MEMORANDUM OPINION AND ORDER
The labyrinthine 305-page, 175-count indictment in this case, nearly two inches thick and weighing almost four pounds, names thirty-eight defendants, thirty-seven of whom are alleged to have been members or associates of the El Rukns, an infamous Chicago street gang. It details a maze of well over 250 factually separate criminal acts committed from 1966 to 1989 in many different locations and, for each act, alleges the participation of varying combinations of defendants and countless unindict-ed co-conspirators. The government’s justification for including many of these otherwise unconnected criminal events in one indictment and one trial is that each was allegedly committed to attain power, control, and wealth for the street gang.
Defendants Jackie Clay, Harry Evans, Henry Leon Harris, Earl Hawkins, Eugene Hunter, Derrick Kees, Anthony Sumner, *1165Freddie Elwood Sweeney, and Ricky Dean Williams have all pleaded guilty. Further guilty pleas are not anticipated. Defendants Roger Bowman, Floyd Davis, Eddie Franklin, Bernard Green, Melvin Mayes, Walter Pollard, and Edward Williams are presently fugitives and may not be apprehended in time for trial. This means that at least twenty-two defendants, but as many as twenty-nine, will be going to trial. Several defendants have filed motions to sever this indictment pursuant to Rule 8(b) or, in the alternative, Rule 14 of the Federal Rules of Criminal Procedure, and the issue before the Court is whether we should permit these 175 diverse charges to be tried in a single mega-trial of unprecedented projected duration. We believe that we should not. Therefore, for the reasons discussed below, although we deny the Rule 8(b) motions, the Rule 14 motions are granted.1
I. Indictment
A. Count One-RICO Conspiracy
Count One of the indictment charges all but one of the defendants2 with conspiracy to violate the Racketeering Influenced and Corrupt Organizations Act (“RICO”), 18 U.S.C. § 1962(d) (1988), and alleges that the El Rukn organization is a racketeering “enterprise” as defined by 18 U.S.C. § 1961(4).3 It describes a cohesive organization with tightly controlled operations and a formal chain of command. As alleged, unindicted co-conspirator Jeff Fort masterminded the activities of the El Rukns and wielded ultimate and unquestioned authority. He was assisted by thirty-five defendants whom the government contends were El Rukn “generals” or “officers,” the organization’s second and third levels of command, respectively. The government contends that the remaining two defendants named in Count One, although not El Rukn members, were otherwise intimately associated with the organization.
Under the direction and control of Fort, the named defendants and other unindicted co-conspirators are alleged to have conducted El Rukn affairs through the commission of an astonishing number of racketeering acts, including at least twenty murders, twelve attempted murders, eleven conspiracies to murder, one act of kidnapping, wide-scale drug trafficking, and numerous acts of obstruction of justice, including one attempt to bribe a judge and several acts of witness intimidation, retaliation, and tampering. As alleged, all of these wide-ranging and diverse offenses are connected, although at times somewhat loosely, to the affairs of the El Rukn street gang.
B. Count Two-Substantive RICO
Count Two charges thirty-six defendants with substantive violations of RICO under 18 U.S.C. § 1962(c)4 and recounts 128 separate acts of racketeering, many of which are also alleged in Count One.5 The number of racketeering acts that each defendant is alleged to have committed ranges from as many as seventy for Melvin Mayes to only two for Isiah Kitchen. The nature of acts charged also varies. Some defendants are charged with numerous violent racketeering acts and only one or two narcotics-related acts, while other defendants are charged solely with narcotics-related acts.
1. Violent Racketeering Acts
According to the allegations in Count Two, the long string of violent racketeering *1166acts began in May 1974, when several defendants murdered Gilbert Connors, the brother of a rival drug dealer, to prevent encroachment on the El Rukn drug trade. To facilitate a cover-up of this crime, the same defendants subsequently killed Gregory Freeman on May 22, 1974, and blamed this murder on a witness to the Connors homicide. As a result, El Rukn members were able to persuade the witness to refuse to testify and the charges in both the Connors and Freeman murder cases were subsequently dropped.
Other murders were committed to enforce and enhance Jeff Fort’s control over the El Rukns. Fort ordered El Rukn members, including several defendants, to kill disloyal members and former El Rukn leaders, including Willie McLilly, Roy Love, and Mickey Cogwell. McLilly and Love were killed on November 29, 1974, and Cogwell was killed on February 25, 1977.
The El Rukns also went to extreme lengths to protect its members from criminal prosecution. In August 1977, eighteen defendants conspired to kill Audrina Thomas to prevent her from testifying against two El Rukn generals charged with murder. Then, in a case of mistaken identity, Thomas’ sister, Rowena James, was shot and killed on September 1, 1977. In January of 1987, four defendants and other El Rukn members kidnapped Patricia McKinley in an attempt to prevent her from testifying against an El Rukn general charged with the murder of Maurice Coleman.
The El Rukns committed a slew of other murders and attempted murders related to narcotics activity. On March 12, 1980, pursuant to Fort’s order, two defendants killed Douglas Ellison because of a dispute involving a narcotics debt. Three months later, eight defendants conspired to murder Lemont Timberlake because he failed to accede to demands to stop dealing narcotics in El Rukn-controlled territory. This plan was carried out on June 17, 1980, when Timberlake was shot and killed in a vacant lot on the south side of Chicago.
From about April 1981 to January 1983, as a result of a drug territorial dispute, twenty-three defendants conspired to kill the top leaders and members of the Titanic Stones, a rival gang, including Eugene Hairston, George Thomas, Barnett Hall, Ray East, Robert East, and Willie Bibbs. During that same time period, Willie Bibbs and Barnett Hall were killed and George Thomas was the target of an attempted murder. In the course of the Thomas attempt, on January 23,1983, five defendants and other gang members killed Charmaine Nathan and shot and attempted to kill Sheila Jackson.
In April 1985, as a result of another drug territorial dispute with a rival gang, twenty-three defendants conspired to kill various members of the King Cobra street gang. In furtherance of this conspiracy, during the last week of April, teams of El Rukns cruised the streets of Chicago in efforts to locate and murder King Cobra members. During this time, several defendants and other El Rukn members murdered Robert Jackson, Rico Chalmers, Glendon McKinley, and Vicki Nolden and attempted to murder Theotis Clark and Andre Chalmers. McKinley and Nolden were innocent bystanders swept up in the wave of violence.
The El Rukns committed other murders and attempted murders as a result of separate drug territorial disputes. On January 7, 1982, Bruce Davis, a rival drug dealer, and his wife Sheila were each shot about six times because Davis had ignored warnings to discontinue narcotics sales in an El Rukn-controlled territory. On February 13, 1983, another rival drug dealer, Chal-mers Tyler, was killed under similar circumstances. In a third incident, on April 28, 1984, the El Rukns killed Jerome Smith and Taiman Hickman. Smith, the leader of a rival gang, had ignored similar warnings regarding narcotics sales in El Rukn-con-trolled territory.
The El Rukns also murdered and attempted to murder rival gang members in retaliation for shootings of El Rukn members. Fort ordered the murders of Melvin and Jerry Ewing in retaliation for the 1981 murder of Carl Taylor, a former El Rukn general. El Rukn defendants subsequently shot and attempted to murder Jerry Little *1167and Phillip Steele on February 23, 1983, in the mistaken belief that they were the Ewing brothers. Later that year, on June 21, in a conspiracy involving nineteen defendants, Ronald Bell was murdered in retaliation for his earlier attempt to murder El Rukn Thomas Pearson.
During 1985 and 1986, the El Rukns conspired to and committed interstate murder-for-hire on behalf of defendant Noah Robinson. Robinson hired the El Rukns to murder Robert Aulston, a former business partner with whom Robinson was having a legal dispute. In exchange for $10,000, several defendants travelled to Texas and attempted to murder Aulston in November 1985. Approximately one month later, Robinson again solicited the El Rukns to commit murder. This time the target was Leroy Barber, for whose murder Robinson was willing to pay $10,000. At the direction of Fort, two groups of defendants travelled to Greenville, South Carolina in attempts to locate and kill Barber. Then, on January 3, 1985, one of these groups found Barber and murdered him. Nearly two years later, in response to a grand jury investigation of this murder, Robinson hired defendant Sweeney to murder Janice Denise Rosemond, a grand jury witness who had been an eyewitness to the Barber murder. On December 4, 1987, in consideration for $5,000, Sweeney stabbed and attempted to murder Rosemond.
In other efforts to cover up the Barber murder, Robinson and other defendants attempted to prevent defendant Harris from cooperating with authorities and providing details of the murder and other El Rukn affairs. On September 30, 1987, Robinson offered to pay Harris $15,000 to provide false information to the authorities concerning Robinson’s involvement in the Barber murder and El Rukn drug trafficking. Later, on October 9, 1987, two defendants, along with other El Rukn members, fired shots into an establishment called Hank’s Fun House Tap in Milwaukee, Wisconsin, which was owned by Harris’ father, in order to intimidate Harris and prevent him from providing information to the authorities. Less than a year later, on two separate occasions in July 1988, Robinson solicited other defendants to murder Harris. Later that same month, in a final effort to persuade Harris to remain silent, Robinson offered to pay Harris to leave the country.
