The indispensable key to understanding the law of double jeopardy is to understand that there is no law of double *517jeopardy. This is not to say, however, that there are not many laws of double jeopardy. Double jeopardy, though it once was, is no longer a single species of legal doctrine. It is, now at least, an entire category; a genus embracing within its ever-widening folds no less than five distinct species: 1) former acquittal, 2) former conviction, 3) simultaneous jeopardy, 4) mistrial/retrial, and 5) collateral estoppel — each tracing back to a different origin, each serving a different purpose, each with a different set of implementing rules. The title “double jeopardy,” moreover, is periodically invoked by false claimants, such as the non-double jeopardy look-alikes of dual sovereignty1 and enhanced punishment2 and by the double jeopardy reject of compulsory joinder or “same transaction.”3 An unfailing compass through the resulting maze of at-times bewildering and frequently contradictory rules and pronouncements is the constant admonition “Think plural.”
The Present Case
Upon this appeal, Michael Butler relies exclusively upon the collateral estoppel sub-variety of the double jeopardy family of defenses. Collateral estoppel, to an extent not shared by true double jeopardy or res judicata, is stubbornly fact-bound. When looking at a first juridical event to *518determine its impact upon a pending second juridical event, we are concerned not with the legal implications of the first event, as we would be in the true double jeopardy or res judicata context,4 but only with findings of fact actually or almost certainly made by lay fact finders. We are indifferent to whether those findings were logical or internally consistent. Our exclusive concern is with, to the extent we can discern it, what was found as a matter of fact and not with what might have been found or what should have been found.
Butler initially went to trial in the Circuit Court for Howard County before a jury presided over by Judge Raymond J. Kane, Jr. He was charged with having been one of two assailants who together, during the early morning hours of August 12, 1989, made a murderous attack on each of two victims. The other assailant, concededly the triggerman, was Kent Tilghman.5 The victim Sherman Chenault died of two gunshot wounds to the neck and head. The victim Sharrell Hudson, girlfriend of the murdered Chenault, was shot in the face, suffering a fractured jaw as a result.
At his first trial, some of the charges against Butler for the attack on Chenault produced jury verdicts. As a result of those verdicts, other unresolved charges relating to the attack on Chenault were disposed of by applying traditional double jeopardy precepts. All of the charges relating to the attack on Sharrell Hudson, on the other hand, were left unresolved by a hung jury. Prior to the scheduled retrial on those charges, Butler interposed the plea in bar of *519collateral estoppel. Judge Cornelius P. Sybert, Jr. denied the plea and this interlocutory appeal followed.6
The Factual Background
The murder victim Sherman Chenault was dealing in drugs. On the fateful night, his attempted purchase of drugs led ultimately to his ambush and murder. Sharrell Hudson, who was simply along for the ride, was a random victim. Chenault, incidentally, had just been released from prison several weeks earlier.
On the afternoon of August 11, Chenault had been at Sharrell Hudson’s house from approximately 4:30 P.M. onward. By telephone, he had arranged to meet with Kent Tilghman at 10 P.M. that evening at a McDonald’s restaurant on Security Boulevard. After the meeting was arranged, Ms. Hudson saw Chenault in the bedroom taking “a whole lot of money” out of a metal box and counting the money. It was subsequently established that “the money” involved in this case was $17,000. After one abortive attempt to find the restaurant, Chenault returned to Ms. Hudson’s house and got her to go along as navigator. He put a gun and the metal box containing the money into a black vinyl bag, which he then placed in the trunk of his mother’s black Riviera. It was in that Riviera that the ultimate shootings took place. Ms. Hudson, apparently lacking a baby-sitter, took her five-year-old daughter along with her.
