dissenting:
Nearly eight years ago we had the occasion to state that “[t]errorism, both domestic and international, is one of the gravest problems of our day.”1 It still is. Events, of which we are all aware, even since the argument of this appeal and continuing to the date of the filing of our opinions in this case, make it increasingly clear that organized terrorism, like organized crime, is not about to go away. It calls for protective measures with a severity proportionate to the danger at hand.
The essential question presented on this appeal is whether the United States is powerless to protect the public from the depredations of appellants who are key members of Los Macheteros, one of the most notorious terrorist organizations in the western hemisphere, based in Puerto Rico. I believe that the United States is not powerless tb provide the needed protection. One critical aspect of such protection has been provided by Congress in the Bail Reform *1010Act of 1984 — specifically in permitting pretrial detention of those defendants found to constitute a danger to the community.
I regret that I am unable to join in the thoughtful, scholarly, innovative opinions of Judge Newman and Chief Judge Feinberg. I am unable to do so simply because I disagree with their views that result in eliminating pretrial detention for dangerousness as an option available to courts, where, as here, an experienced magistrate and an experienced district judge have found that the defendants constitute a danger to the community. In so doing, Judge Newman and Judge Feinberg, in a case of first impression, have struck down as unconstitutional, on its face and as applied, a key provision of the Bail Reform Act.
I therefore dissent most emphatically from my colleagues’ separate views that the Fifth Amendment right to due process of law was violated in the cases of those appellants who are being detained pending trial on the ground that they constitute a danger to the community. Judge Newman and Judge Feinberg both recognize the government interest in preserving the integrity of the trial process and the interest in protecting the community from defendants who have been charged with very serious crimes and who are shown to be likely to commit additional crimes if they are released pending trial. I disagree, however, with the significantly greater weight my colleagues accord to preserving the integrity of the trial process and preventing risk of flight, compared with protecting the community from those who have been found to constitute a danger to the community. My colleagues do not give sufficient weight to the intent of Congress in enacting the Bail Reform Act and to the care exercised by Congress in establishing procedural requirements which prevent unjustified infringement upon the liberty interests at stake.2
I.
The effect of my colleagues’ view that these defendants cannot be detained prior to trial on grounds of dangerousness is effectively to emasculate the heart of the Bail Reform Act. Congress intended the Act to apply to this type of case involving very serious offenses and involving defendants whose predictable future criminal conduct cannot be held in check by imposing conditions on pretrial release. In order to place the Fifth Amendment issue in proper perspective, it may be helpful to describe the activities of those two appellants who, under my colleagues’ view, would be released pending trial — despite the detailed findings of the magistrate and the district court that they should be detained.
Camacho-Negron
The magistrate and the district court enumerated the following reasons for detaining Camacho-Negron. He was a salaried member of Los Macheteros, the terrorist organization involved in sophisticated para-military activities. He had been “approved” for military action, operated under a code-name, and participated in the January 1981 attack on the Muniz Air Base in Puerto Rico where nine aircraft were destroyed. Los Macheteros claimed credit for this attack as well as other terrorist acts involving murder, robbery, and destruction of property. Camacho-Negron had financial dealings with other members of the organization. Evidence tied him to the transportation of the Wells Fargo armed robbery proceeds from upstate New York to Mexico. He had a pistol at the time of his arrest and was equipped to travel to foreign countries.
*1011 Gonzales-Claudio
Gonzales-Claudio had been a longtime member of Los Macheteros. He had a salaried position in the organization, was entrusted with weaponry, and used a codename. Evidence linked him to the December 1979 attack on a Navy bus in which two sailors were killed and nine others wounded, and to the January 1981 attack on the Muniz Air Base. He was involved in smuggling explosives into Puerto Rico from Mexico, and in transporting Wells Fargo armed robbery proceeds from the United States to Mexico. He flaunted the Wells Fargo armed robbery by participating in the gift give-away program and in contacting the press. At the time of his arrest, he resisted and the agents uncovered a pistol. He had an “ability and willingness to use violence against innocent persons” according to the magistrate and the district court.