2. Non-Violent Racketeering Acts
Count Two alleges ninety-seven additional racketeering acts which, with one exception, are narcotics-related offenses. The only exception, designated Racketeering Act 128, charges defendants Melvin Mayes and Alan Knox with using the telephone to facilitate the acquisition of an M-72 series Light Anti-Tank Weapon, commonly referred to as a LAW rocket, in violation of 18 U.S.C. § 1952(a)(3) (1988). The narcotics-related offenses alleged in Count Two include the possession and distribution of narcotics in violation of 21 U.S.C. § 841(a)(1) (1988), the use of a communication facility to further various narcotics offenses in violation of 21 U.S.C. § 843(b) (1988), and interstate travel with intent to further narcotics offenses in violation of 18 U.S.C. § 1952 (1988). A significant majority of these offenses each involve only one or two defendants. Notably, however, Racketeering Act 31 charges thirty-seven defendants with a narcotics conspiracy in violation of 21 U.S.C. § 846 (1988).
C. Count Three-Narcotics Conspiracy
Racketeering Act 31 is repeated as a separate “substantive” offense in Count Three. This Count describes a conspiracy to possess with intent to distribute, and to distribute, multi-kilogram quantities of heroin and cocaine, hundreds of pounds of marijuana, thousands of amphetamine pills, thousands of Talwin and Triplenamin pills, multi-liter quantities of codeine syrup, and large quantities of Phencyclidine (PCP). The government contends that, on a regular basis during the course of this conspiracy, from 1966 to 1989, various defendants purchased narcotics from sources in at least seven states, including Florida, South Carolina, Mississippi, New York, Wisconsin, Illinois, and Michigan. The government further contends that narcotics the defendants sold narcotics from numerous El Rukn-controlled buildings and territories throughout the Illinois cities of Chicago, *1168Evanston, Skokie, and Harvey, and also from Milwaukee, Wisconsin.
D. The Other Counts
With the exception of Counts One Hundred Sixty-Nine through One Hundred Seventy-Four, which charge certain defendants with firearms offenses in violation of 18 U.S.C. §§ 2, 844(d), 924(c), 922(g)(1), and 18 U.S.C.App. 1202(a)(1) (1988), the other 172 counts of the indictment reallege selected acts of racketeering from Count Two as corresponding “substantive” criminal violations. The large majority of the acts realleged are the narcotics-related offenses. Only fourteen of the remaining substantive counts correspond to violent racketeering acts, and these represent only a small portion of the violent crimes alleged in Count Two. Counts Four through Seven correspond to the King Cobra conspiracy and the related murders and attempted murders. Counts Eight through Sixteen correspond to the crimes connected to Robinson, including the attempted murder of Robert Aulston, the murder of Leroy Barber, and the related acts of witness tampering. The last violent substantive count and the final count in the indictment is Count One Hundred Seventy-Five, which corresponds to the McKinley kidnapping.
II. Severance Discussion
Numerous defendants have filed motions to sever this indictment claiming that they are improperly joined pursuant to Rule 8(b), or, in the alternative, that joinder is unduly prejudicial pursuant to Rule 14. We will discuss these motions in turn.
A. Rule 8(b)
Rule 8(b) permits joinder of an unlimited number of defendants “if they are alleged to have participated in the same act or transaction or in the same series of acts or transactions constituting an offense or offenses.” Fed.R.Crim.P. 8(b). In determining whether joinder is proper under this Rule, a court must look to the face of the indictment. United States v. Bruun, 809 F.2d 397, 406 (7th Cir.1987); United States v. Velasquez, 772 F.2d 1348, 1354 (7th Cir.1985), cert. denied, 475 U.S. 1021, 106 S.Ct. 1211, 89 L.Ed.2d 323 (1986). If the indictment alleges a conspiracy, it is well established that all members of that conspiracy are properly joined. United States v. Garner, 837 F.2d 1404, 1412 (7th Cir.1987), cert. denied, 486 U.S. 1035, 108 S.Ct. 2022, 100 L.Ed.2d 608 (1988); United States v. Dounias, 777 F.2d 346, 348 (7th Cir.1985).
Although the defendants agree with this settled principal, they nonetheless argue that they are improperly joined. Relying on Kotteakos v. United States, 328 U.S. 750, 66 S.Ct. 1239, 90 L.Ed. 1557 (1946), they argue that the indictment fails to satisfy Rule 8(b) because it alleges multiple distinct conspiracies, including separate conspiracies to murder, obstruct justice, and distribute narcotics. In Kotteakos, the Supreme Court reversed the convictions of the petitioners where the evidence established at least eight separate conspiracies connected only by a common participant. Id. at 773, 66 S.Ct. at 1252. The Court stated that a defendant has a right “not to be tried en masse for the conglomeration of distinct and separate offenses committed by others_” Id. at 775, 66 S.Ct. at 1253. Thus, according to the defendants here, Kotteakos requires a finding that they are improperly joined.
While their characterization of the indictment is correct, the defendants are wrong that the allegation of separate conspiracies renders joinder improper under Rule 8(b). With the advent of RICO, Congress significantly broadened the scope of the government’s authority to bring defendants together in one indictment. It conferred this broad authority without eviscerating the principals set forth in Kottea-kos or “radically altering] traditional conspiracy doctrine_” United States v. Riccobene, 709 F.2d 214, 224 (3d Cir.), cert. denied, 464 U.S. 849, 104 S.Ct. 157, 78 L.Ed.2d 145 (1983). Instead, Congress simply outlawed a particular conspiratorial agreement, the object of which could include the commission of a wide array of separate and distinct offenses. Section 1962(d) of the RICO statute proscribes agreements “to conduct or participate in the affairs of an enterprise through a pattern of racketeering activity.” United States v. Neapolitan, 791 F.2d 489, 498 *1169(7th Cir.), cert. denied, 479 U.S. 940, 107 S.Ct. 422, 93 L.Ed.2d 372 (1986). A single “pattern of racketeering activity” can include numerous distinct conspiracies. Riccobene, 709 F.2d at 224-25; United States v. Sutherland, 656 F.2d 1181, 1192 (5th Cir.1981), cert. denied, 455 U.S. 949, 102 5.Ct. 1451, 71 L.Ed.2d 663 (1982). Consequently, “a series of agreements that under pre-RICO law would constitute multiple conspiracies could under RICO be tried as a single ‘enterprise’ conspiracy” in violation of § 1962(d).6 Sutherland, 656 F.2d at 1192; see also Neapolitan, 791 F.2d at 496 n. 3, 501; Riccobene, 709 F.2d at 224-25; United States v. Castellano, 610 F.Supp. 1359, 1396 (S.D.N.Y.1985); cf. United States v. Napue, 834 F.2d 1311, 1332 (7th Cir.1987) (if separate agreements “represent stages or different functions to be performed in the formulation of a larger scheme, the object of which is to effectuate a single unlawful result, then there is a single conspiracy”). Therefore, consistent with the rationale of Kotteakos, the defendants in this case are properly joined because they are each alleged members of a single and unifying RICO conspiracy. The separate and distinct underlying conspiracies are simply part of the “same series of acts or transactions constituting [this] of-fense_” Fed.R.Crim.P. 8(b).
Some defendants argue in the alternative that joinder is improper because the indictment charges certain defendants with substantially more conspiratorial ae-tivity than others.7 This argument has no merit. According to the Seventh Circuit in United States v. Whaley:
Various people knowingly joining together in furtherance of a common design or purpose constitute a single conspiracy. While the conspiracy may have a small group of core conspirators, other parties who knowingly participate with these core conspirators and others to achieve a common goal may be members of an overall conspiracy.
830 F.2d 1469, 1474 (7th Cir.1987) (quoting United States v. Varelli, 407 F.2d 735, 742 (7th Cir.1969), cert. denied, 405 U.S. 1040, 92 S.Ct. 1311, 31 L.Ed.2d 581 (1972)), cert. denied, 486 U.S. 1009, 108 S.Ct. 1738, 100 L.Ed.2d 202 (1988). Moreover, mere ignorance of the identities of all of the conspirators or of all the details of the conspiracy is immaterial. United States v. Fitzgerald, 579 F.2d 1014, 1018 (7th Cir.), cert. denied, 439 U.S. 1002, 99 S.Ct. 610, 58 L.Ed.2d 677 (1978); United States v. Arvanitis, 676 F.Supp. 840, 844 (N.D.Ill.1987), affd sub nom. United States v. Gaitanis, 902 F.2d 37 (7th Cir.1990). Thus, regardless of the alleged extent of any one defendant’s involvement, all of the defendants in this case are properly joined for purposes of Rule 8(b) because they each allegedly agreed to the “overall objective,” namely to conduct or participate in the affairs of the so called “El Rukn Nation” through a pattern of racketeering.8 Indeed, the govern*1170ment claims that each of the moving defendants willingly held a formal leadership position, either as a “general” or “officer,” and thus “operated and promoted the [racketeering] activities of the enterprise, and exercised authority over subordinate members.” We conclude that these defendants are properly joined. Accordingly, their motions to sever pursuant to Rule 8(b) are denied.9
B. Rule lip
Having decided that the requirements of Rule 8(b) are met by the allegations in the indictment, we turn to Rule 14. As indicated above, although join-der is technically proper under Rule 8(b), Rule 14 authorizes severance if such join-der is prejudicial to the defendants. United States v. Lane, 474 U.S. 438, 447, 106 S.Ct. 725, 731, 88 L.Ed.2d 814 (1986); Garner, 837 F.2d at 1413. According to Rule 14:
If it appears that a defendant or the government is prejudiced by a joinder of offenses or of defendants in an indictment or information or by such joinder for trial together, the court may order an election or separate trials of counts, grant the severance of a defendant or provide whatever relief justice requires.
