20 Mass. App. Ct. 541

February 15, 1985 - August 12, 1985

Suffolk County


A general verdict of acquittal on a charge of murder did not, under the principle of collateral estoppel, bar the prosecution of the defendant for giving perjured testimony at a preliminary hearing on the same murder charge, the allegedly false testimony being the defendant's denial that he had confessed his guilt to an investigating detective, where the defendant did not meet his burden of proving that the truth of that testimony was necessarily determined by virtue of the general verdict of acquittal. [545-549]

At a perjury trial, there was no error in admitting in evidence the defendant's answer to an allegedly confusing compound question, not itself the statement alleged to be perjured, and any confusion in the defendant's mind resulting from the compound nature of the question went only to the weight of the evidence and not to its admissibility. [550-551]

Discussion of the law relating to the nature and quantum of proof needed for conviction at a trial for perjury. [553-557]

At the perjury trial of a defendant who, at the trial of another individual for murder, had allegedly testified falsely in denying that individual's complicity in the murder, the evidence against the defendant was not of a direct or a clear and compelling character, and thus was insufficient as matter of law to support his conviction. [558]

INDICTMENTS found and returned in the Superior Court Department on May 4, 1982.

The cases were tried before William H. Carey, J.

Michael D. Cutler for the defendant.

Judy G. Zeprun, Assistant District Attorney, for the Commonwealth.

ARMSTRONG, J. The defendant appeals from convictions on two indictments charging that he gave perjured testimony: the

Page 542

first instance on October 5, 1981, at a pretrial hearing in connection with a murder charge then pending against him, the allegedly false testimony being his denial of a statement by which he had confessed his guilt to an investigating detective sergeant named McConkey; the second on March 2, 1982, at the trial of one Evans for the same murder, the allegedly false testimony being his (Coleman's) denial that Evans was with him when he committed the murder. The first appeal is based in part on a claim that the trial of Coleman for perjury (and certain evidence at the trial) was barred by principles of double jeopardy and, derivative therefrom, collateral estoppel. The second appeal is based on a claim that the evidence was insufficient to show beyond a reasonable doubt the falsity of Coleman's testimony at Evans's trial.

The background of the perjury charges, as disclosed in the evidence adduced at the trial, is as follows. On the night of October 9, 1980, shortly before midnight, one Donna Hankins and a male friend, Smith, went to a third-floor apartment of a house on Georgia Street, in the Roxbury section of Boston, where Smith "was staying." The house or apartment was described as being that of Evans or Evans's girlfriend. No one was in the apartment when Hankins and Smith arrived; but some time around or shortly after midnight, Coleman arrived and joined them. Hankins and Smith were "getting high" on marijuana and cocaine; Coleman did not use either but was drinking. Evans arrived in the apartment and asked Hankins if she had two dollars he could use to pay a taxi cab driver. She said no. Evans then asked Smith, who apparently carried a gun, if he wanted to "take [the cab driver] off," a phrase Hankins understood to mean rob the cab driver. Smith declined, but Coleman said he would go. Coleman and Evans left the apartment at approximately 1:30 A.M., Coleman carrying a gun. Approximately ten or fifteen minutes later Hankins heard a sound like a firecracker. She and Smith left the apartment around 3:30 A.M. She did not see a cab in the streets near the apartment but walked down Georgia Street to catch a cab at a Grove Hall taxi stop. She did not hear of the murder of the cab driver, Walter J. Butkiewicz, for a week or so.

Page 543

About 3:00 A.M. an Officer Jones and his partner, in response to a radio call, drove in their cruiser to Cheney Street, which, like Georgia Street, is in the vicinity of Grove Hall, and found the body of the cab driver, Butkiewicz, dead in his cab, shot twice through the head.

The subsequent investigation was conducted by Detectives McConkey and O'Meara. Coleman was arrested several days after the murder for an unrelated offense, and a ballistics test indicated that a .38 caliber handgun found on the ground near Coleman when he was arrested (and for which he was carrying .38 caliber bullets) was the gun that killed Butkiewicz. Coleman first told Detective McConkey that, on the night of the murder (October 9-10), at 12:30 A.M., he had purchased liquor in Grove Hall and then proceeded to Evans's apartment, where he spent the night, not going out again until morning. A day later, on being told the police had found a witness who saw him running from Cheney Street around the time of the murder, Coleman gave another alibi: that at midnight he had gone with Evans to the "Sky Cap," where he had smoked reefers with friends until about 2:30 A.M., when he had gone home. Detective McConkey obtained a warrant for Coleman's arrest for the Butkiewicz murder, then confronted him with the evidence amassed against him. He asked Coleman if he had anything to say, and, after a delay, Coleman, sobbing, is alleged to have said to Detective McConkey: "What do you want me to say? You know what? You're right." This statement was taken by Detective McConkey to be an acknowledgment by Coleman of his guilt, and McConkey committed it to writing as soon as possible after he left Coleman's cell. [Note 1]

The two alibis and the alleged confession had been uncounseled, and before Coleman's trial on the murder charge there

Page 544

was a hearing on a motion to suppress. At that hearing Coleman testified under oath and denied both that he had confessed his guilt to Detective McConkey and that he had stated to McConkey, "What do you want me to say? You know what? You're right." He testified that what he had actually said was, "I know my rights." The motion to suppress was apparently denied, and the alleged confession was put in evidence against Coleman at his murder trial. Despite the introduction of the confession, the jury found him not guilty on the indictment.

