Opinion of the Court
Tried by special court-martial, the accused was convicted, despite his pleas, of one specification alleging possession of marihuana and hashish with the intent to distribute, in violation of 21 U.S.C. § 841(a)(1), and Article 134, Uniform Code of Military Justice, 10 U.S.C. § 934. The adjudged and approved sentence extends to a bad-conduct discharge, confinement at hard labor for 60 days, and reduction to pay grade E-l. The Court of Military Review affirmed in a per curiam opinion.
We granted review of the following issue:
WHETHER THE GOVERNMENT’S DELAY FOR TACTICAL PURPOSES IN BRINGING APPELLANT TO TRIAL UNFAIRLY PREJUDICED APPELLANT AND DENIED HIM HIS RIGHT TO SPEEDY TRIAL UNDER ARTICLE 10 UCMJ AND THE SIXTH AMENDMENT TO THE U.S. CONSTITUTION.
We hold there was nothing improper in the Government’s motives and affirm.
The accused was arrested at his off-base apartment on January 23, 1981, by civilian law-enforcement authorities, after a search of his residence revealed the presence of drugs and other contraband. The search was conducted pursuant to a warrant issued by a civilian magistrate and was based upon information provided by an informant. On March 25, 1981, the civilian charges which had been filed against the accused were dismissed, after a series of scheduled hearings, because the civilian authorities were unable to produce the evidence which had been seized during the search of the accused’s apartment.
The informant was a serviceman working for the Naval Investigative Service (NIS). Although the case against the accused was initially processed by civilian authorities, NIS agents also investigated it. On the day the accused was arrested, NIS agents secured sworn statements from three other servicemen who admitted their own involvement in drug transactions and identified the accused as their supplier. The information available at the time military charges were preferred indicated that the accused was the head of a “drug distribution network.” The Government’s strategy was to try him last in anticipation that the other servicemen would testify against him in accordance with their sworn statements; however, this strategy backfired. After each of their trials, the accused’s alleged partners-in-crime were granted immunity; and each, in remarkably similar fashion, effectively recanted those portions of their earlier sworn statements which linked the accused to crime.
The chronology of events in this case is as follows:
23 Jan 1981: Accused arrested at his off-base apartment by civilian police. Search reveals drugs and other contraband.
25 Mar 1981: Civilian charges against accused dismissed.
8 May 1981: Military charges preferred against accused. (These charges were never referred to a court-martial.)
27 May 1981: Article 32, UCMJ [10 U.S.C. § 832], investigation ordered by convening authority.
17 Jun 1981: Accused placed on legal “hold” effective 16 Aug 81.
10 Jul 1981: First scheduled meeting of Article 32 investigation. Two-week delay requested by civilian defense counsel. (This counsel also represented one of the other servicemen, and it appears that this delay was requested on behalf of that person and not the accused.)
24 Jul 1981: Article 32 investigation began.
14 Aug 1981: New charges preferred against accused. (In essence these charges duplicated those of 8 May 1981, except that a conspiracy specification was added.)
*5516 Aug 1981: Accused’s release from active duty date.
15 Sep 1981: Accused demands immediate trial.
18 Sep 1981: Reply from convening authority.
6 Oct 1981: Initial Article 32 investigating officer reassigned.
8 Oct 1981: Article 32 investigation scheduled to begin. Delayed apparently due to conflict in military defense counsel’s schedule.
15 Oct 1981: Replacement investigating officer challenged for cause and relieved. Basis of challenge and party challenging not of record.
16 Oct 1981: Another investigating officer appointed and some evidence taken. Government counsel requests delay to await results of trial of the last of accused’s alleged associates.
30 Oct 1981: Trial counsel absent on emergency leave.
5 Nov 1981: Trials of all alleged co-actors concluded.
19 Nov 1981: Trial counsel returns from emergency leave.
2 Dec 1981: Additional charges preferred.
4 Dec 1981: Next meeting of Article 32 investigation.
7 Dec 1981: Article 32 investigation terminated.
18 Dec 1981: Charges referred to a special court-martial.
18 Dec 1981-
4 Jan 1982: Accused on annual leave.
7-8, 14, 18-
20 Jan 1982: Trial and conviction of accused.1
The Government seeks to justify the delay on the grounds of prosecutorial discretion in scheduling the trials of the other three servicemen. We note that, in addition to the formal request for trial, the accused made numerous requests for mast with his commanding officer to find out the status of the charges against him.2
Apart from the normal anxieties of awaiting trial and some minor inconveniences, it appears that the accused’s life and daily routine at the air station were largely unaffected by this series of events. In fact, he was allowed to take Christmas leave on the very eve of his pending trial. On the other hand, the accused asserts that because of his involuntary retention in the service beyond his normal date of separation from active duty he was unable to avail himself of several civilian job offers and the opportunity to attend one of several universities which had accepted his application.