Both appellants are members of a well-organized terrorist group that has a network available for safe-houses, weapons storage and travel. I agree that the district court on remand should consider whether these two appellants who were found to pose a danger to the community also pose a risk of flight.
II.
I do not agree with my colleagues that the Fifth Amendment prohibits the pretrial detention of the two appellants who are being detained only on the ground of danger to the community (Camacho-Negron and Gonzales-Claudio). I shall first address Judge Newman’s view that even a short period of detention in such a case violates the due process clause of the Fifth Amendment. I shall then turn to Judge Feinberg’s view that the length of detention in this case results in punishment before an adjudication of guilt, in violation of the due process clause.3
(A)
Judge Newman would hold as unconstitutional under the due process clause of the Fifth Amendment even a brief period of pretrial detention based on a defendant’s danger to the community. I do not agree that such detention fails to comport with the Fifth Amendment concept of fundamental fairness. In my view, the constitutional parameters set forth by the Supreme Court, the legislative history of the Bail Reform Act, and the procedural safeguards provided by the Act permit pretrial detention where a defendant poses a serious risk to the safety of the community.
Bell v. Wolfish, 441 U.S. 520, 534 n. 15 (1979), left open the question whether a government objective other than ensuring a defendant’s presence at trial may justify pretrial detention under the due process clause. Since then, the detention authorized by the New York Family Court Act of juveniles likely to commit crimes was upheld in Schall v. Martin, 467 U.S. 253 (1984). As my colleagues point out, the Supreme Court in Schall recognized the state’s role as parens patriae and the seventeen-day detention limitation in that decision. But in analyzing the substantive component of due process — whether the legislation is compatible with fundamental fairness — the Schall Court first questioned whether the statute “serve[s] a legitimate state objective” and, second, whether the procedural safeguards are adequate. 467 U.S. at 263-64. The Court then stated:
“ ‘A court must decide whether the disability is imposed for the purpose of punishment or whether it is but an incident of some other legitimate governmental purpose.’ Absent a showing of an express intent to punish on the part of the State, that determination generally will turn on ‘whether an alternative purpose *1012to which [the restriction] may rationally be connected is assignable for it, and whether it appears excessive in relation to the alternative purpose assigned [to it].’ ”
467 U.S. at 269 (quoting Bell v. Wolfish, 441 U.S. 520, 588 (1979); Kennedy v. Mendoza-Martinez, 372 U.S. 144, 168-69 (1963)). Whether we analyze the interest of the government under a “legitimate” purpose test or under a “weighty public interest” test, see Schall, supra, 467 U.S. at 291 (Marshall, J., dissenting), I believe that the government has sustained its burden in the instant case. The government has a “legitimate and compelling interest” in protecting the community from crime. Schall, supra, 467 U.S. at 264 (quoting De Veau v. Braisted, 363 U.S. 144, 155 (1960)).
When Congress enacted the Bail Reform Act, it responded conscientiously to the disturbing rate of recidivism among released defendants4 and the need to give as much consideration to the danger a defendant may pose in the pretrial release determination “as the likelihood that he will not appear for trial.” S.Rep. No. 225, 98th Cong., 2d Sess. 6, reprinted in 1984 U.S. Code Cong. & Ad.News 3182, 3189. Congress also recognized that the prior law of bail resulted in disparate treatment among defendants based on ability to post bond. In discussing pretrial release determinations prior to the 1984 Act, the Committee observed:
“If a court believes that a defendant poses such a [grave] danger, it faces a dilemma — either it can release the defendant prior to trial despite these fears, or it can find a reason, such as risk of flight, to detain the defendant (usually by imposing a high money bond). In the Committee’s view, it is intolerable that the law denies the judges the tools to make honest and appropriate decisions regarding the release of such defendants.”
S.Rep. No. 225, 98th Cong., 2d Sess. 5, reprinted in 1984 U.S.Code Cong. & Ad. News 3182, 3188. In my view, removing from a court’s options the right to detain based on dangerousness only invites a court to detain a dangerous person on the ground of risk of flight. Unlike the “clear and convincing” evidence standard applied in detaining a person on the ground of dangerousness, the government need establish a risk of flight by only a preponderance of the evidence. Congress clearly intended that judicial officers should consider separately whether an individual is a danger to the community or whether he poses a risk of flight. Judge Newman’s proposed holding, I fear, would resurrect the very predicament which Congress attempted to halt.