Evans had also been indicted for the murder of Butkiewicz, but his trial was severed from that of Coleman and took place after Coleman's acquittal. Coleman had not testified at his own trial, but he appeared at Evans's trial as a witness for the defense. There he testified that he was in fact Butkiewicz's killer and that he had been alone at the time of the killing. [Note 2] He was asked if he had confessed to Detective McConkey, and he affirmed that he had and specifically that he had uttered the words, "What do you want me to say? You know what? You're right." The jury acquitted Evans. They had not been told that Coleman had previously been tried and acquitted of the murder.

Coleman was then charged with perjury, the two indictments alleging, first that he had lied at the suppression hearing when

Page 545

he denied confessing to Detective McConkey and, second, that he had lied at the Evans trial when he testified that Evans was not with him when he killed Butkiewicz. The Commonwealth's evidence principally supporting the first charge consisted of the defendant's testimony at the Evans trial that he had confessed to Detective McConkey and the testimony of Detective McConkey that Coleman had confessed to him. The evidence principally supporting the second charge was the testimony of Donna Hankins, which paralleled her testimony at Evans's trial, to the effect that Coleman and Evans had left the Georgia Street apartment together armed and with the express purpose of "taking off" the waiting cab driver and that she had heard a sound like fireworks shortly thereafter. The judge denied motions for required findings on both indictments and, after the jury had returned guilty verdicts, sentenced Coleman to eighteen to twenty years in State prison on the first indictment and to life imprisonment on the second.


Coleman bases his collateral estoppel defense on Ashe v. Swenson, 397 U.S. 436 (1970), which held that collateral estoppel in criminal cases is an integral part of the protection against double jeopardy guaranteed by the Fifth and Fourteenth Amendments to the United States Constitution, and on United States v. Nash, 447 F.2d 1382 (4th Cir. 1971), and United States v. Drevetzki, 338 F.Supp. 403 (N. D. Ill. 1972), each of which applied the holding of Ashe v. Swenson to the trial of a perjury charge based on an acquitted defendant's allegedly false testimony at his prior trial. [Note 3] Coleman's alleged

Page 546

perjury in denying the jail confession did not occur at his murder trial, when he did not testify, but at a pretrial hearing; but his collateral estoppel argument is based on the fact that Detective McConkey testified at the murder trial that he (Coleman) had confessed to the crime [Note 4] and the jury nevertheless acquitted him. This, Coleman argues, necessarily implies that the jury rejected McConkey's testimony that the defendant had confessed to him, a proposition on which the first perjury charge depends.

The cases interpreting the scope and application of the Ashe v. Swenson collateral estoppel doctrine in criminal cases are not always easily understood and reconciled (see Commonwealth v. Kline, 19 Mass. App. Ct. 715 , 717 & n.2 [1985]), and they present particularly serious policy problems when applied to defendants who win acquittal through perjured testimony. See Commonwealth v. Mitchell, 15 Mass. App. Ct. 577 , 582-583 (1983), quoting from Adams v. United States, 287 F.2d 701, 703 (5th Cir. 1961) [Note 5] Nevertheless, the particular

Page 547

facts of this case permit decision on a narrow basis that does not implicate the difficult policy considerations that faced the courts in the Nash and Drevetzki cases.

The collateral estoppel principle on which Coleman relies is "that when an issue of ultimate fact has once been determined by a valid and final judgment, that issue cannot again be litigated between the same parties in any further lawsuit." Commonwealth v. Scala, 380 Mass. 500 , 503 (1980), quoting from Ashe v. Swenson, 397 U.S. at 443. "[C]ollateral estoppel requires the concurrence of three circumstances: (1) a common factual issue; (2) a prior determination of that issue in litigation between the same parties; and (3) a showing that the determination was in favor of the party seeking to raise the estoppel bar." Commonwealth v. Lopez, 383 Mass. 497 , 499 (1981). The burden of showing these circumstances is on Coleman as the person raising the bar. Id.

The first condition is met here. The Commonwealth adduced evidence at the Coleman murder trial that Coleman had confessed being the killer to McConkey; thus, the question whether Coleman had so confessed was a factual issue at the murder trial. It was also an issue at the perjury trial, the gist of the charge being that Coleman lied when he denied at the suppression hearing that he had confessed to McConkey.

