite of the apple." As the law of evidence evolved, courts began to exclude the defendant's sworn testimony based on its assumed unreliability. Defendants were considered presumptively biased. Although the absence of defense counsel forced defendants to argue on their own behalf at trial, common-law courts ruled in the latter half of the sixteenth century that the jury was not to assign any evidentiary value to the defendant's factual assertions. See 9 Holdsworth, supra, at 195. In fact, criminal defendants could not testify under oath at all until 1872, and they could not testify under oath in all criminal cases until 1898. See Frederic W. Maitland & Francis C. Montague, A Sketch of English Legal History 176 & n. 3 (1915). Accordingly, until 1872, the state could not have compelled the defendant to testify under oath at trial.

Second, until the mid-sixteenth century, all nonparty witness testimony in common-law criminal trials was voluntary. The Chancery Court invented the subpoena in the fourteenth century to compel the attendance of nonparty witnesses in equity cases. See 9 Holdsworth, supra, at 184. Common-law courts, however, did not obtain this power until 1562. Before then, nonparty witnesses testified, if at all, only reluctantly, mainly because they assumed potential liability for false imprisonment if the opposing party prevailed at trial. See 9 Holdsworth, supra, at 130--31. Because all nonparty witness testimony was voluntary until the mid-sixteenth century, nonparty witnesses could not be compelled to incriminate themselves.

Therefore, introduction of the jury trial did not immediately give rise to a need for the privilege against self-incrimination. Jury trial, however, did set the stage for an implicit recognition of a privilege against self-incrimination in the sixteenth century. As the jury began to hear evidence, courts began to consider what types of evidence should be excluded. New investigative techniques, including the preliminary examination of the accused and compulsory process for witnesses, produced incriminatory testimony. The common-law courts reacted to these procedural innovations by restricting the permissible scope of the defendant's interrogation.C.

The procedural innovations began in 1383. In that year, Parliament passed the first statutes authorizing JPs to examine suspects and witnesses before trial in minor criminal matters. These preliminary examination statutes covered a wide array of subject matters, from heresy to poaching, and established a variety of procedures to be applied, depending on the nature of the case. Most of the early statutes concerned misdemeanors and simply authorized JPs to take testimony and decide guilt themselves, dispensing with the trial altogether. The majority of these statutes did not mention whether or not the examination was to be under oath. Other statutes were more specific and actually required the JPs to take sworn testimony from the witnesses and the suspect.

For example, a 1414 statute authorized the JPs to interrogate laborers under oath to discover if they had committed any crimes. See John H. Langbein, Prosecuting Crime in the Renaissance: England, Germany, France 68 (1974). In the sixteenth century, Parliament authorized common-law courts to interrogate under oath accused bankrupts, abusers of warrants, and other specific types of criminals. See 8 John H. Wigmore, Evidence in Trials at Common Law § 2250, at 285--86 (McNaughton rev. ed. 1961). At first, most common-law courts accepted the practice of examining both suspects and witnesses under oath. See John H. Wigmore, Nemo Tenetur seipsum Prodere, 5 Harv.L.Rev. 71, 76-78 (1891).

In 1554, however, a discourse began among judges about the propriety of sworn interrogation. In that year, Parliament enacted the Bail Statute. See An Act appointing an Order to Justices of Peace for the Bailement of Prisoners, 1 & 2 Phil. & M., ch. 13, § 1 (1554) (Eng.) [hereinafter Bail Statute]. In the following year Parliament enacted the Committal Statute. See An Acte to take the Examination of Prysoners suspected of Manslaughter or Felonye, 2 & 3 Phil. & M., ch. 10, § 1 (1555) (Eng.) [hereinafter Committal Statute]. These statutes systematically applied the apparatus of the preliminary examination to felonies. The purpose of the Bail Statute and the Committal Statute was to augment the system of private accusations in cases of serious crime where no accuser stepped forward. See Langbein, supra, at 39.

The system established by the Bail Statute and the Committal Statute functioned as follows: if the suspects or witnesses were not already before the court, the court could compel their attendance by issuing a warrant to the sheriff to apprehend them and bring them into court. See Michael Dalton, The Countrey Justice 400-01 (Legal Classics ed.1996) (1655). Once the parties were present, the court examined them and recorded only the information that could be used to prove the guilt of the accused. See Committal Statute, supra, at § 1; see also Bail Statute, supra, at § 1 (providing similar language). JPs had no authority to discharge accused felons. Until the nineteenth century, in fact, they were required to either jail or bail the accused--the preliminary hearing was not a probable cause hearing. See J.M. Beattie, Crime and the Courts in England 1660--1880, at 271 (1986).