At the outset we observe that there is no doubt that the Government perfected extended personal jurisdiction over the accused as a result of its timely commencement of action with a view towards trial. See United States v. Fitzpatrick, 14 M.J. 394 (C.M.A.1983); United States v. Self, 13 MJ. 132 (C.M.A.1982). However, while an accused has no absolute right to immediate separation from the service at the expiration of his enlistment, the Government has an obligation to proceed to trial with all deliberate speed, absent the accused’s acquiescence in the retention. United States v. Fitzpatrick, supra at 397; United States v. Hutchins, 4 M.J. 190, 192 (C.M.A.1978).
In general, accountability for speedy-trial tracking begins at the time of pretrial restraint or preferral of charges, whichever comes first.3 Para. 215e, Manual *56for Courts-Martial, United States, 1969 (Revised edition); see also United States v. Marion, 404 U.S. 307, 320, 92 S.Ct. 455, 463, 30 L.Ed.2d 468 (1971); United States v. Burton, 21 U.S.C.M.A. 112, 44 C.M.R. 166 (1971); United States v. Williams, 16 U.S.C.M.A. 589, 37 C.M.R. 209 (1967). Because there is no evidence of record that authorities of either the state or the United States were acting as agents for each other, the Government was not accountable for the time following the accused’s arrest and charging by the state authorities. Cf. United States v. Reed, 2 M.J. 64, 67 (C.M.A.1976); United States v. Williams, 12 U.S.C.M.A. 81, 30 C.M.R. 81 (1961). Thus, the period of government accountability begins with the preferral of charges on May 8, 1981, even though these were not the charges on which the accused ultimately was tried. See United States v. Avalos, 541 F.2d 1100, 1108-09 (5th Cir.1976), cert. denied, 430 U.S. 970, 97 S.Ct. 1656, 52 L. Ed.2d 363 (1977).
Courts have wrestled with the concept of “speedy trial” which is, and has been, far easier to define than to apply to a given factual situation. We have tried to follow the concepts developed by the Supreme Court of the United States in several cases. See United States v. Johnson, 17 M.J. 255 (C.M.A.1984), for an analysis of these precedents. In Barker v. Wingo, 407 U.S. 514, 92 S.Ct. 2182, 33 L.Ed.2d 101 (1972), the Supreme Court identified four factors which might be used to determine whether an accused has suffered a deprivation of his right to a speedy trial: “[l]ength of delay; the reason for the delay, the defendant’s assertion of his right, and prejudice to the defendant.” These factors are to be balanced “on an ad hoc basis.” Id. at 530, 92 S.Ct. at 2191 (footnote omitted).
The quantum of delay is “a triggering mechanism” for identifying a “presumptively prejudicial” delay. Id. Delays of as little as five or six months have caused the federal courts to inquire into the remaining Barker factors. See United States v. Saintil, 705 F.2d 415, 418 (11th Cir.), cert. denied, 464 U.S. 855, 104 S.Ct. 171, 78 L.Ed.2d 155 (1983); United States v. Ransom, 465 F.2d 672, 673 n. (D.C.Cir.1972). In the instant case, the respective parties calculate vastly different periods of delay as chargeable to the other. However, save for a one-week delay attributable to military defense counsel and the accused’s two- and-one-half-week Christmas leave (which the Government was, in any event, in a position to prevent), this record reveals no delay attributable to the defense between May 8, 1981, and January 7, 1982. Therefore, we are satisfied that the case merits our balancing of the remaining Barker factors.4
Looking at the second Barker factor — “the reason for the delay” — we conclude that it was primarily caused by the Government’s desire to obtain additional testimony against the accused. We find nothing improper in this motive, as such. A prosecutorial decision to await the results of another trial to obtain the testimony of a witness is, by itself, a neutral factor. See United States v. Fairchild, 526 F.2d 185 (7th Cir.1975), cert. denied, 425 U.S. 942, 96 S.Ct. 1682, 48 L.Ed.2d 186 (1976); United States v. Baumgarten, 517 F.2d 1020 (8th Cir.), cert. denied, 423 U.S. 878, 96 S.Ct. 152, 46 L.Ed.2d 111 (1975); United States v. Johnson, 3 M.J. 143 (C.M.A.1977). But cf. United States v. Lara, 520 F.2d 460, 464 (D.C.Cir.1975) (prosecutors obtained dismissal of an indictment in one district in order to indict the defendant *57in another district which they believed would be more sympathetic to their cause); Arrant v. Wainwright, 468 F.2d 677, 681 (5th Cir.1972) (prosecutor delayed trial “because he ‘did not want to see ... [the defendant] be acquitted of this crime’ ”).