The second circumstance is not satisfied, however, unless it can be ascertained that the jury determined that factual issue at the murder trial. The problem is the usual one, that the jury at the murder trial spoke not through express findings of fact

Page 548

but through a general verdict. "Assessing the significance of a jury verdict in some criminal cases may involve . . . `sheer speculation.' But the records of other trials are such as to indicate plainly, when `viewed with an eye to all the circumstances of the proceedings,' Sealfon v. United States, 332 U.S. 575, 579 [1948], that a jury verdict of acquittal is determinative of a particular issue that was contested at that trial." Hoag v. New Jersey, 356 U.S. 464, 475-476 (1958) (Warren, C.J., dissenting). As the burden of showing collateral estoppel is on the person raising the bar, collateral estoppel will not be applied unless the record shows that the issue was "necessarily decided at the first trial." Commonwealth v. Benson, 389 Mass. 473 , 478, cert. denied, 464 U.S. 915 (1983). Typically, collateral estoppel will apply only to "ultimate" facts rather than "evidentiary" facts, "because of the tenuous and speculative relationship between the result in the prior proceeding and the evidence proposed to be presented in the subsequent prosecution." Id., at 481-482.

Those principles are determinative of the collateral estoppel claim relating to the confession. Detective McConkey's testimony at the Coleman murder trial was that he had visited Coleman at the Charles Street jail and had laid out the evidence that tended to incriminate Coleman, that Coleman had sat silent for a long time, perhaps fifteen or twenty minutes, and that then Coleman, sobbing, had said, "What do you want me to say? You know what? You're right." McConkey, interpreting these words as a confession, wrote them down at his first opportunity some fifteen minutes later. McConkey's testimony was evidence from which the jury could make findings and draw inferences, but the jury were not called upon to make a finding whether the words attributed to Coleman were in fact spoken by him or whether, if spoken, they were correctly interpreted by McConkey as intended to acknowledge guilt. The words were not wholly self-explanatory and unequivocal. In the context described by McConkey, the jury could reasonably doubt that the words were intended as a confession, or, if so intended, the jury could have taken into account the fact that this was the third statement made by Coleman as to events the

Page 549

night of the Butkiewicz murder, all uncounseled, and thus doubted the reliability of the words as a confession. Compare, on concerns for reliability in such circumstances, Miranda v. Arizona, 384 U.S. 436, 470 (1966), citing Crooker v. California, 357 U.S. 433, 443-448 (1958) (Douglas, J., dissenting). Thus, it was not impossible for a rational juror to conclude that McConkey accurately reported the words spoken by Coleman, or even to conclude that the words were intended by Coleman as a confession, yet still acquit Coleman by finding little reliability in a confession given in such circumstances and little persuasiveness in the other, circumstantial evidence offered against Coleman. It follows that Coleman has not "met [his] burden of proving that this fact was necessarily determined by virtue of the general verdict of acquittal." Commonwealth v. Benson, 389 Mass. at 481. The judge ruled correctly that the Commonwealth was not precluded by principles of collateral estoppel from prosecuting Coleman under the first perjury indictment. [Note 6], [Note 7]

Page 550

Coleman's remaining contention with respect to the first perjury conviction is that the question asked him at the Evans trial which elicited his acknowledgement that he had confessed to McConkey was a compound question, confusing to a witness who wished to give affirmative answers to some parts and negative answers to others, and thus was too vague to support a perjury prosecution. [Note 8] (Coleman, in fact, gave an unqualified affirmative answer, logically implying the correctness of each component part.)

In some circumstances there could be merit to such an argument if the answer to a confusingly compounded question were itself the statement alleged to be perjured. The untruthful import of an allegedly perjured statement should be direct and clear, not implied or suggested. Bronston v. United States, 409 U.S. 352, 359 (1973) (statement literally true but unresponsive, and found by jury to have been intended to mislead examiner, not perjurious). Commonwealth v. Giles, 353 Mass. 1 , 15-16 (1967) (answer shown ambiguous by later answers may not be basis for perjury prosecution). Commonwealth v. Geromini, 357 Mass. 61 , 64-65 (1970) (same). Here, however, the statement alleged to be perjured was the denial of the confession at the suppression hearing. The denial was clear and unequivocal as to the words recorded by Detective McConkey. Read in the light of Coleman's other answers at the suppression hearing (as required by the Giles and Geromini cases), it is clear that Coleman was denying having made any confession

Page 551

to McConkey in any form of words. He acknowledged McConkey's account of the setting, substituting for the statement recorded by McConkey ("What do you want me to say? You know what? You're right.") another statement carrying a wholly different import: "I know my rights." Any confusion resulting from the compound nature of the question asked at the Evans trial would go merely to the weight of the evidence that the allegedly perjured statement was in fact false.