Fed.R.Crim.P. 14; see also ABA Standards for Criminal Justice, Joinder and Severance § 13-3.1(b)(1) (1980) (severance should be granted whenever it “is deemed to promote a fair determination of a defendant’s guilt or innocence of each offense”). This Rule must be read against the backdrop of Rule 2, which provides that the Federal Rules of Criminal Procedure “are intended to provide for the just determination of every criminal proceeding. They shall be construed to secure simplicity in procedure, fairness in administration and the elimination of unjustifiable expense and delay.” Fed.R.Crim.P. 2.
The District Court is given wide latitude in determining whether a Rule 14 severance is appropriate. The decision is within its “sound discretion” and “will not be overturned absent a clear showing of an abuse of that discretion.” United States v. Caliendo, 910 F.2d 429, 437 (7th Cir.1990); see also United States v. McAnderson, 914 F.2d 934, 948 (7th Cir.1990); United States v. Penson, 896 F.2d 1087, 1094 (7th Cir.1990). This deference is due in part to the fact that the “balancing of the cost of conducting separate trials and the possible prejudice inherent in a single trial is best conducted by the trial court_” Caliendo, 910 F.2d at 437 (quoting United States v. Moya-Gomez, 860 F.2d 706, 754 (7th Cir.1988), cert. denied, — U.S. -, 109 S.Ct. 3221, 106 L.Ed.2d 571 (1989)). To determine if severance is appropriate in the present case, we must weigh the public interest in a joint trial of the twenty-two to twenty-nine defendants against the possibility of undue prejudice or confusion arising from such a trial. See United States v. *1171Zanin, 831 F.2d 740, 744 (7th Cir.1987); United States v. Rivera, 825 F.2d 152, 159 (7th Cir.1987), cert. denied, 484 U.S. 979, 108 S.Ct. 494, 98 L.Ed.2d 492 (1987); United States v. Racente, 503 F.2d 543, 549 n. 8 (7th Cir.) (must balance “possible prejudice” against “inconvenience and expense to the Government, the court and the jurors”), cert. denied, 419 U.S. 1048, 95 S.Ct. 623, 42 L.Ed.2d 642 (1974); see also 8 R. Cipes, I. Hall & M. Waxner, Moore’s Federal Practice ¶ 14.02(1), at 14-5 (2d ed. 1990).
1. Public Interest
a. Advantages of Joint Trial
To strike the appropriate balance, we recognize, as we must, the oft-cited “strong public interest in having persons jointly indicted tried together, especially where the evidence against the defendants arose out of the same acts or series of acts.” United States v. Turk, 870 F.2d 1304, 1306 (7th Cir.1989) (quoting United States v. Oxford, 735 F.2d 276, 280 (7th Cir. 1984)); see also Caliendo, 910 F.2d at 437. Joint trials generally “reduce the expenditure of judicial and prosecutorial time [and] ... the claims the criminal justice system makes on witnesses, who need not return to court for additional trials.” United States v. Buljubasic, 808 F.2d 1260, 1263 (7th Cir.), cert. denied, 484 U.S. 815, 108 S.Ct. 67, 98 L.Ed.2d 31 (1987). For this reason, joint trials are considered “an essential element of the quick administration of justice.” United States v. Walters, 913 F.2d 388, 393 (7th Cir.1990). Certainly, “[i]f every defendant who wanted a severance was given one, the slow pace of our court system would go from a crawl to paralysis.... ” Id.
Joint trials are also favored because of their purported effect on the accurate determination of culpability. According to the court in Buljubasic, joint trials
reduce the chance that each defendant will try to create a reasonable doubt by blaming an absent colleague, even though one or the other (or both) undoubtedly committed a crime. The joint trial gives the jury the best perspective on all of the evidence and therefore increases the likelihood of a correct outcome.
808 F.2d at 1263; see also Richardson v. Marsh, 481 U.S. 200, 208-210, 107 S.Ct. 1702, 1708, 95 L.Ed.2d 176 (1987).
b. Special Concerns Raised by Prospect of Joint Mega-Trial
It has long been assumed that the advantages referred to above adequately support a strong presumption in favor of joint trials and against severance. See Caliendo, 910 F.2d at 437; see also Bruton v. United States, 391 U.S. 123, 143, 88 S.Ct. 1620, 1631, 20 L.Ed.2d 476 (1968) (White, J., dissenting) (“[unquestionably, joint trials are more economical and minimize burdens on witnesses, prosecutors, and courts” and “avoid delays in bringing those accused of crime to trial”). Thus, to prevail in a motion for severance, a defendant ordinarily “must show that she could not possibly have a fair trial without a severance.” Caliendo, 910 F.2d at 437; see also United States v. Holleman, 575 F.2d 139, 142 (7th Cir.1978) (“severance need be granted only for the most compelling reasons”); Arvanitis, 676 F.Supp. at 847. However, the recent proliferation of complex, multi-de-fendant trials in this district and others, prompted in large part by RICO, has raised doubts about the foundations of this onerous burden. See generally Federal Bar Council Committee on Second Circuit Courts, A Proposal Concerning Problems Created by Extremely Long Criminal Trials (1989) (hereinafter Extremely Long Criminal Trials). Some courts, when faced with a multitude of defendants indicted together under the expansive RICO umbrella, have questioned the wisdom of blindly embracing the purported advantages of a joint trial while, at the same time, disregarding the manifest difficulties presented by what is commonly called a “mega-trial.” See, e.g., United States v. Mancuso, 130 F.R.D. 128, 131 (D.Nev. 1990); United States v. Gallo, 668 F.Supp. 736, 754 (E.D.N.Y.1987).
A careful review of these difficulties reveals a point of diminishing returns with respect to the net benefits of a joint *1172trial as the number of defendants and complexity of the indictment increases. Accordingly, at some point, the oft-cited advantages of a joint trial are outweighed by the manifest disadvantages of a large and protracted trial. Based on this abstract supposition, the Second Circuit, in United States v. Casamento, 887 F.2d 1141, 1152 (2d Cir.1989), cert. denied, — U.S. -, — U.S. -, 110 S.Ct. 1138, 2175, 107 L.Ed.2d 1043, 109 L.Ed.2d 504 (1990), adopted a presumption against a joint trial and for severance when faced with the prospect of a mega-trial. It counseled that in the event the estimated length of the government’s case is more than four months, the prosecutor should “present a reasoned basis to support a conclusion that a joint trial of all the defendants is more consistent with the fair administration of justice than some manageable division of the case into separate trials for groups of defendants.” Id. If, in addition to an estimated length of more than four months, the case involves more than ten defendants, “the prosecutor should make an especially compelling justification for a joint trial.” Id.
Although the approach of the Second Circuit is necessarily arbitrary, it is well-taken. It correctly recognizes that the broad societal disadvantages of large and protracted joint trials, some of which are referred to below, may outweigh their arguable advantages. See generally Dawson, Joint Trials of Defendants in Criminal Cases: An Analysis of Efficiencies and Prejudices, 77 Mich.L.Rev. 1379 (1979). Accordingly, a strong presumption in favor of joint trials is not justified in the context of an inordinately complex mega-trial like the one proposed here, where the principal nexus between the charges is that the defendants allegedly were associated with the same criminally-oriented gang.
c. Disadvantages of Joint Mega-Trial
One disadvantage, which is particularly relevant in this case, is the significant exacerbation of the public cost of providing defense counsel to each defendant. All but one of the defense attorneys currently representing a defendant in this case are being federally funded. Testimony directly implicating most of their clients is projected to last not more than a few weeks, and, for many, not more than a few days. Thus, in a single trial, most of these twenty-one to twenty-eight appointed attorneys would be compelled to sit idly for the duration of a lengthy trial where the vast majority of evidence deals solely with the criminal activities of other attorneys’ clients. See United States v. Phillips, 664 F.2d 971, 1017 n. 68 (5th Cir.1981) (defendant complains that only 120 pages of 12,-000 page transcript concerned his activities), cert. denied, 457 U.S. 1136, 102 S.Ct. 2965, 73 L.Ed.2d 1354 (1982); United States v. Morrow, 537 F.2d 120, 137 (only twenty-five pages of transcript devoted to testimony concerning defendant’s activities out of a total record of “over fifty volumes”), cert. denied, 430 U.S. 956, 97 S.Ct. 1602, 51 L.Ed.2d 806 (1977); United States v. Kelly, 349 F.2d 720, 759 (2d Cir.1965) (defendant’s name first mentioned three months into a nine-month trial and then only sporadically thereafter), cert. denied, 384 U.S. 947, 86 S.Ct. 1467, 16 L.Ed.2d 544 (1966); see also Extremely Long Criminal Trials, at 2 (discussing unidentified case where a defendant’s name was not mentioned until after six months of trial). The added expense of this idle attorney time would needlessly escalate the total public cost of providing these attorneys. Indeed, if this proposed joint trial were to last as long as projected (two years by defense counsel, one year by the Court, and five to six months by the prosecution), this added expense would be exorbitant. According to the estimates of defendant Derrick Porter 10 and based on the current hourly rate for court-appointed counsel of $40 out-of-court and $60 in-court,11 the cost of twenty-one court-appointed attorneys for a one-year trial is at least $1.6 million. Here, a significant majority of this total would be *1173wasted as compensation to defense counsel for idle time.