In Barker, most of the five-year delay between arrest and trial was due to the prosecution’s efforts to obtain a conviction and thus utilize the testimony of Barker’s co-actor. The Supreme Court indicated that this may be a permissible reason for delay under ordinary circumstances. The delay in that case was, of course, extraordinary.
In the present case, it appears that the Government was willing to delay trial in the accused’s case until after trial of his alleged co-actors in the reasonable expectation of acquiring admissible evidence on serious additional charges. With the benefit of hindsight, it is obvious that the ends of justice would have been better served if the Government had proceeded to trial with the evidence at hand,5 instead of delaying trial to obtain additional evidence. In fact, the charges finally referred to trial were based on evidence available at the time of the arrest in January 1981. That the Government’s strategy failed, however, does not turn a permissible reason for delay into an impermissible one. Once the witnesses recanted their statements incriminating the accused at his Article 32 investigation, the unsubstantiated charges were promptly withdrawn, and the remaining charges were referred to a special court-martial.
The third Barker factor weighs in the accused’s favor, as he did demand trial. This indicates that he actually desired a speedy trial, unlike the situation in Barker where the defendant, for tactical reasons or otherwise, did not. The Supreme Court found no violation of Barker’s sixth-amendment right to a speedy trial, expressing reluctance to rule that a defendant was denied a speedy trial that he did not want.
The final factor to be considered is prejudice to the accused. Barker stated that there are
[three] interests of defendants which the speedy trial right was designed to protect. ... : (i) to prevent oppressive pretrial incarceration; (ii) to minimize anxiety and concern of the accused; and (iii) to limit the possibility that the defense will be impaired. Of these, the most serious is the last, because the inability of a defendant adequately to prepare his case skews the fairness of the entire system.
Id. 407 U.S. at 532, 92 S.Ct. at 2192 (footnote omitted). As to the first interest, there was no “oppressive pretrial incarceration” here. As to the second, there is.no indication that the “anxiety and concern” suffered by this accused exceeded the norm. As to the third, there is no basis in this record for contending that the accused’s defense was impaired by the passage of time. If anything, it was the prosecution’s case that deteriorated.
Further, although he was not confined during the relevant period, the accused claims specific prejudice due to his involuntary retention on active duty. We have held that mere retention beyond expiration of service date for the purpose of administering military justice is not, by itself, tantamount to arrest or confinement.6 United States v. Rachels, 6 M.J. 232 (C.M.A.1979); United States v. Amundson, 23 U.S.C.M.A. 308, 49 C.M.R. 598 (1975). However, while the retention of a service-member beyond the expiration of his enlistment does not establish prejudice per se, it “is a circumstance to be considered in determining whether he has been prejudiced by pretrial delays.” United States v. *58Douse, 12 M.J. 473, 481 (C.M.A.1982) (Everett, C.J., concurring in'the result). Here, the prejudice to the accused was slight. He continued to receive pay and allowances; he was allowed leave, liberty and the other privileges of his military status; and he performed his normal duties.
Applying the Barker factors to the instant case, we are not persuaded that the accused was denied his right to a speedy trial. Although the accused was involuntarily held for trial almost five months past the expiration of his enlistment, the delay was for a legitimate purpose and any prejudice was minimal. We are, therefore, unwilling to reverse the decision of the court below on this ground.
The decision of the United States Navy-Marine Corps Court of Military Review is affirmed.
Chief Judge EVERETT concurs.