Moreover, the question asked at the Evans trial was not, in context, confusing. It was compound in the sense that it called upon Coleman to say whether he remembered several propositions in addition to the words attributed to Coleman by McConkey. A denial would have left uncertain which of the propositions were true or false, but Coleman's affirmative answer had no such ambiguity. Here it is of significance that Coleman was an astute witness, not confused by complicated questions and not hesitating to express his inability to understand questions which were objectively confusing. [Note 9] In addition he was assisted by counsel appointed to advise him at the Evans trial, and Coleman conferred with him before answering the immediately preceding question. The jury could properly take all these facts into account in assessing the weight to be given Coleman's acknowledgment of the recorded words at the Evans trial. They could also take into account that the acknowledgement was consistent with and supported the thrust of Coleman's testimony at the Evans trial: namely, that he murdered Butkiewicz and that Evans was not with him when he committed the murder.

Page 552


Coleman's alleged perjury at Evans's trial -- his denial of Evans's complicity in the murder -- does not give rise to a claim of collateral estoppel for the reason, among others, of lack of mutuality, Coleman not having been a party in the Evans case. Commonwealth v. Cerveny, 387 Mass. 280 , 284 (1982), citing Standefer v. United States, 447 U.S. 10 (1980). [Note 10] Commonwealth v. Royce, ante 221, 228 n.5 (1985). Rather, Coleman's contention is that the evidence was insufficient as matter of law to warrant a finding that his denial of Evans's complicity (or even presence) was perjured. In particular, he focuses on an absence of proof that Evans was in fact with him at the scene of the murder.

The evidence on this point most favorable to the Commonwealth was as follows. Donna Hankins testified that Coleman and Evans left the Georgia Street apartment together at around 1:30 A.M., that their express purpose was to rob a waiting cab driver, that Coleman was armed with a handgun, and that she heard a noise like a firecracker some fifteen minutes later. The cab driver, Butkiewicz, was in fact murdered that night sometime before 3:00 A.M., shot with a pistol found, inferentially, in Coleman's control several days later. Butkiewicz's body was found in his cab parked on Cheney Street, in the vicinity of Georgia Street. Coleman and several other young men were seen running from Cheney Street around 2:40 or 2:45 A.M., [Note 11] warranting an inference that Coleman's testimony at the Evans trial, to the extent that he claimed to have been alone at the time of the murder, was false. [Note 12]

Page 553

If a perjury indictment were subject solely to the usual rule as to burden of proof in criminal cases (proof from which a rational jury could draw an inference of guilt beyond a reasonable doubt, Commonwealth v. Latimore, 378 Mass. 671 , 677 [1979]), we would conclude that the jury, on the stated evidence, properly could have found beyond a reasonable doubt that Evans was with Coleman at the time of the murder, and thus that Coleman testified falsely when he denied Evans's presence. The jury, by giving credence to Donna Hankins's testimony except for the 1:30 A.M. departure time, could infer that the firecracker sound ten to fifteen minutes after the departure of Coleman and Evans together was in fact the sound of the gunshot that killed Butkiewicz and that Evans, having set out merely minutes before with the intention of participating in the armed robbery of Butkiewicz, must have been one of those who fled with Coleman from Cheney Street a few minutes before the arrival of the police.

In a perjury prosecution, however, the customary standard of proof in criminal cases has traditionally been supplemented by a special requirement relating to the nature and quantum of proof needed for conviction. See Brightman v. United States, 386 F.2d 695, 697-698 (1st Cir. 1967). The usual statement of the rule is that one may not be convicted of perjury except on the directly opposing testimony of either two witnesses or one witness and, in addition, "independent evidence [of] strong corroborating circumstances, of such a character as clearly to turn the scale and overcome the oath of the defendant and the legal presumption of his innocence." Commonwealth v. Parker, 2 Cush. 212 , 223-224 (1848). See also Commonwealth

Page 554

v. Douglass, 5 Met. 241 , 243 (1842); [Note 13] Commonwealth v. Pollard, 12 Met. 225 , 228 (1847). [Note 14] "It is not necessary that there should be two living witnesses in contradiction of the statement of the defendant to justify a conviction of perjury. It is sufficient if, in addition to one directly opposing witness, corroborating circumstances sufficient to turn the scale and overcome the oath of the defendant and the legal presumption of his innocence are proved." Commonwealth v. Butland, 119 Mass. 317 , 324 (1876). If the so called "quantitative rule" is still applicable in this Commonwealth, the proof of the Evans trial perjury is insufficient, as matter of law, to warrant conviction, there not having been even one directly opposing witness at the perjury trial. (For example, no witness was produced to testify that Evans was one of those who was seen fleeing from Cheney Street with Coleman after the murder.)