On that second expedition, they found both McDonald’s and the would-be sellers. Kent Tilghman had been driven to the rendezvous in a Toyota Célica owned and operated by Darcell Butler, the appellant’s sister. It was arranged that Chenault and his entourage in the Riviera would follow Tilghman and Darcell Butler in the Célica back to the home *520of the appellant’s mother. The appellant was present when the two groups arrived at his mother’s house. The appellant and Sharrell Hudson were, it turned out, old acquaintances; Ms. Hudson, incidentally, was inherently suspicious of the appellant and kept a wary eye on him throughout the evening. Chenault took the $17,000 from the metal box and handed it to Tilghman, who placed it in a yellow trash bag. According to Ms. Hudson, the appellant observed this transfer of cash. Immediately thereafter, Tilghman and the appellant adjourned upstairs for two or three minutes. What then followed seems to have been a scenario arranged for Chenault’s benefit. A beeper contact followed by a telephone contact was ostensibly made with Tilghman’s brother, Milton Tilghman, who was presumably setting up the time and place for the actual delivery of the drugs that Chenault had just constructively purchased. When the last signal apparently came in at about midnight, all concerned parties headed for the rendezvous in two separate cars.
While waiting for the telephoned communications to come in, Kent Tilghman had made several trips to or out the door, apparently with Sherman Chenault. Just preceding one such trip, the appellant was observed by Ms. Hudson looking out the door. By process of elimination, it appears likely that the yellow trash bag containing $17,000 had ended up in the trunk of the Toyota Célica before the group departed the elder Mrs. Butler’s home.
Sherman Chenault drove his Riviera, with girlfriend Sharrell Hudson in the right front seat, with drug seller Kent Tilghman in the rear passenger compartment, and with Ms. Hudson’s daughter in the back with Tilghman. The appellant Butler followed, alone in his sister’s Célica. The two cars drove to the 6600 block of Waning Moon Way in the Village of Owen Brown in Columbia. Kent Tilghman directed Chenault to pull into the parking lot of an apartment complex. The appellant Butler remained poised in the Célica on the road leading into the parking lot.
How the appellant came to be piloting the auxiliary vehicle had evidentiary significance. According to the ap*521pellant, he and his wife simply wanted Kent Tilghman to give them a ride home from the elder Mrs. Butler’s house to their own home in Baltimore. Kent Tilghman, however, conditioned the providing of a ride home upon the appellant’s willingness to drive out to Columbia in the auxiliary vehicle and then to chauffeur Kent Tilghman back to the departure point. Despite the logistical asymmetries of both time and distance, the appellant agreed. Why no one else, such as his sister Darcell, could have driven him and his wife home was not explained. Why it would not have been more logical for Sherman Chenault, who also lived in the Baltimore area, to have driven Kent Tilghman back to Baltimore, after their mutual trip to Columbia, was also not explained, except by the theory that both Kent Tilghman and the appellant knew that Sherman Chenault would, by that time, be dead.
The appellant’s driving of the extra car to Columbia arguably had such sinister significance. He was not, at first glance, a logical choice in that the Célica was not his car and in that he did not, moreover, even own a driver’s license. If the yellow trash bag containing $17,000 was in the trunk of the Célica, however, Kent Tilghman needed a trusted minion at the wheel. The appellant apparently had the qualifications. He and Kent Tilghman had served time together in a correctional facility in Hagerstown. The appellant had served eight years for robbery. Tilghman, furthermore, was romantically involved with the appellant’s sister Darcell. After Kent Tilghman and the appellant had their brief meeting upstairs at the Butler home, the appellant did not even have to ask to borrow the keys to the Célica. He already had them in his pocket.
The fact that the midnight rendezvous was to be in a quiet residential comer of Columbia was also very relevant. The stated reason given to Sherman Chenault and Sharrell Hudson for the selected rendezvous point was that brother Milton Tilghman, the apparent drug source, ostensibly lived in Columbia. Extensive records checks by the Howard County police, however, turned up no connection of any sort *522between anyone named Milton Tilghman and the town of Columbia or any other location in its vicinity. There was a Milton Tilghman, on the other hand, living in Baltimore. The appellant, significantly, acknowledged that he, as part of a job he had once had installing pools, had worked on a job not only in Columbia generally but in the small Village of Owen Brown specifically. The appellant, inferentially, knew the remote area selected for the execution.