Another significant disadvantage is the havoc this case would wreak upon this Court’s already burdened docket. A trial of the length proposed by the government would obviously prevent this Court from fulfilling its obligations to litigants in other pending cases, both civil and criminal. We would be required to preside over just one case for an indefinite and prolonged period of time, which, in turn, would result in unconscionable delays in many of our other cases.12 See Gallo, 668 F.Supp at 755; see also Mancuso, 130 F.R.D. at 131. Moreover, the trial management tasks presented by such a case promise to be unwieldy.13 A case of this nature would require the prompt attendance of dozens of persons, any one of whose absence would bring the proceedings to an abrupt halt. No day in court would be free from the ominous specter of yet another delay caused by the lateness or absence of any of the more than one hundred witnesses, eighteen jurors and alternates, twenty-two to twenty-nine defendants, twenty-two to twenty-nine defense counsel and tens of court, security, and government personnel. Additionally, the myriad of inevitable motions, objections, and sidebar discussions would in all likelihood result in further lengthy and unpredictable delays.
A trial of the mammoth proportions advocated by the government would also place enormous personal burdens on the jurors, the defendants, defense counsel and the Court. Sitting as a juror in a case of *1174this nature increases the normal inconvenience of jury duty to almost unacceptable proportions. To be taken away from personal and occupational endeavors for a year or more to “sit stoically and silently for hours every day, day after day," is an unconscionable disruption of a person’s personal and employment life. Gallo, 668 F.Supp. at 754. This disruption would be magnified exponentially if, as is a possibility in this case, it became necessary to sequester the jury. In a lengthy, multi-de-fendant trial, this entire process is “draining, disorienting, exhausting, and often demoralizing.” Id.; see also ABA Report, Jury Comprehension in Complex Cases 31-33 (Dec.1989) (hereinafter Jury Comprehension). These are uniquely inordinate burdens to place upon citizens assuming the civic responsibility of jury duty. Moreover, it is unrealistic to expect that this onerous and unfair hardship will not impact adversely on the jury’s proper functioning. It should be self-apparent that a juror frustrated14 or hostile 15 because of an unduly prolonged trial will not be a fair and conscientious juror. See Extremely Long Criminal Trials, at 2 (speculating that jury’s verdict acquitting each of the two dozen defendants in United States v. Accetturo, 842 F.2d 1408 (3rd Cir.1988), after a fifteen-month trial, represented “either an angry attack on the prosecution for the length of the trial or an abdication of the jurors’ responsibilities occasioned by their inability or refusal to deliberate over the enormous volume of evidence”).
The defendants themselves would also be unnecessarily burdened. At least seventeen of the twenty-two defendants currently in custody are being detained solely as a result of this indictment. The time spent in jail prior to trial has already been prolonged by the expansive discovery required and the large number16 and excessive complexity of pretrial motions. These defendants, presumed innocent, are entitled to as swift a rendering of justice as the circumstances allow. Prolonged pretrial incarceration and a trial of undue length strains this right to its breaking point. Gallo, 668 F.Supp. at 754.
The more than twenty court-appointed defense attorneys would also be deleteriously affected by a lengthy trial. It is one thing to expect urban defense attorneys with heavy overhead expenses to work at the reduced hourly rates mandated by the Criminal Justice Act17 when the case’s duration can be measured in weeks. It is quite another matter, however, to expect these attorneys to survive economically on these comparatively spartan hourly fees in a case anticipated to last as long as a year or two. Not only would their present earnings be diminished, but there would be long-range adverse effects on their practices due to such a lengthy full-time commitment. These severe repercussions to their professional careers would be particularly damaging in this case where most of the attorneys are sole practitioners.
The judge’s ability to rule objectively and dispassionately would also be unduly compromised. As the court in Gallo stated:
The grinding tension of such a long, complex case, particularly where the judge is making rulings which are continuously on the borderline of probative force and prejudice, is debilitating.... [Moreover, *1175a]fter the ease has been under way for some time, the judge is increasingly concerned about the possibility of a mistrial, which would require yet another extended period in tying up the court. Thus, as the trial develops, it becomes increasingly difficult for the court to view objectively the case and its evidentiary decisions without the added tension of avoiding errors that might result in a mistrial or reversal.
Id. at 755; cf. United States v. Ianniello, 866 F.2d 540, 542, 543-44 (2d Cir.1989) (after thirteen-month trial, case remanded for hearing regarding claims that Judge, in ex parte communications to jury, improperly admonished that there could be no hung jury).
Thus, it is clear that a “monster” trial such as the one proposed here presents uniquely significant inefficiencies and hardships. These disadvantages far outweigh any arguable advantages of a joint trial, including the potential reduction in aggregate trial time, which in this case in any event is doubtful. See Mancuso, 130 F.R.D. at 131; Gallo, 668 F.Supp. at 756-758. Accordingly, a strong presumption in favor of a joint trial is not justified in the context of a mega-trial. Indeed, since several separate and shorter trials involving smaller groups of defendants would largely eliminate the disadvantages of such a trial, the public interest supports a strong presumption against a joint mega-trial and in favor of severance.18 Casamento, 887 F.2d at 1152.
2. Prejudice to Defendants
A Rule 14 analysis also requires us to assess the potential prejudice of a joint trial to the defendants. Every joint trial involves some inherent degree of prejudice. United States v. Morales, 868 F.2d 1562, 1573 (11th Cir.1989); United States v. McLaurin, 557 F.2d 1064, 1074 (5th Cir. 1977), cert. denied, 434 U.S. 1020, 98 S.Ct. 743, 54 L.Ed.2d 767 (1978); see also Spencer v. Texas, 385 U.S. 554, 562, 87 S.Ct. 648, 653, 17 L.Ed.2d 606 (1967) (“all joint trials ... furnish inherent opportunities for unfairness ... ”). Ordinarily, we would consider whether the defendants have demonstrated that this prejudice outweighs the presumption in favor of a joint trial. Zanin, 831 F.2d at 744; Rivera, 825 F.2d at 159. But this is not an ordinary case. Since we have concluded the public interest weighs against a joint trial in this case, the defendants have no presumption to overcome. Indeed, in the context of this mega-trial, the scales are tipped decidedly in favor of severance.19
The Seventh Circuit has provided some guidance in assessing the degree of prejudice presented by the prospect of a joint trial. It has identified four situations in which this prejudice may require severance under Rule 14:
(1) antagonistic defenses conflicting to the point of being irreconcilable and mutually exclusive; (2) massive and complex evidence making it almost impossible for the jury to separate evidence as it relates to each defendant when determining each defendant’s innocence or guilt; (3) a *1176co-defendant’s statement inculpating the moving defendant; and (4) gross disparity in the weight of the evidence against defendants.
Garner, 837 F.2d at 1413; see also United States v. Oglesby, 764 F.2d 1273, 1276 (7th Cir.1985). Satisfaction of any one of these criteria may mandate severance. Id.
a. “Massive and Complex Evidence”
We will begin our discussion of possible prejudice with the Garner court’s second specified concern, “massive and complex evidence,” because this concern most compellingly requires severance in this instance. The necessity of severance in this context reflects the precept that we use “every safeguard to individualize each defendant in his relation to the mass.” Kotteakos, 328 U.S. at 773, 66 S.Ct. at 1252. The critical question then is whether “it is within the jury’s capacity ... to follow admonitory instructions and to keep separate, collate and appraise evidence relevant only to each defendant.” Oglesby, 764 F.2d at 1276; United States v. Fernandez, 892 F.2d 976, 990 (11th Cir.1989) (question is whether jury could assess each defendant’s culpability based “solely upon that defendant’s own acts, statements and conduct”). In other words, is it reasonable to conclude that as a practical matter the jury will give each defendant the individual justice that the law demands? We find that it is not reasonable to so conclude.
It is fanciful to believe that any jury would be able, or even willing, to intelligently and thoroughly deliberate over the enormous volume of evidence expected at a single trial of this action. In its present form, the trial would involve twenty-two to twenty-nine defendants accused of over 150 factually separate criminal acts spanning a period of over twenty years and involving at least twenty-five different provisions of the state and federal penal codes. The government concedes that the volume of evidence at such a trial would be “massive,” and we find that solely by virtue of its volume the evidence would be equally “complex.” 20 After this long and arduous trial, the jury would be required to sift through a virtual warehouse of evidence to determine what items were presented against which defendant and as to which criminal act. It would then be obliged to resolve a plethora of difficult factual issues and to strictly apply the detailed and complex law as provided by hundreds of pages of jury instructions. The inevitable length of such a trial dramatically increases the difficulty of this Herculean task. Both common sense and scientific study dictate that as the volume of evidence and corresponding length of trial increases, the degree and quality of jury comprehension decreases proportionately. See, e.g., Jury Comprehension, at 25, 31-32. To expect any jury to accurately recall and appraise the vast amount of detailed testimonial and documentary evidence it heard many months or even a year earlier is unrealistically optimistic.