More recent cases have tended to criticize the quantitative rule and to question its continued applicability. See, e.g., Commonwealth v. Gale, 317 Mass. 274 , 277-278 (1944); Commonwealth v. Fine, 321 Mass. 299 , 302-303 (1947); Commonwealth v. Giles, 353 Mass. at 20; and, most recently, Commonwealth v. Hawley, 380 Mass. 70 , 81-82 (1980). The historical reasons for the development of the rule no longer obtain (see 7 Wigmore, Evidence Sections 2041-2042 [3d ed. 1940]), [Note 15] and it has

Page 555

been suggested that its application should "be limited to the situation for which it was designed, namely, to prevent a conviction of perjury when there is no evidence other than the word of one witness against that of the defendant. It has no place in a case in which the falsity of defendant's testimony can be established by evidence of a different kind." Perkins, Criminal Law c. 5, Section 3(7), at 465 (2d ed. 1969). The Model Penal Code Section 241.1(6) (1980), retains the rule in such a limited form: "Corroboration. No person shall be convicted of an offense under this Section [i.e., perjury] where proof of falsity rests solely upon contradiction by testimony of a single person other than the defendant."

The reason for retention of the rule after its historical underpinnings had vanished is probably explained by a widely felt, not unreasonable, fear that if a witness could be found guilty of perjury merely on the contradictory testimony of another witness, every witness would face potential reprisal from those against whom he testifies; [Note 16] and in every case in which there is conflicting testimony on a contested and material question of fact, the transcript would contain substantially all the evidence needed to indict, try, and convict the conflicting witnesses for perjury. Such a regime would ultimately affect the accessibility of witnesses, who would inevitably be discouraged from testifying in court. See Bronston v. United States, 409 U.S. at 359, citing Study of Perjury, reprinted in Report of New York Law Revision Commission, Legis. Doc. No. 60, at 249 (1935). See also Weiler v. United States, 323 U.S. 606,

Page 556

609 (1945). The quantitative rule is thus an expression of a societal policy to encourage witnesses to come forward with their testimony freely, without fear of reprisal, by protecting them from the threat of a perjury conviction on anything less than compelling proof of knowing falsity. [Note 17]

The quantitative rule may justifiably be thought a crude tool for effectuating the underlying policy and be rejected for that reason. See, especially, Commonwealth v. Hawley, 380 Mass. at 82, and Nolan, Criminal Law Section 601, at 401 (1976). We have no doubt that some, more refined, rule must be devised to serve the policy. Some of the possible contours of such a rule may be found in the various lines of authority which have recognized exceptions to the quantitative rule. Those best recognized are (1) where the witness accused of giving perjured testimony has himself given contradictory testimony under oath on another occasion; [Note 18] (2) where the witness's testimony is shown to be untrue by substantially direct documentary or physical evidence; [Note 19] and (3) where falsity is not in its nature

Page 557

susceptible of direct proof, and the circumstantial evidence of falsity "is of such a character as to be virtually positive or direct" (Annot., Perjury -- Circumstantial Evidence 88 A.L.R. 2d 852, 875 [1963]). [Note 20] Where the quantitative rule has been abandoned altogether, courts have nevertheless insisted that the circumstantial evidence of the perjury be of a compelling character, objectively inconsistent with the innocence of the defendant. Compare Commonwealth v. Parker, 2 Cush. at 223-224 ("independent evidence [of] strong corroborating circumstances, of such a character as clearly to turn the scale"); Laughran v. Kelly, 8 Cush. 199 , 202 (1851) ("independent evidence of such a character as clearly to turn the scale"). Listed in the margin are Federal cases which have stated, in varying forms of words, the same underlying concept concerning the quality of the evidence necessary to support a perjury prosecution. [Note 21]

Page 558

The evidence with respect to the alleged Evans trial perjury did not have the requisite compelling character. There was no direct evidence that Evans was with Coleman at the time of the murder. Only inference supports the proposition that the murder was perpetrated shortly after Coleman and Evans left the Georgia Street apartment, rather than an hour or more later. Evans was not among those indicated by the evidence to have been seen fleeing from Cheney Street at 2:45 A.M., when the murder may have occurred. No medical evidence was offered at the perjury trial to show the time of death. The evidence leaves problematical whether the killing occurred on Georgia Street or on Cheney Street; the evidence fails to indicate the distance between the Georgia Street apartment and the place where the Butkiewicz cab was parked on Cheney Street. If, as we hold, the evidence needed to sustain a perjury conviction must be of a direct or clear and compelling character, the Commonwealth, with respect to the alleged Evans trial perjury, failed to meet its burden of proof. It follows that the conviction on that indictment cannot stand.