As they came to a halt in the Village of Owen Brown but with the motor still running, Chenault asked Tilghman if his brother was there yet. Tilghman replied, “No, not yet” and immediately began shooting. Ms. Hudson did not see a gun but felt a sharp pain on the left side of her head. Chenault’s body fell on top of her. She heard car doors shutting and a car drive away. She looked up and saw that the Célica was gone. She was able to get her daughter out of the car and to knock on the door of an adjacent apartment. The police and an ambulance responded to the scene.
The First Set of Verdicts
After the uneventful granting of a motion for judgment of acquittal on Count 7, charging the transportation of a handgun, eight charges contained in seven counts remained against the appellant. Pour of them involved the attack on Sherman Chenault. The jury reached a verdict on two of the charges and was hung on the other two:
Attack on Sherman Chenault
Count 1 (1st Degree Murder) — Hung
Count 1 (Second Degree Murder) — Not Guilty
Count 3 (Use of Handgun) — Hung
Count 8 (Accessory After Fact) — Guilty
The apparent anomaly between a hung jury on first degree murder and an acquittal on second degree murder is readily explainable. The evidence, the State’s closing argument and the judge’s instructions on the alternative theory of felony-murder all highlighted the State’s primary reliance on felony-murder perpetrated in the course of an *523armed robbery. The murderous mens rea under that theory does not entail any intent to kill at all but only the intent to perpetrate the underlying felony. Any co-felon, including the appellant seated behind the wheel of the getaway car, would thus be endowed with a murderous mens rea even if he never remotely contemplated that a killing would take place. Second-degree murder, by contrast, requires the specific intent either to kill or to commit grievous bodily harm against the victim. Although second-degree murder of the intent-to-kill variety is thereby a lesser, included offense subsumed within premeditated murder, it is not a lesser included offense within felony-murder. Thus, the acquittal for second-degree murder is in no way inconsistent with the jury’s irresolution on felony-murder; nor, under double jeopardy principles, does it have any foreclosing effect on further proceedings with respect to felony-murder. That acquittal of second-degree murder was a non-factor in what followed. Wooten-Bey v. State, 67 Md.App. 606, 508 A.2d 1010 (1986); Wooten-Bey v. State, 308 Md. 534, 520 A.2d 1090 (1987). And see Huffington v. State, 302 Md. 184, 486 A.2d 200 (1983).
The other four charges arose out of the attack on Sharrell Hudson. The jury was hung on all four charges:
Attack on Sharrell Hudson
Count 2 (Assault to Murder) Hung
Count 4 (Use of Handgun) Hung
Count 5 (Assault) Hung
Count 6 (Battery) Hung
The Double Jeopardy Plea
When the State indicated its intention of retrying him on the six charges that had resulted in mistrials, the appellant interposed the plea in bar of double jeopardy. His argument was based upon the theory that the conviction for being an accessory-after-the-fact to the murder of Chenault had established, as a matter of law, that the appellant could not have been a principal in the murder. Osborne v. State, 304 Md. 323, 327 n. 3, 499 A.2d 170 (1985); Hawkins v. *524State, 87 Md.App. 195, 200-201 n. 8, 589 A.2d 524 (1991).7 With respect to the two remaining charges involving the attack on Sherman Chenault, Judge Sybert granted the motion to dismiss, apparently on grounds of traditional double jeopardy. It is unnecessary for us to analyze the propriety of that ruling. That disposition with respect to the charges involving Chenault was, in any event, of no influence on the other charges involving Sharrell Hudson.
One overarching principle of double jeopardy law is that with respect to crimes against the person involving multiple victims, each victim is, for double jeopardy computations, a separate unit of prosecution. Thus, none of the charges from the package of crimes against Sherman Chenault is remotely the “same offense” as any of the charges from the separate package of crimes against Sharrell Hudson. When the “same offense” is not involved, the law of double jeopardy is not involved, with the single and exclusive exception of its shadowy, interloping, post-1970 poor relation called collateral estoppel.
Lest the analysis become shamefully muddied, therefore, from this point on in any appellate opinion, no double jeopardy principle and no double jeopardy precedent should be invoked unless it be one dealing explicitly and exclusively with collateral estoppel as such. This is only a collateral estoppel case and no other aspect of double jeopardy is involved.