The difficulty of this task would be increased further by the inevitable admission of large amounts of evidence on a limited basis. While much of the anticipated evidence would be technically relevant to the RICO charges under Federal Rule of Evidence (“Evidence Rule”) 401 as to all of the defendants, a large part of it would surely be excluded under Evidence Rule 403 against most of the defendants because it would be cumulative or prejudicial.21 Other evidentiary rules would also compel the surgical admission of evidence against only selected defendants. With respect to co-conspirator’s statements under Evidence Rule 801(d)(2)(E), for example, the evidence would be admitted on an especially complicated basis due to the multitude of conspir*1177acies charged as subparts to the over-arching RICO conspiracy. See Gallo, 668 F.Supp. at 751. These conspiracies, as factually convoluted as any ever charged in this district, involve varying combinations of defendants, exist for differing periods of times, and involve defendants who allegedly join and leave the conspiracies at various times.
All of this would require the jury to understand and apply a web of hundreds of complicated and intricate limiting instructions to evidence that it may barely recall in the first place. The task, akin to solving a hideously complicated puzzle, would be overwhelming. See id. at 752 (suggesting jury could not heed the “extraordinarily intricate limiting instructions” of mega-trial without aid of a computer). Although in most cases “it is within the jury’s capacity to follow the trial court’s limiting instruction requiring separate consideration for each defendant and the evidence admitted against him,” United States v. Diaz, 876 F.2d 1344, 1358 (7th Cir.1989), there are some cases “in which the risk that the jury will not, or cannot, follow instructions is so great, and the consequences of failure so vital to the defendant, that the practical and human limitations of the jury system cannot be ignored.” Bruton, 391 U.S. at 135, 88 S.Ct. at 1627. This is such a case. The sheer volume and complexity of the anticipated instructions make it impossible to ignore the limitations of the jury system. See United States v. Lane, 474 U.S. 438, 450 n. 13, 106 S.Ct. 725, 733 n. 13, 88 L.Ed.2d 814 (1986) (“in the context of mass trials ..., limiting instructions on evidence admissible only as to one defendant might in some circumstances be inadequate to prevent prejudice”) (emphasis removed); see also Blumenthal v. United States, 332 U.S. 539, 559-60, 68 S.Ct. 248, 257-58, 92 L.Ed. 154 (1947); Gallo, 668 F.Supp. at 753. The real danger, of course, is that the jury would, in spite of all precautions, consider evidence admissible only as to some defendants in determining the culpability of others and in the process convict some defendants because of their association with the mass. Another equally unacceptable probability is that the jurors may convict or acquit some defendants as a result of their confusion in sorting out the mass of evidence, charges, and jury instructions.
Notwithstanding the immense difficulties facing the jury that we have just charted, the government argues that the jury would not be overwhelmed by its task. The government contends that with its proposed organized presentation of the evidence and the use of various visual aids, the jury, after a trial projected to last a year or more, would be able to a make a thoughtful and considered determination of the culpability of each of the twenty-two or more defendants and be able to intelligently apply the more than twenty-five distinct provisions of the various relevant penal codes to the more than 150 separate criminal acts committed over the course of more than twenty years. This proposition strikes us as a great deal of make-believe. Unquestionably, the human mind is limited in its ability to accept, synthesize, and recall information accurately. To try this case in its present form would, at the very least, push these abilities to their uncertain limits.' This places in jeopardy the most important objective of a criminal trial: accurate fact-finding. We cannot place the liberty of twenty-two or more individuals at stake under such circumstances. Thus, because there is likely to be “massive and complex evidence making it almost impossible for the jury to separate evidence as it relates to each defendant when determining each defendant’s innocence or guilt,” we find that severance is appropriate under Rule 14. Garner, 837 F.2d at 1413.
b. Disparity of Evidence
Severance is also independently appropriate in this case because of the “gross disparity in the weight of the evidence against [the] defendants.” Id. A “gross disparity” in the evidence presents a danger that some defendants will suffer “spillover prejudice” due to the accumulation of evidence against other defendants. When thát occurs, a defendant may suffer a transference of guilt merely due to his association with a more culpable defendant. *1178Id. Thus, we again must consider whether it is within the jury’s capacity to adhere to instructions requiring separate consideration of the evidence admitted against each defendant.22 Moya-Gomez, 860 F.2d at 765. Generally, unless the disparity of evidence is unquestionably severe, courts presume that the jury will faithfully observe such instructions. See United States v. Hedman, 630 F.2d 1184, 1200 (7th Cir. 1980), cert. denied, 450 U.S. 965, 101 S.Ct. 1481, 67 L.Ed.2d 614 (1981). In this case, however, the evidentiary disparity is unquestionably severe and, consequently, we will not presume the jury has the necessary capacity.
At one end of the spectrum, three defendants have been charged with the commission of less than five racketeering acts under RICO. This group includes Isiah Kitchen, who is charged with only two narcotics-related acts and a separate weapons possession offense. Four other defendants are charged with only five racketeering acts. Moving along the spectrum, several other defendants are charged with twelve or more racketeering acts. Among this group is James Speights, who is accused of nineteen racketeering acts, including participation in seven murder conspiracies and one witness kidnapping. At the far end of the spectrum, Alan Knox is charged with thirty-nine racketeering acts. These acts include two murders, participation in ten murder conspiracies, and the illegal purchase of a LAW rocket with the intent to perform terrorist acts against the United States on behalf of the Libyan government. If fugitive Melvin Mayes is apprehended, the apparent disparity will be increased further because he is charged with seventy racketeering acts. Thus, it is evident from the vastly disproportionate number of criminal acts charged against these defendants that there is likely to be a “gross disparity” in the weight of the evidence presented. If the Seventh Circuit’s directive to sever where there is a “gross disparity” of the evidence is to have any meaning at all, it must apply to this case.
The government argues that even if the evidence is sufficiently disparate to warrant severance, the danger of “spillover prejudice” is mitigated in this case because in any event most of the evidence would be admissible at each of the separately severed trials. It contends that it would be entitled to offer evidence of all of the charged racketeering activity at each separate trial to prove the entire pattern of racketeering activity and the full scope of the enterprise. See United States v. Garver, 809 F.2d 1291, 1298 (7th Cir.1987); United States v. Persico, 621 F.Supp. 842, 853 (S.D.N.Y.1985). This argument is unpersuasive simply because all the evidence would not be admissible at the separate trials. A great deal of evidence technically admissible under Evidence Rule 401 would certainly be excluded at each severed trial because of its lack of connection to the active criminal deeds of any of the defendants at those particular trials. This exclusion is dictated by Evidence Rule 403 which mandates the exclusion of evidence whose probative value is substantially outweighed by the “danger of unfair prejudice” and by “considerations of undue delay, waste of time, or needless presentation of cumulative evidence.” Fed.R.Evid. 403. Thus, because of its functional inadmissability, the technical admissability of the evidence at separate trials does not militate against severance here.
Therefore, due to extreme evidentiary disparity, a trial in its present form unquestionably raises the specter of “spillover prejudice” and the resulting danger of “guilt by association” such that each defendant’s fundamental right to an independent determination of his guilt or innocence is jeopardized. Severance is appropriate here on this ground alone.
*1179c. Antagonistic Defenses
The potential for “antagonistic defenses conflicting to the point of being irreconcilable and mutually exclusive” is also an important concern in this case. Garner, 837 F.2d at 1413. Generally, severance is not required where there is a mere hostility between defenses and some finger-pointing between defendants. United States v. Turk, 870 F.2d 1304, 1306 (7th Cir.1989); see also Buljubasic, 808 F.2d at 1263 (“finger-pointing is an acceptable cost of the joint trial and at times is even beneficial because it helps complete the picture before the trier of fact”). Severance is only required where the hostile defenses are “irreconcilable and mutually exclusive” such that the “acceptance of one defendant’s defense will preclude the acquittal of the other defendant.” Bruun, 809 F.2d at 407; see also Turk, 870 F.2d at 1306; United States v. Pacheco, 794 F.2d 7, 9 (1st Cir.1986) (to require severance, antagonism must go “beyond mere fingerpointing into the realm of fundamental disagreement over core and basic facts”). While it is true, as the government points out, that no defendant has done more than speculate about the potential for this type of antagonism, we think that it is nonetheless appropriate to consider this factor.
The sheer number of defendants to be tried makes conflicting trial strategies and a significant degree of finger-pointing a virtual certainty. This potential prejudice is compounded here because the jury may not comprehend the individual defense theories due to the mass confusion caused by twenty-two or more attorneys vigorously pursuing his or her own course while seeking his or her own client’s acquittal. Moreover, if any of these separate courses became too antagonistic during the trial so as to then mandate severance, a great waste of time and judicial resources would result. Although this potential prejudice may not require severance at this initial stage, we believe that the real possibility of it eventually occurring weighs heavily in favor of severing this trial. See Gallo, 668 F.Supp. at 751.
C. Balance Struck in Favor of Severance
In summary, the potential prejudice against the defendants in this case and the administrative and fiscal advantages of separate trials outweigh any of the possible advantages of conducting a joint trial. Indeed, a trial of this magnitude would be appropriate in only the rarest circumstances. Nonetheless, in spite of its inherent prejudice and affront to the public interest, the mega-trial too often has been improperly tolerated and even endorsed by some participants in our judicial system.