On indictment no. 039591, the judgment is reversed, the verdict is set aside, and judgment is to be entered for the defendant. On indictment no. 039590, the judgment is affirmed.

So ordered.


[Note 1] The interval may have been as much as fifteen to twenty minutes. Detective McConkey closed off the interrogation as soon as he obtained the statement, followed somewhat cumbersome security procedures for leaving the Charles Street jail, wrote down the statement on entering his cruiser, and recorded that he and Detective O'Meara had heard it. Detective McConkey later testified to his certainty that the written statement was a verbatim rendition of Coleman's words.

[Note 2] The full transcript of Coleman's testimony at Evans's trial, which was marked as an exhibit but not put in evidence (selected portions were read to the jury), shows that he told the jury that he and Evans left the Georgia Street apartment together at about 12:30 A.M., that the cab Evans had arrived in had left, that he and Evans walked to the Sky Cap, Evans leaving after a half hour or so and Coleman staying until after 2:00 A.M. when a friend drove him to his home. Thereafter he left his house and walked to Cheney Street alone, came upon the Butkiewicz cab, and robbed and killed Butkiewicz. Coleman's testimony placed the murder at around 2:30 A.M. or 2:45 A.M., approximately an hour later than would be inferred from the testimony of Donna Hankins (at the perjury trial; the record of her testimony at Evans's trial is not before us). We can only speculate, in the absence of the transcripts of either of the two murder trials (except for Coleman's testimony at Evans's trial), that Coleman's testimony may have tied in better than Hankins's with the testimony of witnesses who were said to have seen Coleman fleeing from Cheney Street around 2:45 A.M. and, inferentially, notified the police.

[Note 3] In Ashe v. Swenson, the defendant was accused of being one of four armed robbers who burst in on a poker game and robbed the six players. Charged with six counts of armed robbery, the defendant was tried and acquitted on one of the counts. The question was whether collateral estoppel barred his being tried on the other five. In the Nash case postal inspectors, to apprehend a thief, placed marked quarters in an envelope, saw the defendant take the envelope to the ladies' room, and later searched her, finding the marked quarters. She testified she got the quarters from a money-changing machine, suggesting that the real thief must have inserted the quarters and that she had the misfortune to receive them when changing a dollar. A jury acquitted her of theft despite evidence that the machine could not recycle coins inserted into it. She was then charged with perjury. In the Drevetzki case, the defendant, charged with theft, testified at his trial and denied having admitted his guilt to a special agent. The latter testified that the defendant had admitted his guilt, but the defendant was acquitted. He was then charged with perjury for denying his confession.

[Note 4] The transcript of McConkey's testimony at the Coleman murder trial is not before us; but defense counsel elicited from McConkey at the prejury trial that Coleman's jailhouse statements as testified to at the perjury trial were put before the murder trial jury by McConkey's testimony there. The defendant relies on that proposition in this court.

[Note 5] "On the one hand the concern exists that allowing an acquittal to afford any sort of insulation for perjury will be giving defendants an uncontrollable license to testify falsely. The resulting detriment to the reliability of evidence and more so, to the stability of the judicial process, would only be enhanced by the obvious fact that the more persuasively flagrant the defendant's fabrication, the greater his chances of total exoneration. This completes a vicious circle since the successful acquittal on the substantive offense would immunize him as to the very falsehoods which brought it about. On the other hand some apprehension exists that allowing prosecution for perjury will actually give the State a second shot at the defendant for the same wrong. The mere fact, this argument continues, that one charge relates to the doing of an act and the other to a denial of having done it or to affirmative proof that it was not so done is not sufficient basis on which to make a distinction. This is particularly true where the same or substantially the same evidence is presented in both cases . . . . This, we see, approaches closely, whether acknowledged or not, an intuitive feeling akin to double jeopardy despite the fact that the two are distinct." 287 F.2d at 703, (The last sentence, of course, was written nine years before Ashe v. Swenson, when it was generally understood that double jeopardy was not involved where the first and second offenses were separate: i.e., where the "evidence necessary to sustain a second indictment would [not] have been sufficient to secure a conviction on the first." Hoag v. New Jersey, 356 U.S. 464, 467 [1958]).

[Note 6] The Drevetzki case (see note 3, supra) is distinguishable because the court concluded on the facts of that case that the jury at the first trial (for theft) necessarily had to decide the question whether Drevetzki had confessed to the special agent. If the case had not been dismissed on that ground, the government might well have encountered difficulty in complying with the substantive standards for proof in a perjury case, which are discussed later in this opinion in connection with the alleged Evans trial perjury. Factually, this case is distinguishable from Drevetzki because of the present defendant's later sworn testimony (contrary to his testimony at the suppression hearing) acknowledging that he had confessed to Detective McConkey.