In view of the strong interest in preventing reasonably predictable criminal activity by defendants charged with serious crimes, and the non-punitive nature of the detention, the remaining issue is whether the procedures afforded the detainees provide sufficient protection against erroneous and unnecessary deprivations of liberty. See Schall, supra, 467 U.S. at 274; Mathews v. Eldridge, 424 U.S. 319, 335 (1976).
The Bail Reform Act provides significant mechanisms and protections against unjustified restraints. The statute is far more specific than the juvenile detention statute addressed in Schall. Detention under the Act does not apply to all defendants charged with serious crimes. Rather, the Act is intended to apply to “a small but identifiable group of particularly dangerous defendants as to whom neither the imposition of stringent release conditions *1013nor the prospect of revocation of release can reasonably assure the safety of the community or other persons. It is with respect to this limited group of offenders that the courts must be given the power to deny release pending trial.” S.Rep. No. 225, 98th Cong., 2d Sess. 6-7, reprinted in 1984 U.S.Code Cong. & Ad.News 3182, 3189. The judicial officer must be convinced that conditions on release are not adequate. A hearing on this issue must be held. 18 U.S.C. § 3142(e), (f). Of course, probable cause to believe that the accused committed the crime must be established. Gerstein v. Pugh, 420 U.S. 103, 114 (1975); Morrissey v. Brewer, 408 U.S. 471, 489 (1972). The determination that no conditions of release will protect adequately the public must be supported by clear and convincing evidence, 18 U.S.C. § 3142(f), whereas risk of flight need only be shown by a preponderance of evidence. This is not an ad hoc determination. The judicial officer must consider the factors set forth in § 3142(g), including the seriousness of the charge and its violent nature, the weight of the evidence, the history and characteristics of the accused, and the nature and seriousness of the danger posed to the community. There must be an evidentiary basis for the facts supporting a judicial officer’s conclusion that pretrial detention is necessary. Moreover, the defendant at a detention hearing has the right to counsel or appointed counsel if he cannot afford an attorney, the right to present witnesses on his behalf, the right to cross-examine witnesses and the right to present information by proffer or otherwise. 18 U.S.C. § 3142(f). In addition, subsection (i) reinforces the deliberation requirement imposed on the judicial officer and aids the process of appellate review. The detention order must include written findings of fact and the specific reasons for detention. As further protections against restraints on liberty, the statute requires that, to the extent possible, pretrial detainees must be separated from those persons who are serving their sentences after conviction. A detainee also must be afforded a reasonable opportunity to consult with counsel. Finally, a detainee is entitled to prompt appellate review of a detention order. 18 U.S.C. § 3145(b).5
These effective procedural safeguards, combined with the strong government interest at stake, in my view, support the constitutionality of pretrial detention based on a defendant’s proven danger to the community.
(B)
I also disagree with Judge Feinberg’s view that the length of detention in this case violates the due process rights of those appellants being held on the ground of danger to the community.
Not only must pretrial detention serve a legitimate government interest to satisfy due process — a burden which I believe the government has sustained — but it is clear that there can be no punishment prior to an adjudication of guilt. Bell v. Wolfish, supra, 441 U.S. at 535. Restraints on liberty are not necessarily punitive. In my view, the restraints imposed here are required by the extraordinary need to protect the public from especially dangerous individuals. The Bail Reform Act on its face states no intent to punish pretrial detainees. The legislative history of the Act is devoid of any intention to punish. Congress did not seek to foster traditional goals of punishment such as retribution, deterrence, or rehabilitation. Congress’ sole concern was the safety of the community. Pretrial detention does not pun*1014ish past conduct but rather protects the community from predictable future conduct. United States v. Edwards, 430 A.2d 1321, 1332 (D.C.App.1981) (en banc), cert. denied, 455 U.S. 1002 (1982); United States v. Hazzard, 598 F.Supp. 1442, 1451 (N.D.Ill.1984). Congress carefully incorporated the analysis of Edwards in the Bail Reform Act. The nonpunitive intent is further illustrated by Congress’ mandate that pretrial detainees, to the extent possible, be separated from those persons who are serving sentences after a conviction. 18 U.S.C. § 3142(i)(2). I agree with those courts which have held that such pretrial detention based on dangerousness is not punishment. See, e.g., United States v. Portes, 786 F.2d 758, 767 (7 Cir.1986); United States v. Maull, 773 F.2d 1479, 1485 (8 Cir.1985); United States v. Kouyoumdjian, 601 F.Supp. 1506, 1511 (C.D.Cal.1985); United States v. Hazzard, supra, 598 F.Supp. at 1451.