In the first instance, of course, the decision to bring a multitude of defendants and charges within the scope of a single indictment is within the province of the prosecutor. See United States v. Batchelder, 442 U.S. 114, 124, 99 S.Ct. 2198, 2204, 60 L.Ed.2d 755 (1979). Some courts and commentators have urged prosecutorial restraint with regard to the use of this discretion. See, e.g., United States v. Agueci, 310 F.2d 817, 840 (2d Cir.1962) (solution to problem of mass trials “is largely in the hand of the United States Attorney,” who has initial charging discretion), cert. denied, 372 U.S. 959, 83 S.Ct. 1013, 10 L.Ed.2d 11 (1963); 8 R. Cipes, I. Hall & M. Waxner, Moore’s Federal Practice ¶ 8.06(2), at 8-28 (1990); Second Circuit Judicial Conference Panel Discussion on the Problems of Long Criminal Trials, 34 F.R.D. 155, 161 (1963) (comments of Judge Weinfeld). Indeed, a prosecutor has an obligation to use some measure of restraint and to consider the effect of a sizable indictment on the integrity of the resulting trial. As stated by Justice Sutherland in Berger v. United States:
The United States Attorney is the representative not of an ordinary party to a controversy, but of a sovereignty whose obligation to govern impartially is as compelling as its obligation to govern at all; and whose interest, therefore, in a criminal prosecution is not that it shall win a case, but that justice shall be done.
295 U.S. 78, 88, 55 S.Ct. 629, 633, 79 L.Ed. 1314 (1935); see also United States v. Bell, 901 F.2d 574, 578 (7th Cir.1990) (“As an *1180officer of the court, the [assistant United States attorney] has some responsibility to ensure, as far as may be reasonably possible, the integrity of the proceedings.”); ABA Standards for Criminal Justice § 3-1.1(c) (1980) (“The duty of the prosecutor is to seek justice, not merely to convict.”); cf. United States v. Foskey, 636 F.2d 517, 526 (D.C.Cir.1980) (with respect to admissability of evidence pursuant to Evidence Rule 403, “assistant United States attorney must step back from his or her partisan role and make these determinations in an objective and fair-minded fashion before proffering the evidence”). In this case, the government demonstrated little or no concern for any such obligation when it charged thirty-eight defendants in this convoluted 175-count, 305-page indictment.
Unfortunately, prosecutors have long been willing to bring cases of this nature despite the violence they do to the notion of a fair trial.23 Indeed, prosecutors generally have demonstrated an increasing “penchant for drawing together evermore complex and extensive conspiracies into a single indictment.” Kopituk, 690 F.2d at 1320; see also Extremely Long Criminal Trials, at 1. This phenomenon can be explained by the fact that prosecutors have significant incentives to bring mass indictments, not the least of which is the consequent ability to procure a large number of convictions in what they perceive to be the shortest amount of time. The regrettable truth, however, is that these incentives24 carry far greater weight in the charging decision than any concern for a fair and manageable trial. It seems unlikely, then, that the recent increase in mega-trials will soon be curbed at the initiative of the prosecution.
It is also unlikely that appellate courts will be able to offer a great deal of assistance in this regard. After a lengthy trial resulting in conviction, the denial of severance is, understandably, rarely overturned.25 In the Seventh Circuit, the matter is judged by an “abuse of discretion" standard and “will be disturbed on appeal only if it results in manifest and substantial prejudice.” Bruun, 809 F.2d at 407; see also United States v. Briscoe, 896 F.2d 1476,1516 (7th Cir.1990) (appellant has “extremely difficult burden” when challenging denial of severance); Anderson v. United Parcel Service, 915 F.2d 313, 315 (7th Cir.1990) (“The district court’s decision must strike us as fundamentally wrong for an abuse of discretion to occur.”). As a practical matter, this onerous burden is to some degree more difficult to meet in the context of a mega-trial where there is a natural reluctance to render moot such a massive commitment of time and money. See 8 R. Cipes, I. Hall & M. Waxner, Moore’s Federal Practice ¶ 8.06(2), at 8-28 (1990). Simply put, appellate courts, by virtue of their *1181role in the judicial system, are not in a position to effectively deter the mega-trial. The resulting and almost invariably silent tolerance of such trials at the appellate level perpetuates their continued existence. But see Casamento, 887 F.2d at 1149 (court upholds results of mega-trial, but finds strong presumption against such trials).
Thus, it seems that the trial court is the place where the tide of mega-trials must be stemmed.26 It certainly has the authority to do so given its virtually unbounded discretion to grant severance in such cases. Indeed, the trial court is compelled to use this discretion to “secure simplicity in procedure, [and] fairness in administration” and, consequently, should never tolerate a trial that involves a multitude of charges against a throng of defendants. Fed.R. Crim.P. 2. The defendants’ motions for severance are granted under Rule 14.
III. Severance Plan Overview
A. Government’s Opportunity for Input
Having determined that severance is appropriate, we next turn to the specifics of the severance plan. Because at this stage of the proceedings the government is the only party with complete knowledge of the evidence that it will present in this case, it is in the best position to suggest an efficient and effective severance plan. However, unfortunately, it has declined our previous requests to participate in the formulation of such a plan. On two separate occasions, we gave the government an opportunity to set forth a scheme to sever this unwieldy indictment. Taking guidance from the Second Circuit in Casamento, we issued an order on May 10, 1990 requesting that the government
propose a plan to sever this case into several trials (by defendants and/or counts) in the event the defendants' motions to sever are granted ... [and to] consider whether the interests of justice would be adequately served by limiting the prosecution of individual defendants to charges that can be proven expeditiously and that, in the event of conviction, carry exposure to maximum penalties.
Cf Casamento, 887 F.2d at 1152. Again, at a pretrial conference on August 31, 1990, we requested suggestions from the government regarding an appropriate plan. Unfortunately, on both of these occasions the government refused and instead insisted that the case could only be tried as indicted.27
B. Plan Structure
Therefore, without input from the government, we adopt a severance plan that attempts to address the considerations discussed above and to strike a balance between other practical and competing concerns. To begin, we have sought to drastically reduce the volume of evidence against each defendant unrelated to his active participation in El Rukn affairs and his alleged crimes. At the same time, we have sought to ensure that the introduction of evidence is not so restricted as to unduly impair the government’s case against any defendant or its opportunity to secure a conviction carrying significant penalties. Additionally, it is important to limit the duplication of *1182evidence at the separate trials and, accordingly, minimize aggregate trial time and the corresponding expenditure of both public and judicial resources.
With these considerations in mind, the defendants will be divided into five non-overlapping groups to be tried at five separate trials. While the government will be permitted to go forward at all five trials with Counts One and Two, the RICO conspiracy and substantive RICO counts respectively, it will not be permitted to try all of the racketeering acts charged in these counts at every trial. Instead, some acts will not be tried at all and the remaining acts will be divided between the five trials and, for the most part, tried only once. Of course, non-RICO substantive counts will be tried along with their corresponding RICO racketeering acts. Likewise, substantive counts that correspond to racketeering acts that are not tried will be held in abeyance, to be tried at a subsequent phase of trials if if becomes necessary.28
C. Overlapping Evidence
While a division of criminal offenses will substantially limit the overlapping presentation of evidence, the repetition of certain evidence is unavoidable due to the nature of RICO charges. An essential element of both RICO conspiracy and substantive RICO is the existence of an “enterprise that affected interstate commerce.” Neapolitan, 791 F.2d at 500. According to the court in Neapolitan an “enterprise” is
an association having an ascertainable structure which exists for the purpose of maintaining operations directed toward an economic goal that has an existence that can be defined a part [sic] from the commission of the predicate acts constituting the “pattern of racketeering activity.”
Id. (quoting United States v. Anderson, 626 F.2d 1358, 1372 (8th Cir.1980), cert. denied, 450 U.S. 912, 101 S.Ct. 1351, 67 L.Ed.2d 336 (1981)); see also Delta Truck & Tractor, Inc. v. J.I. Case Co., 855 F.2d 241, 243 (5th Cir.1988), cert. denied, 489 U.S. 1079, 109 S.Ct. 1531, 103 L.Ed.2d 836 (1989); United States v. Weinberg, 656 F.Supp. 1020, 1024 (E.D.N.Y.1987) (“enterprise must have ongoing formal or informal organization, and its various associates must function as a continuing unit”). Because the El Rukn organization is the alleged “enterprise” for the RICO charges against each defendant, its existence must be proven at all five trials. Presumably, this may entail a presentation of some of the same evidence five times. Nonetheless, the gravamen of proof necessary to RICO convictions will not be repeated because, unlike proof of an enterprise common to all RICO defendants, the remaining elements of proof under RICO substantially pertain only to individual criminal activity-
To convict a defendant under substantive RICO, the government must prove, in addition to proof of an enterprise: (1) that the defendant was employed by or associated with the enterprise; (2) that he participated, either directly or indirectly, in the conduct of the affairs of the enterprise; and (3) that he participated through a pattern of racketeering activity, i.e., through the commission of at least two racketeering acts. United States v. Alvarez, 860 F.2d 801, 818 (7th Cir.1988), cert. denied, 490 U.S. 1051, 109 S.Ct. 1966, 104 L.Ed.2d 434 (1989); Kopituk, 690 F.2d at 1323.
The elements of RICO conspiracy are analogous to the elements of substantive RICO and similarly focus upon individual criminal activity. To prove RICO conspiracy, however, the government need not show that a defendant committed two racketeering acts or even that he agreed to personally commit two such acts. Instead, it is enough that the defendant manifested “his agreement to the objective of a violation of RICO.” Neapolitan, 791 F.2d at
*1183498. Thus, a defendant is guilty of RICO conspiracy when he enters an “agreement to conduct or participate in the affairs of an enterprise and an agreement to the commission of at least two predicate acts” by anyone associated with the enterprise. Id. at 499.