[Note 7] That our decision is put on this narrow basis is not meant to intimate that, if we were required to decide the broader point, we would accept the assumption of the Nash and Drevetzki cases that the collateral estoppel principle of Ashe v. Swenson was intended to apply to a criminal defendant's perjured testimony by which he obtains an acquittal. A sensible middle ground to accommodate the competing interests articulated in the Adams case (see note 5, supra) can be found in the view expressed by Winter, J., concurring in the Nash case (447 F.2d at 1386-1387), to the effect that the purpose of the double jeopardy clause and the related collateral estoppel principles is sufficiently served if the defendant's perjury must be proved by new, compelling evidence different from or additional to that heard by the acquitting jury. The evidence supporting Coleman's first perjury conviction (i.e., his sworn testimony at the Evans trial by which he repudiated his sworn testimony at the suppression hearing) would meet that criterion whether or not Coleman's acquittal at his own murder trial necessarily implied the jury's rejection of McConkey's testimony concerning Coleman's jailhouse confession. In contrast, the government in both the Nash and Drevetzki cases appears to have relied in the perjury prosecutions solely on the evidence previously rejected by the juries which had heard the original criminal prosecutions.

[Note 8] The transcript reads: "Q. And, sir, after [Detective McConkey] outlined [the evidence] to you, there was a long pause, and you didn't say anything for a period of five or ten or maybe fifteen minutes, and then the officer said to you, `We are going to have to leave now, David. Do you wish to say anything?' And at that time you said to this officer: `What do you want me to say? You know what? You're right.' Do you remember that happening? A. Yes."

[Note 9] This observation, based on the limited sections of transcript that were read in evidence at the perjury trial, is strongly reinforced by a reading of Coleman's entire testimony at the suppression hearing and at the Evans trial (the transcripts of that testimony were marked for identification and are thus before us although they were not admitted in evidence due to the objection of defense counsel). Coleman also testified at a voir dire during the perjury trial.

[Note 10] Mutuality is not always required for collateral estoppel in civil cases. See Fireside Motors, Inc. v. Nissan Motor Corp. in U.S.A., 395 Mass. 366 , 372 n.6 (1985), and cases cited.

[Note 11] This statement is based on Detective McConkey's testimony concerning what he had told Coleman prior to Coleman's confession. That testimony was properly admitted to show Coleman's state of mind but was subject to a limiting instruction because of its hearsay nature. No such instruction having been sought or given, the testimony was in the case for all purposes. See Liacos, Massachusetts Evidence 74-75 and 441 (5th ed. 1981).

[Note 12] Not admitted in evidence was testimony offered by the Commonwealth to the effect that there was no damage to the cab from a second bullet that grazed Butkiewicz but did not lodge in his body. That testimony, coupled with evidence that the cab doors were shut and the windows were up, would have cast further doubt on Coleman's story that he shot Butkiewicz twice from outside the cab while Butkiewicz was seated in the cab and suggested that Butkiewicz was himself outside the cab when shot. Such an inference would suggest (although not compel) an inference that Coleman had accomplices to assist in seating Butkiewicz's body in the cab after death.

[Note 13] "The reason of the rule in cases of perjury is, that the same effect is to be given the testimony of the party accused, as to that of the accusing witness, so that if there be no other proof, the scale of evidence is poised; there being witness against witness, oath against oath." 5 Met. at 243.

[Note 14] The Pollard case held that the two-witness or one-witness-plus-independent-corroboration rule applies only to the element of falsity, not to the elements of (1) the taking of the oath, (2) the jurisdiction of the tribunal, (3) the words spoken, (4) their materiality, or (5) the knowing character of any falsity.

[Note 15] When the rule originated, defendants in criminal cases could not be witnesses in their own behalf; thus there was generally no oath to offset that of the accuser. In perjury trials, however, the oath taken by the defendant (when he gave the allegedly perjured testimony) offset the oath of a single accuser, and the case stood in equipoise (see note 13, supra) until the oath of another accuser was weighed in. After criminal defendants were made competent to testify in their own behalf, the equipoise rationale (oath against oath) could be applied equally to any crime.

[Note 16] Wigmore (3d ed. 1940), supra, quoting from Best, Evidence Sections 605-606 (1849), states: "But when we consider the very peculiar nature of this offense [perjury], and that every person who appears as a witness in a court of justice is liable to be accused of it by those against whom his evidence tells, who are frequently the basest and most unprincipled of mankind . . . we shall see that the obligation of protecting witnesses from oppression, or annoyance, by charges, or threats of charges of having borne false testimony, is far paramount to that of giving even perjury its deserts. To repress that crime, prevention is better than cure; and the law of England relies, for this purpose, on the means provided for detecting and exposing the crime at the moment of commission, -- such as publicity, cross-examination, the aid of a jury, etc., . . . and it throws every fence round a person accused of perjury."