*525 A Shaky Pedigree: The Rush to Federalization
It is not to demean the importance of the collateral estoppel protection in a criminal setting to point out that its Fifth Amendment pedigree, and thereby its status as a part of double jeopardy law, is exceedingly suspect. Collateral estoppel, as a first cousin of res judicata, is a legal doctrine furthering the interest in finality that had its origins in civil practice. The Supreme Court first applied the doctrine in the context of a criminal case in 1916 in United States v. Oppenheimer, 242 U.S. 85, 37 S.Ct. 68, 61 L.Ed. 161 (1916). The opinion of Justice Holmes disclaimed any reliance on the Fifth Amendment and, in doing so, rejected the government’s argument that the Fifth Amendment Double Jeopardy Clause necessarily exhausts the panoply of protections available to a criminal defendant. The Supreme Court held that the enunciation of certain constitutional protections in the Double Jeopardy Clause did not imply that other protections, brought over from civil law, were not also available:
“The safeguard provided by the Constitution against the gravest abuses has tended to give the impression that when it did not apply in terms, there was no other principle that could. But the 5th Amendment was not intended to do away with what in the civil law is a fundamental principle of justice ...” (citation omitted).
37 S.Ct. at 69.
Our present point is that although collateral estoppel became a valuable protection as part of federal criminal law, it was not a part of double jeopardy law. Indeed, the Double Jeopardy Clause guards against a person being placed twice in jeopardy of life or limb “for the same offense.” Collateral estoppel does not involve double exposure for the same offense. If it did, it would be called “direct estoppel” and not “collateral estoppel.” Cook v. State, 281 Md. 665, 668, 381 A.2d 671 (1978). By definition, it deals with the possibly foreclosing effect of an earlier adjudication of a “different offense.” *526The Supreme Court further developed its collateral estoppel protection in a criminal context in Collins v. Loisel, 262 U.S. 426, 43 S.Ct. 618, 67 L.Ed. 1062 (1923); Sealfon v. United States, 332 U.S. 575, 68 S.Ct. 237, 92 L.Ed. 180 (1948); United States v. Williams, 341 U.S. 58, 71 S.Ct. 595, 95 L.Ed. 747 (1951) and Yates v. United States, 354 U.S. 298, 77 S.Ct. 1064, 1 L.Ed.2d 1356 (1957). In none of those opinions did the Supreme Court give the slightest intimation that the constitutional law of double jeopardy was remotely involved. In the Collins v. Loisel opinion, Justice Brandéis pointed out that the protection extended did not arise out of double jeopardy law and that the Double Jeopardy Clause “was not intended to supplant the fundamental principle of res judicata in criminal cases.” 262 U.S. at 430, 43 S.Ct. at 619. The point again is not that the protection did not exist but simply that it was not an aspect of double jeopardy protection. In Sealfon v. United States, Justice Douglas expressly stated that the collateral estoppel protection was applicable even though the Double Jeopardy Clause was not.
How then did this relatively extensive body of doctrine that traditionally had been separate and distinct from double jeopardy become miraculously transformed into an aspect of double jeopardy? The fortunes of the doctrinal pedigree were wrapped up in the larger fortunes of the massive federalization of criminal procedure that occurred between 1960 and 1970. Hoag v. New Jersey, 356 U.S. 464, 78 S.Ct. 829, 2 L.Ed.2d 913 (1958), was the first attempt to make the federal collateral estoppel law apply to a state criminal prosecution. The Supreme Court acknowledged that the petitioner had no legitimate double jeopardy claim. “At the outset it should be made clear that Petitioner has not been twice put in jeopardy for the same crime.” 356 U.S. at 466, 78 S.Ct. at 832. The Supreme Court had serious reservations about whether its collateral estoppel law, even as a part of general due process, was of constitutional dimension:
*527“Despite its wide employment, we entertain grave doubts whether collateral estoppel can be regarded as a constitutional requirement. Certainly this Court has never so held.”