I also disagree with Judge Feinberg’s view that passage of time alone transforms what initially was regulatory detention into punishment, resulting in a due process violation. See also United States v. Colombo, 777 F.2d 96, 101 (2 Cir.1985) (at some point, the length of detention “might” violate due process).6 While Judge Feinberg applies the specific factors set forth in Kennedy v. Mendoza-Martinez, 372 U.S. 144, 168-69 (1963), more recently the Supreme Court in Schall focused on whether there was an express intent to punish, whether an alternative purpose could rationally be assignable to the restraint on liberty, and whether it is excessive. Schall, supra, 467 U.S. at 269. Congressional intent and the importance of preventing reasonably predictable future criminal conduct7 do not change with the passage of time. Although the length of the restraint on liberty increases with time, the form and motivation for restraint do not change with passage of time. No alternative purpose comes into play.
A determination of the constitutionality of pretrial detention based on dangerousness, moreover, should not be determined only on the basis of time as it affects the due process prohibition against punishment before an adjudication of guilt. The protections against unnecessary infringements on liberty also must include an examination of the Speedy Trial Act, 18 U.S.C. §§ 3161-3174, and the Sixth Amendment right to a speedy trial. Delays in bringing a defendant to trial may be the result of a variety of factors. The reasons for the delay must be examined. It does not seem logical to me that a defendant whose perhaps unnecessary pretrial motions result in excludable time under the Speedy Trial Act, see 18 U.S.C. §§ 3164, 3161(h), can transform regulatory detention into punishment and therefore demand his own release. It may be that either the defendant, or the government, or both, has increased the length of time before trial. An analysis of the validity of the detention should consider whether the defendant has moved for a severance *1015or an expedited trial, Colombo, supra, 777 F.2d at 101, the extent of and necessity for pretrial motions made on behalf of the government and the defendant, and whether one side’s strategy has added needlessly to the complexity of the case. United States v. Accetturo, 183 F.2d 382, 388 (3 Cir.1986).
While a defendant cannot be detained indefinitely prior to an adjudication of guilt, the Supreme Court has recognized that due process necessarily is a flexible concept. Gerstein v. Pugh, supra, 420 U.S. at 123. And Congress intended to incorporate this flexibility in the Bail Reform Act by requiring judicial officers to analyze the lawfulness of pretrial detention on a case by case basis. S.Rep. No. 225, 98th Cong., 2d Sess. 19, reprinted in 1984 U.S. Code Cong. & Ad.News 3182, 3201-02. An appropriate analysis therefore must include such factors as the seriousness of the crime with which the defendant is charged, the strength of the government’s proof as to the individual’s danger to the community, and the strength of the government’s case on the merits. A judicial officer must have the tools to detain a defendant who, for one reason or another, may not pose a risk of flight, but who is so likely to commit crimes if released that no conditions imposed on release are adequate to protect the safety of the community. As Judge Feinberg states, Congress may not have envisioned the length of time a defendant could be detained pending trial when it analyzed the interrelationship between the Speedy Trial Act and the Bail Reform Act. Nevertheless, that does not relieve the courts of the duty to monitor closely motions for extensions and excludable time.
III.