The language of the statutes did not tell JPs whether or not to interrogate suspects and witnesses under oath. In practice, JPs consistently threatened, cajoled, and badgered suspected felons into confessing at the preliminary hearing. JPs recorded any incriminating statements and presented them to the jury at trial. Reading the record of the examination was the first step in most criminal trials during the sixteenth and seventeenth centuries. See 8 Wigmore Evidence, supra, § 2250, at 286. The suspect's statements during the examination were apparently treated as admissions. Magistrates were not required to warn suspected felons of the potentially dire consequences of their pretrial statements, however, until 1848. See Sir John Jervis' Act (Administration of Criminal Justice), 11 & 12 Vict., ch. 42, § 18 (1848) (Eng.).

Nevertheless, JPs disagreed for several decades about whether suspected felons should be administered an oath before the interrogation. For the first two decades after the enactment of the statutes, some courts apparently took sworn testimony while others did not. In the 1580s, however, a consensus began to emerge that commonlaw courts should not interrogate suspected felons under oath during the preliminary examination. See Levy, supra, at 107. In support of this new rule, the common-law judges of the seventeenth century reiterated the arguments of the ecclesiastical lawyers of the 1200s, particularly the nemo tenetur principle.

Under the ancient accusatorial system, a suspected felon could only be brought to trial on the basis of a private accusation or an indictment by an inquest of jurors. Chapter 28 of the Magna Carta established the requirement of an accusation early in the history of the common law. Some common-law judges opined that allowing the court to interrogate suspected felons under oath during the preliminary examination would undermine the requirement of an accusation. They reasoned that, in effect, sworn interrogation forced the suspect to take an oath before a private citizen or an inquest had accused him:

The offender himself shall not be examined upon oath, for by the Common Law, Nullus tenetur seipsum prodere: Neither was a mans fault to be wrung out of himself (no not by examination only) but to be proved by others, untill the Stat. of 2 & 3. Ph. & M. cap. 10. gave authority to the Justices of peace to examine the Felon himself.

Dalton, supra, at 369. Thus, the common-law courts rejected the path taken by the ecclesiastical courts in establishing the inquisitional procedure.

Against the background of the jury trial and testimonial disqualification, this application of the nemo tenetur principle by the common-law courts effectively conferred on defendants an implied privilege against self-incrimination. As originally conceived, the nemo tenetur principle allowed a court to force properly accused defendants to swear to their innocence at trial. The ecclesiastical courts essentially had followed this model: once a private accuser, an inquest of clerics, or the bishop summoned a defendant, the trial court formally interrogated the defendant and other witnesses under oath and based its verdict on their testimony.

Common-law courts, by contrast, gave defendants the option of exoneration through jury trial rather than through oath or battle. Moreover, the disqualification of defendants from testifying under oath at trial prevented the state from calling them as witnesses. Under the common-law rule, then, the nemo tenetur principle barred the court from interrogating defendants under oath at the pretrial hearing, and proof by jury plus witness-disqualification allowed defendants to avoid giving incriminatory testimony at trial. In combination, these policies prevented the state from compelling defendants to testify against themselves.

The implied privilege against self-incrimination for defendants contrasted with the treatment of nonparty witnesses. As noted above, JPs had the power to compel the attendance of witnesses at the examination by issuing a warrant to the sheriff to bring them before the court. See id. at 401. The Bail Statute and the Committal Statute further authorized magistrates to bind over prosecution witnesses who were before them to give testimony at trial against the defendant. See Committal Statute, supra, at § 2; see also Bail Statute, supra, at § 1 (providing similar language). In other words, after recording the testimony of prosecution witnesses, the examining magistrate could require those witnesses to post a bond to secure their attendance at trial.24 If a witness refused to post a bond, then the magistrate could incarcerate the witness to ensure attendance. See Dalton, supra, at 397.