356 U.S. at 471, 78 S.Ct. at 834. In Hoag, the Supreme Court ultimately declined to hold that its collateral estoppel protection was a part of fundamental due process and, therefore, binding upon the states.
Twelve years after Hoag v. New Jersey, Ashe v. Swenson, 397 U.S. 436, 90 S.Ct. 1189, 25 L.Ed.2d 469 (1970), presented a factual situation that was indistinguishable from Hoag. The primary change between 1958 and 1970 was in the makeup of the Court. If federal collateral estoppel law applied, the defendant, Ashe, would clearly prevail. As a matter of general due process, however, the Court was obviously reluctant to overturn a precedent but twelve years old.
One opportunity presented itself. A year earlier, Benton v. Maryland, 395 U.S. 784, 89 S.Ct. 2056, 23 L.Ed.2d 707 (1969), had overruled Palko v. Connecticut, 302 U.S. 319, 58 S.Ct. 149, 82 L.Ed. 288 (1937), and held for the first time that the Double Jeopardy Clause of the Fifth Amendment was part of the due process of law guaranteed by the Fourteenth Amendment against State action. Incorporated into the Fourteenth Amendment, thereby, was not simply the core principle of double jeopardy but all of the federal case law over the years that had interpreted and implemented the Double Jeopardy Clause. If, therefore, federal collateral estoppel law were deemed to have been historically a part of the federal double jeopardy protection, it would ipso facto become binding on the states with no necessity of overruling Hoag v. New Jersey. It could be incorporated indirectly rather than directly. The analysis by which the Supreme Court then overturned 54 years of precedent, including disclaimers from such Olympians as Holmes and Brandéis that collateral estoppel was not double jeopardy, was casual to the point of being cavalier:
*528“The ultimate question to be determined, then, ... is whether this established rule of federal law is embodied in the Fifth Amendment guarantee against double jeopardy. We do not hesitate to hold that it is. For whatever else that constitutional guarantee may embrace, it surely protects a man who has been acquitted from having to 'run the gantlet’ a second time.” (footnote and citations omitted) (emphasis supplied).
397 U.S. at 445-446, 90 S.Ct. at 1195. The nonchalant, almost ipse dixit tone of so significant a holding occasioned the outraged cry of Chief Justice Burger in dissent:
“[T]he only expressed rationale for the majority’s decision is that Ashe has ‘run the gantlet’ once before. This is not a doctrine of the law or legal reasoning but a colorful and graphic phrase, which, as used originally in an opinion of the Court written by Mr. Justice Black, was intended to mean something entirely different. The full phrase is ‘run the gantlet once on that charge ... ’; it is to be found in Green v. United States, [355 U.S. 184, 190, 78 S.Ct. 221, 225, 2 L.Ed.2d 199] (1957), where no question of multiple crimes against multiple victims was involved____ This Court held nothing more than that Green, once put in jeopardy — once having ‘run the gantlet ... on that charge’ — of first degree murder, could not be compelled to defend against that charge again on retrial.
Today’s step in this area of constitutional law ought not to be taken on no more basis than casual reliance on the ‘gantlet’ phrase lifted out of the context in which it was originally used. This is decision by slogan.” (citation omitted) (emphasis in original) (emphasis supplied).
397 U.S. at 465, 90 S.Ct. at 1205 (dissenting opinion by Burger, C.J.).8
*530The point to be made in recounting this constitutional revisionism is that when two or more distinct bodies of doctrine — stemming from different origins, serving differ*531ent albeit related purposes, and implemented by different rules — are generically labeled with the same “umbrella term,” there arises the recurring semantic danger that a statement in the case law, correct in its original context of one variety of double jeopardy law, may be randomly misapplied to some other variety where it does nothing but generate confusion. This is why any discussion of collateral estoppel should meticulously confine itself to nothing but collateral estoppel cases.