Therefore, the only element of RICO conspiracy and substantive RICO that will necessitate the repetition of evidence is the existence of an enterprise. Evidence of the other elements will, for the most part, differ according to each defendant’s individual association with, and participation in the affairs of, the enterprise. Thus, while the government will need to prove the existence of the enterprise at all five trials,29 it will need to prove the remaining elements of Counts One and Two only once at the separate trials of each group of defendants.
D. Evidentiary Limitations and Counts in Abeyance
Before detailing the specifics of the severance plan, another matter merits some discussion. The complexity of this indictment and the alleged wide-ranging involvement of many of the defendants makes it impossible to divide both the defendants and the alleged criminal offenses into non-overlapping groups and, at the same time, permit the government to try every defendant for every racketeering act charged under RICO. The indictment charges eleven large and distinct conspiracies involving various combinations of defendants, each of whom is alleged to have committed numerous other crimes and to have been a member of other conspiracies. Consequently, there are no discreet groups of defendants who are charged with discreet and separate crimes and, therefore, no neatly separable groups. Thus, to avoid overlapping evidence, it is necessary to exclude evidence of many racketeering acts that each defendant allegedly committed, as well as to hold in abeyance the substantive counts that correspond to such acts.
There are certain minimum limitations, however, to the number of racketeering acts that may be excluded from evidence. As discussed above, a substantive RICO conviction requires proof of a “pattern of racketeering” and the commission of at least two racketeering acts. Alvarez, 860 F.2d at 818; Kopituk, 690 F.2d at 1323. Thus, to maintain the possibility of such a conviction for all defendants, the severance plan must allow the government the opportunity to prove at least two racketeering acts against each defendant. However, because a substantive RICO violation does not require proof of any more than two acts in establishing a “pattern of racketeering,” evidence of additional racketeering acts becomes increasingly less important and less probative.30 Accordingly, at some point, the probative value of such evidence will be substantially outweighed by the “danger of unfair prejudice” and by “considerations of undue delay, waste of time, or needless presentation of cumulative evidence.” Fed.R.Evid. 403. Therefore, under Evidence Rule 403, we will exclude evidence of many of the multitude of racketeering acts charged against the defendants. See Kelsay v. Consolidated Rail *1184Corp., 749 F.2d 437, 443 (7th Cir.1984) (admissibility of evidence pursuant to Evidence Rule 403 is “within the sound discretion of the trial court”).
Additionally, this limit on evidence is necessary to the viability of this or any other severance plan of this nature. To hold otherwise and to allow the government to prove every racketeering act that it has alleged against every defendant would in effect allow the government to retry the entire case at every trial. This, of course, is unacceptable because it would largely defeat one of the most compelling reasons for granting severance. Under such circumstances, this Court has the authority and broad discretion to exclude evidence of cumulative racketeering acts in order to control the separate trials. See Gentile v. County of Suffolk, 129 F.R.D. 435, 458 (E.D.N.Y.1990); cf. Fed.R.Evid. 102 (“rules shall be construed to secure fairness ..., [and] elimination of unjustifiable expense and delay”); Fed.R.Evid. 104(a) (court not bound by rules of evidence in determining admissability); Fed.R.Evid. 611(a) (“court shall exercise reasonable control over ... presenting [of] evidence so as to ... avoid needless consumption of time”).
Of course, as we have stated, we will hold in abeyance the substantive counts that correspond to racketeering acts that we exclude from evidence. This aspect of the plan may, indeed, eliminate the need to ever spend time and resources trying these crimes. Under the severance scheme we have devised, each defendant faces a potential jail term that can be measured in decades and many face potential life sentences for RICO convictions based on murder. See 18 U.S.C. § 1963(a) (1988); Ill.Rev.Stat. ch. 38, para. 9-1 (1989). For all practical purposes, convicting a defendant of an offense that carries a substantial penalty will obviate the need to try him for the remaining crimes charged against him. If a defendant is not convicted, the government will have the opportunity at that time to determine whether it wishes to proceed with the counts held in abeyance.
IV. Details of Severance Plan
As set forth in detail below, we have devised a severance plan to meet these concerns and the other concerns of the Court and the parties.31 We believe that this plan provides the most efficient and effective method of trying this case. While it reduces the potential prejudice that each defendant may suffer, it still gives the government more than sufficient opportunity to convict every defendant of crimes that carry substantial penalties. It also reduces the evidence that must be presented at more than one trial.32
A. Trial One
Defendants Louis Hoover, Felix Mayes, Derrick Porter, and Noah Robinson will be tried together at Trial One. The connection between these defendants is that they are all alleged to have participated in crimes in which Robinson is the central figure. This group will be tried for these crimes, which include the attempted murder of Robert Aulston in Texas and the murder of Leroy Barber in South Carolina as well as the various acts of witness tampering committed against defendant Henry Leon Harris and Janice Rosemond. These crimes are charged as substantive Counts Eight through Sixteen and alleged in Count Two as corresponding Racketeering Acts 23, 24, and 26-30.
To insure the possibility of Count Two substantive RICO convictions for Hoover, F. Mayes, and Porter, however, it is necessary to allow the government to prove additional racketeering acts. Thus, the alleged murders of turncoat El Rukns Willie McLilly, Roy Love and Mickey Cogwell and the *1185alleged conspiracy to murder Lemont Tim-berlake will also be tried at Trial One. These offenses, which are not charged as separate substantive counts, are alleged as Racketeering Acts 3, 4 and 8. The addition of these acts gives the government the opportunity to prove that both F. Mayes and Porter committed at least two racketeering acts. Although both of these defendants are charged with other racketeering acts, these acts were chosen because of the limited alleged participation of defendants who are not included in this group. Only two of the remaining defendants to be tried who are not fugitives are charged with participation in these crimes.
Because Hoover is charged with the commission of only one violent offense, the government will also be permitted to try a narcotics-related offense against him. Thus, at Trial One the government may prosecute Racketeering Act 78 and corresponding substantive Count Sixty, one of the two narcotics-related offenses with which Hoover is charged, wherein Hoover allegedly used a telephone to facilitate a narcotics offense in violation of 21 U.S.C. § 843(b) (1988). The other narcotics-related offense with which he is charged is Count Three and Racketeering Act 31, the broad narcotics conspiracy. The other three Trial One defendants are also alleged members of this conspiracy. Nonetheless, this offense will be held in abeyance with respect to this group to avoid overlapping evidence because, as will be explained below, it will tried at another trial. This is the only count that is held in abeyance. The government has the opportunity to obtain convictions for the remaining twelve counts charged against this group, including substantive RICO.
B. Trial Two
The following eight defendants will be tried at Trial Two: Jeff Boyd, Edgar Cook-sey, Andrew Craig, Charles Green, Alan Knox, Sammy Knox, Michael Sardin, and James Speights.33 The primary connection between these defendants is that they were all members of the alleged King Cobra conspiracy, charged as substantive Count Four and alleged as Racketeering Act 19. Although there are other defendants charged in this conspiracy, they are not included here because none of them is charged with any of the additional King Cobra-related crimes, which include the murders of Robert Jackson, Rico Chalmers, Glendon McKinley, and Vicki Nolden, the attempted murder of Andre Chalmers, and the assault of Theotis Clark. Five of the defendants in this group are charged with some participation in these crimes, charged as Counts Five, Six, and Seven and alleged in Count Two as Racketeering Acts 20, 21, and 22.
The other three defendants in the group, S. Knox, Sardin, and Speights, although not charged with any of these additional crimes, are included because of their alleged involvement in the McKinley kidnapping, charged as Count One Hundred Seventy-Five and also alleged as Racketeering Act 25. This is important because the other alleged kidnapper, Boyd, was included in this group due to his involvement in the additional King Cobra-related crimes. Thus, in the interest of keeping the alleged McKinley kidnappers together, S. Knox, Sardin, and Speights are placed in this group.
The offenses tried at Trial Two will include, in addition to the McKinley kidnapping and all of the King Cobra crimes, the Ronald Bell murder conspiracy, the murders of Gilbert Connors and Gregory Freeman, various narcotics-related offenses, and a weapons offense. Five of the eight defendants in this group are named as members of the Bell conspiracy, alleged as Racketeering Act 17 but not charged as a separate substantive count. The Connors and Freeman murders, alleged as Racketeering Acts 1 and 2, are included here because the only defendant standing trial *1186who is charged with these murders is group member S. Knox. The other defendants charged with these murders have pleaded guilty.
The narcotics offenses that will be tried at this trial include two counts of cocaine distribution in violation of 21 U.S.C. § 841(a)(1) and six counts of the use of a telephone to facilitate a narcotics offense in violation of 21 U.S.C. § 843(b). Sardin and Speights will be tried for cocaine distribution as alleged in Counts One Hundred Fifty-Nine and One Hundred Sixty and corresponding Racketeering Acts 125 and 126. These two defendants will also be tried for a separate weapons offense, charged in Count One Hundred Seventy-Two, because this offense is allegedly connected to Count One Hundred Sixty. The six telephone counts relate to a series of alleged phone conversations that took place on May 1, 1986 involving Boyd, A. Knox, and Speights and to phone conversations that took place on May 9 and May 21, 1986 involving S. Knox and A. Knox. These crimes are charged as Counts Sixty-Five, Sixty-Six, Sixty-Seven, Eighty-Two, Eighty-Four, and Eighty-Eight and are also alleged as, or as a part of, Racketeering Acts 83, 84, 85, 97, and 98.