[Note 17] The commentator cited by Wigmore (see note 16, supra) concludes that "[t]he result [of the two-witness rule and other technical impediments to a perjury prosecution] accordingly is that in England little difficulty, comparatively speaking, is found in obtaining voluntary evidence for the purposes of justice; and although many persons may escape the punishment awarded by law to perjury, instances of erroneous convictions for it are unknown, and the threat of an indictment for perjury is treated by honest and upright witnesses as a `brutum fulmen' [i.e., an empty threat]."

[Note 18] See 7 Wigmore Section 2043 (3d ed. 1940); Annot., Perjury -- Circumstantial Evidence 88 A.L.R. 2d 852, 870-871 (1963); Perkins, Criminal Law 465-466 (2d ed. 1969). Commonwealth v. Bornstein, 269 Mass. 181 , 182-184 (1929), although not discussing the quantitative rule, is probably an example of the application of this exception. In the present case Coleman's perjury at the suppression hearing, discussed supra, also falls within this exception. Contrast Hammer v. United States, 271 U.S. 620, 626-628 (1926).

[Note 19] See Annot., Perjury -- Circumstantial Evidence 88 A.L.R. 2d 864-868 (1963); Perkins, op. cit. at 465 n.3; United States v. Wood, 39 U.S. (14 Pet.) 430, 437, 441 (1840); United States v. Collins, 272 F. 2d 650, 652-653 (2d Cir. 1959), cert. denied, 362 U.S. 911 (1960); United States v. Bergman, 354 F. 2d 931, 934 (2d Cir. 1966); United States v. Thompson, 379 F.2d 625 (6th Cir. 1967). The Nash case (see note 3, supra) furnishes an example of physical evidence that showed compellingly the impossibility of the defendant's sworn testimony.

[Note 20] See Annot., Perjury -- Circumstantial Evidence, 88 A.L.R. 2d at 870; Perkins, Criminal Law 465 n.3 (2d ed. 1969); United States v. Otto, 54 F.2d 277, 279 (2d Cir. 1931); United States v. Bergman, 354 F.2d 931, 934 (2d Cir. 1966).

[Note 21] Brightman v. United States, 386 F.2d 695, 697 (1st Cir. 1967) ("the corroboration must be of a substantial nature . . . `It is sufficient if the corroborating evidence tends to establish the defendant's guilt, and if such evidence together with the direct evidence is "inconsistent with the innocence of the defendant."'" [emphasis original] [citation omitted]). United States v. Collins, 272 F.2d 650, 652 (2d Cir. 1959), cert. denied, 362 U.S. 911 (1960) ("The test should be rather whether the evidence is of a quality to assure that a guilty verdict is solidly founded . . . [Here,] the testimony, although in a sense circumstantial, is absolutely inconsistent with Collins' innocence."). United States v. Bergman, 354 F.2d 931, 934 (2d Cir. 1966) ("circumstantial evidence of a highly reliable order"). United States v. Maultasch, 596 F.2d 19, 25n.9 (2d Cir. 1979) ("other evidence of independent probative value, circumstantial or direct, that is `of a quality to assure that a guilty verdict is solidly founded'" [citation omitted]). United States v. Neff, 212 F.2d 297, 306-307 (3d Cir. 1954) ("strong, clear, convincing and direct"). United States v. Rose, 215 F.2d 617, 621 (3d Cir. 1954) (same). Allen v. United States, 194 F.2d 664, 668 (4th Cir. 1912) ("to convict of perjury the government must produce testimony of a more direct and positive character than is required to justify a verdict of guilty of other offenses"). Van Liew v. United States, 321 F.2d 674, 679 (5th Cir. 1963) ("the Government had the burden of proving by clear, convincing and direct evidence to a moral certainty and beyond a reasonable doubt that the [d]efendant . . . . purposefully misstated the fact . . ."). United States v. Forrest, 639 F.2d 1224, 1226 (5th Cir. 1981) (". . . . the evidence must be strong, clear, convincing and direct. Where the government seeks to establish perjury by the testimony of one witness and corroborating evidence, the latter must be independent of the former and inconsistent with the innocence of the defendant.") United States v. Edwards, 443 F.2d 1286, 1294 (8th Cir.), cert. denied, 404 U.S. 944 (1971) ("substantial evidence excluding every other hypothesis than that of guilt"). United Stats v. Koonce, 485 F.2d 374, 381 (8th Cir. 1973) ("clear, convincing and direct evidence"). United States v. Armilio, 705 F.2d 939, 941 n.5 (8th Cir.), cert. denied, 104 S.Ct. 235 (1983) ("This `clear, convincing and direct evidence' language does not articulate a standard for weighing the evidence, but requires that the government present particular kinds of evidence . . . . "). Hart v. United States, 131 F.2d 59, 61 (9th Cir. 1942) ("clear, convincing, and direct").