Collateral Estoppel Under Maryland Common Law
Maryland has also long recognized the collateral estoppel defense as one available in a criminal trial as part of the common law of this state. It was always recognized, however, as something separate and distinct from the common law of double jeopardy in this state. Cook v. State, 281 Md. 665, 668 n. 2, 381 A.2d 671 (1978) (“Appellant alleges no double jeopardy violation in the present case.”); Rouse v. State, 202 Md. 481, 484-485, 97 A.2d 285 (1953) (“The appellant concedes expressly that former jeopardy would not prevent his conviction in the present case, because that defense can be invoked only when the crimes are the same in law and in fact.”); Scarlett v. State, 201 Md. 310, 318-319, 93 A.2d 753 (1953); State v. Coblentz, 169 Md. 159, 164, 180 A. 266 (1935) (“The defendant does not contend that a plea of former jeopardy could be interposed in this case, nor could it be, for the statutory offenses for which the defendant was indicted are so dissimilar as not to justify such a plea.”). Indeed, on three earlier occasions collateral estoppel-type arguments were raised under the rubric of double jeopardy and were soundly rejected. Gilpin v. State, 142 Md. 464, 121 A. 354 (1923); Novak v. State, 139 Md. 538, 115 A. 853 (1921); Watson v. State, 105 Md. 650, 66 A. 635 (1907).
In any event, it is not the narrow Maryland version of comfortable 19th Century double jeopardy law but the vastly distended federal version that now commands our attendance.
*532 The Federal Case Law
The Supreme Court, since Ashe v. Swenson, has considered collateral estoppel on six occasions.9 Three of those were summary dispositions rendered by per curiam opinions. Simpson v. Florida, 403 U.S. 384, 91 S.Ct. 1801, 29 L.Ed.2d 549 (1971) (although a defendant may rely on an earlier acquittal to invoke collateral estoppel at the subsequent trial of a closely related offense, the State may not rely on an even earlier conviction to cancel out the decisional impact of the intervening acquittal); Harris v. Washington, 404 U.S. 55, 92 S.Ct. 183, 30 L.Ed.2d 212 (1971) (evidentiary error arguably leading to an earlier acquittal does not preclude defendant from relying on that acquittal for collateral estoppel purposes); Turner v. Arkansas, 407 U.S. 366, 92 S.Ct. 2096, 32 L.Ed.2d 798 (1972) (doctrine of collateral estoppel applies even when state procedural rules precluded joining all charges against a defendant in a single trial).
*533The two cases of One Lot Emerald Cut Stones v. United States, 409 U.S. 232, 93 S.Ct. 489, 34 L.Ed.2d 438 (1972), and United States v. One Assortment of 89 Firearms, 465 U.S. 354, 104 S.Ct. 1099, 79 L.Ed.2d 361 (1984), both dealt, for purposes of their collateral estoppel analysis, with the distinction between the criminal and civil burdens of persuasion.10 Both cases held that an earlier acquittal on a criminal charge would not collaterally estop a subsequent civil forfeiture proceeding notwithstanding the fact that the forfeiture would depend upon the establishment of facts already decided in the defendant’s favor in the criminal trial. The holdings were that the failure of the prosecution, in a criminal trial, to prove a fact beyond a reasonable doubt did not necessarily establish that the fact in issue may not well have been proved at least by a preponderance of the evidence, all the persuasion that is required for a forfeiture.11 See in this regard, Dunn v. State, 65 Md.App. 637, *534644-647, 501 A.2d 881 (1985) (One strong reason, among others, that an acquittal in a criminal trial does not collaterally estop the State from proving a probation violation based upon the same alleged criminality is that the finding of a violation may be based upon a lower standard of persuasion than is required for the criminal conviction).
Dowling v. United States, 493 U.S. 342, 110 S.Ct. 668, 107 L.Ed.2d 708 (1990), was the first Supreme Court decision since Ashe v. Swenson to provide a plenary examination of collateral estoppel principles. Ashe and Dowling together represent, effectively speaking, the sum total of collateral estoppel law in its new, constitutional manifestation. The core meaning was stated by Ashe v. Swenson, 397 U.S. at 443, 90 S.Ct. at 1194:
“[WJhen [1] an issue of ultimate fact has [2] once been determined by [3] a valid and final judgment, that issue cannot again be litigated [4] between the same parties in any future lawsuit.” (emphasis supplied).