Although the defendants in this group, individually and collectively, have been charged with other narcotics-related offenses, many of the crimes chosen for trial here were selected because they were allegedly committed by at least two members of the group. Others were chosen simply to increase the number of Count Two racketeering acts for which certain defendants are tried. The remaining substantive counts charged against this group are held in abeyance.
C. Trial Three
The following five defendants will be tried together at Trial Three: Henry Andrews, Thomas Bates, George Carter, Roland Lewis, and Melvin Tillman. Also, in the event any or all of them are apprehended in time, the following fugitive defendants will be tried with this group: Floyd Davis, Eddie Franklin, Bernard Green, Melvin Mayes, Walter Pollard, and Edward Williams. Each of these defendants was an alleged member of the King Cobra murder conspiracy charged as Count Four and also alleged as Racketeering Act 19. However, as noted above, none of these defendants is charged with the commission of the additional related crimes or with the McKinley kidnapping.
Although it will cause some overlap in the evidence between Trials Two and Three, this group will also be tried for the King Cobra conspiracy. This limited overlap allows the government to pursue Count Four against all alleged conspirators, thereby reducing the number of counts held in abeyance. The offenses tried at this trial will additionally include the Bruce Davis murder conspiracy and the Jerome Smith conspiracy, alleged as Racketeering Acts 12 and 18. Every member of this group except defendants Lewis and Tillman and fugitive defendant M. Mayes is named as a member of both the Smith and Davis conspiracies.
With the exception of the Count Three narcotics conspiracy and the narcotics-related offenses included at Trial Two, all narcotics-related offenses charged against this group will also be tried here. An exception to this will be made for fugitive M. Mayes, who is charged with sixty-five narcotics-related offenses. Even if he is apprehended in time for trial, most of these charges will be held in abeyance.
Carter, Davis, Lewis, and Williams will each be tried for individual counts of cocaine distribution in violation of 21 U.S.C. § 841(a)(1). These crimes are charged as Counts Seventeen, Twenty-Four, Twenty-Six, Twenty-Seven, and Thirty-One and are also alleged as Racketeering Acts 34, 41, 43, 44, and 48. Tillman will also be tried for a violation of § 841(a)(1), which is alleged as Racketeering Act 49 but is not charged as a separate substantive count. Additionally, M. Mayes and B. Green will be tried for three counts of using a telephone to facilitate a narcotics offense in violation of 21 U.S.C. § 843(b), alleged as Counts Fifty-Two, Sixty-One, and Eighty-Six and also as Racketeering Acts 70, 79, and 99. Bates will be tried for one count *1187of violating § 843(b), charged as Count Forty-Eight and also alleged as Racketeering Act 66. The remaining counts charged against the defendants in this group are held in abeyance.
D. Trial Four
Defendants Jerome Crowder, Lawrence Crowder, William Doyle, and J.L. Houston will be tried together at Trial Four. None of these defendants is alleged to have been involved with any of the crimes that tied together the defendants in the first three trials, namely the alleged Robinson crimes, the King Cobra conspiracy crimes, and the McKinley kidnapping. While there is some overlap between the other offenses to be tried at those trials and the offenses allegedly committed by the Trial Four group, it is a limited overlap extending only to the Jerome Smith, Ronald Bell, and Bruce Davis murder conspiracies. Nonetheless, to avoid overlapping evidence, this group will not be tried for these crimes.
The offenses to be tried against this group will include all of the remaining violent offenses. This includes the Titanic Stones, Audrina Thomas, Chalmers Tyler, and Ewing brothers murder conspiracies and the related murders and attempted murders. None of these crimes, alleged as Racketeering Acts 5, 6, 9, 10, 13, 14, 15, and 16, are charged as a separate substantive count. According to Count Two, L. Crowder and Doyle were members of three of these murder conspiracies, while J. Crowder and Houston were members of all four.
Additionally, J. Crowder and Doyle each will be tried for narcotics-related crimes. These will include Racketeering Act 32, charging that Doyle allegedly traveled in interstate commerce to distribute cocaine in violation of 18 U.S.C. § 1952, and Racketeering Act 33, charging him with cocaine distribution in violation of 21 U.S.C. § 841(a)(1). Neither of these crimes is alleged as a separate substantive count. Jerome Crowder will be tried for two counts of using a telephone to facilitate a narcotics offense in violation of 21 U.S.C. § 843(b), which are charged as Counts One Hundred Forty-Six and One Hundred Forty-Nine and also alleged as Racketeering Acts 116 and 118.
Neither Houston nor L. Crowder are charged with additional narcotics related crimes except the large narcotics conspiracy charged in Count Three and corresponding Racketeering Act 31. J. Crowder and Doyle are also alleged members of this conspiracy. Nonetheless, Count Three is held in abeyance as to these defendants because it will be tried in the fifth and final trial. The only other substantive counts held in abeyance are a number of narcotics-related offenses charged against J. Crow-der, who in any event will be tried for six racketeering acts.
E. Trial Five
Isiah Kitchen and, in the event he is apprehended in time, Roger Bowman will be tried together at Trial Five. Unlike the rest of the defendants, Kitchen and Bowman have not been charged with the commission of any violent crimes. Their participation in the affairs of the El Rukns is allegedly limited to narcotics-related offenses. The prejudice that either defendant would face at a trial in which the government was allowed to present evidence of numerous violent offenses, including murders, attempted murders, kidnapping, or obstruction of justice, is unacceptable. Thus, Kitchen and Bowman will be tried separately and the government will be permitted to go forward with all charges against these defendants, including the RICO charges in Counts One and Two. However, as is the case in the previous trials, the evidence presented in support of the RICO charges largely will be limited to the criminal offenses committed by these defendants.
These offenses allegedly include participation as co-conspirators in the narcotics conspiracy in violation of 21 U.S.C. § 846 charged as Count Three and alleged as corresponding Racketeering Act 31. Additionally, Bowman is charged with three counts of using a telephone to further narcotics-related offenses in violation of 21 U.S.C. § 843(b), charged as Counts Sixty, Sixty-One, and One Hundred Forty-Six and also alleged as corresponding Racke*1188teering Acts 78, 79, and 116. Kitchen is charged with six counts of violating § 843(b) and one count of possession with intent to distribute cocaine in violation of 21 U.S.C. § 841(a)(1). These crimes are charged as Counts One Hundred Sixty-One through One Hundred Sixty-Eight, and are also alleged collectively as corresponding Racketeering Act 127. Kitchen is also charged with possession of firearms in violation of 18 U.S.C. § 922(g)(1), the one substantive offense with which he is charged that is not alleged as a corresponding racketeering act. None of the Counts charged against these defendants are held in abeyance.
F. Trial Sequence
Finally, we must decide the sequence and timing of these trials. The cases will be tried in the sequence that they are set forth in Section IV of this opinion. Trial One will commence at the date previously scheduled. Trials Two through Five will follow Trial One in sequence. However, because most of the defendants at Trials Two through Five have already been in custody for a significant period of time, we will explore with the Executive Committee of this Court the feasibility of assigning these trials to other District Court Judges to be tried simultaneously.34 See Casamento, 887 F.2d at 1152 (separately severed trials “may, in the judge’s discretion, be assigned for simultaneous trial before other judges”).35 If simultaneous trials are not feasible and the cases are instead tried sequentially, we will revisit, as promised earlier, the pretrial detention orders of the defendants whose trials will be delayed further. Thus, in the event of sequential trials, the Court will order that Magistrate Rosemond reconsider the pretrial detention orders of those defendants in trials Two through Five.
V. Conclusion
The indictment in this case is a massive 305-page “book” detailing the alleged criminal activity of the thirty-eight named defendants. The apparent justification for this unprecedented and confused joinder of a great number of allegedly bad actors charged with a great number of disparate crimes committed over a long period of time in a single trial is that these defendants were all associated with the same street gang and that the fruits of their crimes somehow enured to the general benefit of the gang. Although the drafting of this indictment shows a certain degree of creativity and its return may have made dramatic press conference fodder, these are the easiest tasks in the preparation of a joint prosecution. Now the charges must be tried and the feasibility of attempting to prove them in one trial must be faced squarely.
The oft-stated rule is that defendants who are charged together should be tried together. Nevertheless, whether to grant a severance is left to the sound discretion of the trial court. To say that we should not exercise this discretion to sever these defendants and charges into several separate manageable trials is tantamount to saying that under no circumstances should a trial court ever exercise its discretion to sever. The large number of defendants, the large number and complexity of the charges, and the projected length of the proposed mega-trial presented in this indictment mandate that we order separate trials. Were we not to so rule, the defendants could not receive fair trials, the jury would not be able to properly perform its function, the costs of appointed counsel would be exorbitant, and the damage to the orderly administration of other litigation assigned to this Court’s docket would be considerable. This is too high a price to pay for the government’s ill-conceived desire to prosecute all these defendants in one spectacular trial extravaganza.
Thus, for the multiple and detailed reasons discussed in this opinion, we deny the *1189motions to sever pursuant to Rule 8(b) and grant the motions to sever pursuant to Rule 14. However, in the interest of implementing an optimum severance plan that fairly reflects the interests of all of the parties, we will temporarily stay execution of this order. For the next forty-eight hours, we will accept proposals from any of the parties regarding suggested improvements to our severance plan provided that these suggestions are consistent with the goals of this opinion.36 It is so ordered.