If a witness before the court agreed to be bound over but failed to appear, JPs apparently had several options. First, they could order the forfeiture of the witness's bond. Second, they could jail the witness for contempt. See id. at 401. Third, they could substitute the transcript of the witness' prior sworn testimony for the expected trial testimony, provided that the JP who presided over the preliminary examination was present to attest to the accuracy of the transcript. See Langbein, Prosecuting Crime, supra, at 30 (citing 2 Thomas Smith, De Republica Anglorum 79-80 (L. Alston ed., 1906) (1565)). Fourth, JPs could order a continuance until the prosecution could persuade the witness to testify. Fifth and finally, JPs could order the release of the defendant, but they rarely chose this option. See id. Thus, the Bail and Committal Statutes expanded the courts' power to compel the testimony of prosecution witnesses in criminal cases.25

Despite the potential for compelled self-incrimination, the common-law courts did not recognize a privilege for nonparty witnesses until the second half of the seventeenth century. See 9 Holdsworth, supra, at 199, 229. For example, in the 1580 trial of Thomas Tresham for harboring a heretic, the Lord Chief Justice opined that nonparty witnesses could be compelled to testify under oath at trial, regardless of the nature of their testimony, because they were not themselves in criminal jeopardy. See Levy, supra, at 105. While JP manuals of the time cited the nemo tenetur maxim for the proposition that judges should not question suspects under oath at the preliminary examination, those same manuals instructed JPs to interrogate witnesses under oath. See, e.g., Dalton, supra, at 369.

Therefore, with regard to defendants, an implied privilege against self-incrimination originated in the application of the nemo tenetur principle to preliminary examinations under the Committal and Bail Statutes. While the common-law courts generally applied the nemo tenetur maxim to interrogation of the defendant at the preliminary investigation, they simultaneously required sworn testimony from nonparty witnesses, apparently without regard to the potential for self-incrimination. This contrasting treatment indicates the presence of a limited privilege against self-incrimination for defendants. Nevertheless, the common-law courts did not recognize an explicit privilege against self-incrimination, for defendants or nonparty witnesses, until the late seventeenth century.

Discussion of an explicit privilege against self-incrimination began in the early seventeenth century and arose in two contexts. First, the English common-law courts began to attack the jurisdiction of the prerogative tribunals established by the monarchy for the enforcement of religious and political orthodoxy. The common-law courts did not challenge these rival courts in defense of an individual right to remain silent in the face of self-incriminating questions. Rather, the common-law courts sought to block the Crown from undermining common-law jurisdiction by creating special courts that could operate free from the constraints of common-law procedure.

Second, the religious nonconformists of the early seventeenth century asserted the privilege as a method for obstructing the prosecution of crimes of conscience. Opponents of the Church of England cited the nemo tenetur maxim as an argument against the royal prerogative courts--the Star Chamber and the Commission--that were enforcing establishmentarian dogma. For both the common-law courts and the religious dissenters of the period, the implied privilege against self-incrimination in common-law criminal cases came to represent a limitation on the power of centralized government to take action against private citizens.

The scope of the Church's jurisdiction over the laity did not become settled until the seventeenth century. During the thirteenth century, the Church assumed broad jurisdiction over several types of disputes, including the correction of sinners. Ecclesiastical jurisdiction included prosecution of fornication, adultery, incest, defamation, usury, simony, perjury, heresy, and sorcery. See id. at 129-30. With ecclesiastical jurisdiction sometimes came the severe inquisitional procedure. See generally, Levy, supra, at 47-48.

Protests against the severity of the inquisitional procedure may have led Parliament to pass the Prohibitio Formata de Statuto Articuli Cleri in the early 1300s. See Levy, supra, at 48-49. This statute purported to restrict the use of the inquisitional oath and thereby limit the jurisdiction of ecclesiastical tribunals. The statute provided that JPs should not allow "that laypersons in their bailiwick be called somewhere to give testimony on their oath, except in causes matrimonial and testamentary." See Wigmore, The Privilege Against Self-Incrimination, supra, at 611-612 (emphasis added).

The statute, however, did not specify how JPs were to prevent such interrogations, and, in practice, ecclesiastical courts continued to prosecute sin among the laity, particularly after 1382. See generally Levy, supra, at 49. In fact, by statute, ecclesiastical courts formally received jurisdiction over the crime of heresy in 1401. See Wigmore, The Privilege Against Self-Incrimination, supra, at 6