To summarize:
I would uphold pretrial detention on the ground of a defendant’s reasonably predictable danger to the community in the case where no conditions of release can protect the public adequately. The government interest in protecting the community is compelling. The carefully drafted procedural requirements in the Bail Reform Act militate against unjustified encroachment on the liberty interests at stake. Further, Congress expressed no intent in the Bail Reform Act to punish past behavior. I do not agree that passage of time alone transforms regulatory detention into punishment. In my view, an analysis of the lawfulness of pretrial detention necessarily must include an examination of the Sixth Amendment right to a speedy trial, the Speedy Trial Act, and the reasons for delay. The reasonably predictable danger a detainee poses to the community surely does not decrease with time. Efforts should be directed at bringing such a defendant to trial and monitoring closely the reasons for delay.
I would affirm the district court’s detention orders of appellants Camacho-Negron and Gonzales-Claudio on the ground of their danger to the community and the ineffectiveness of conditions imposed on release. From the refusal of Judges Newman and Feinberg — in their separate opinions — to do so, I respectfully but emphatically dissent.
. United States v. Basic, 592 F.2d 13, 37, (2 Cir.1978) (concurring opinion).
. This dissent is confined to my disagreement with Judge Newman and Judge Feinberg in holding that the Fifth Amendment due process rights of appellants Camacho-Negron an3 Gonzales-Claudio were violated. There is no dispute with respect to the other six appellants. I concur in the judgment of the Court affirming their detention orders.
I also agree with my colleagues’ rejection of appellants’ procedural challenges to the application of the Bail Reform Act and with my colleagues’ view that the Eighth Amendment right to bail provision is not determinative of the critical issue on this appeal.
. As must be evident, neither the view of Judge Newman nor that of Judge Feinberg, as summarized in this paragraph, represents a majority view. Each stands alone, unsupported by each other, and of course unsupported by the author of this dissenting opinion. This unfortunately will present a dilemma to district judges who will not know what the law of the Circuit is— until the Supreme Court is able to speak to the issues.
. Congress reported that “[i]n a recent study of release practices in eight jurisdictions, approximately one out of every six defendants in the sample study were rearrested during the pretrial period — one-third of these defendants were rearrested more than once, and some were rearrested as many as four times." S.Rep. No. 225, 98th Cong., 2d Sess. 6, reprinted in 1984 U.S. Code Cong. & Ad.News 3182, 3189 & n. 15 (citing Lazar Institute, Pretrial Release: An Evaluation of Defendant Outcomes and Program Impact 48 (Wash.D.C. Aug. 1981)).
. These procedural safeguards, including an individual determination of probable cause to believe the defendant has committed a serious crime, an individual detention hearing, the right to counsel, the right to present evidence, the right to cross-examine witnesses, and the burden of “clear and convincing evidence” imposed on the government, were not present in the case of the internment of the Japanese-Americans during World War II. Judge Newman’s allusion to Korematsu v. United States, 323 U.S. 214 (1944), fails to recognize these significant distinctions. While Korematsu indeed may be a regrettable blemish in the history of American jurisprudence, obviously the form of detention authorized under the Bail Reform Act, with its attendant procedural protections, is not such a departure from American ideals of individual liberty.
. I think it is significant to note that in Colombo, at the time of argument, appellant had been detained on the ground of danger to the community for seven months — nearly the same time as appellants Camacho-Negron and Gonzales-Claudio have been detained on the same ground in the instant case. And yet the Colombo court declined to reach the constitutional question regarding the length of Colombo’s detention because "at the time of oral argument, Colombo had not availed himself of the opportunity to move to expedite his trial or for a severance." United States v. Colombo, supra, 777 F.2d at 101. For aught that appears in the record before us in the instant case, none of appellants being detained on the ground of danger to the community has moved to expedite the trial or for a severance.
. The Schall Court recognized the ability of a judicial officer to make a reasonable prediction of future criminal conduct:
"Our cases indicate, however, that from a legal point of view there is nothing inherently unattainable about a prediction of future criminal conduct. Such a judgment forms an important element in many decisions, and we have specifically rejected the contention, based on the same sort of sociological data ... that it is impossible to predict future behavior and that the question is so vague as to be meaningless.”
467 U.S. at 278-79 (footnote omitted; citations omitted).