“Ultimate Fact” Versus “Evidentiary Fact”
The first question that arises is that of what is “an issue of ultimate fact?” Every fact that a defendant has *535once had litigated in his favor is not constitutionally shielded from being relitigated against him a second time around. Some facts, however, are so shielded. The requirement is that they be “ultimate.” The availability of the collateral estoppel defense in Ashe and the unavailability of that defense in Dowling effectively bracket the target of “an issue of ultimate fact.”
Bob Ashe, and others, allegedly pulled off an armed holdup of an all-night poker game, relieving six separate poker players of money and other property. Ashe went to trial initially for the robbery of one of the victims. The only trial issue was that of his criminal agency. The eyewitness identification was exceedingly shaky. A jury returned a verdict of not guilty. When the State subsequently sought to try Ashe for the robbery of one of the other poker players, he interposed the plea of collateral estoppel.
To prove him guilty of the second robbery, it would have been necessary to prove that he was there at the robbery scene. The only explanation for the earlier acquittal, however, was that the first jury had necessarily concluded, as a matter of fact, that he was not there. Having received the benefit of a determination in the trial of one offense that he had not been at the crime scene, he sought to bar the State from relitigating that fact against him at the trial of a different offense. He succeeded. Absent some theory of constructive participation not there pertinent, presence at the crime scene was “an issue of ultimate fact.”
Reuben Dowling also received the benefit of an initial determination of fact in his favor. He was tried, in the Virgin Islands, for the armed robbery of Ms. Vena Henry. He was acquitted. He was subsequently tried for the armed robbery of a bank, which had occurred several weeks before the robbery of Ms. Henry. Because of a retrial necessitated by an earlier hung jury, however, the bank robbery trial in issue came after the trial for the Vena Henry robbery. At the bank robbery trial, the government introduced the testimony of Ms. Henry describing, in some *536detail, Dowling’s robbery of her. Various circumstances— the wearing of a similar ski mask, the carrying of a similar gun, the association with a confederate named Delroy Christian — were a common denominator of the two robberies. Ms. Henry was able to make a better identification of Dowling, moreover, than were the eyewitnesses from the bank.
Understandably, Dowling was chagrined at having offered against him testimony of a crime for which he had been acquitted. The Supreme Court held, for several reasons, that collateral estoppel did not bar the introduction of Vena Henry’s testimony. His participation in the robbery of her — the issue of fact that he sought to have barred from relitigation — was not an issue of ultimate fact at the trial of the bank robbery (although it had, of course, been an issue of ultimate fact at the earlier trial). It was simply an “evidentiary fact” offered, as “other crimes” evidence, to strengthen the identification of him as the bank robber. Collateral estoppel does not bar the relitigation of all factual issues. It is required that the issue as to which preclusion is sought be one of ultimate fact. The Supreme Court was very clear as to this limitation on the collateral estoppel doctrine, holding at 493 U.S. at 348, 110 S.Ct. at 672:
“[U]nlike the situation in Ashe v. Swenson, the prior acquittal did not determine an ultimate issue in the present case. This much Dowling concedes, and we decline to extend Ashe v. Swenson and the collateral-estoppel component of the Double Jeopardy Clause to exclude in all circumstances, as Dowling would have it, relevant and probative evidence that is otherwise admissible under the Rules of Evidence simply because it relates to alleged criminal conduct for which a defendant has been acquitted.”
In Simms v. State, 83 Md.App. 204, 213-216, 574 A.2d 12 (1990), the defendant at an earlier sentencing proceeding had received the benefit of a trial judge’s fact finding that the evidence of a prior conviction was legally insufficient and that that conviction could not, therefore, be used for *537sentence enhancing purposes. Following a separate conviction in another county, the State sought to prove that same prior conviction against the defendant for purposes of sentence enhancement. We held that the existence of the earlier conviction was not an issue of ultimate fact capable of engaging the gears of collateral estoppel.12
In Cook v. State, 281 Md. 665, 381 A.2d 671 (1978), the Court of Appeals held that an earlier suppression ruling in a defendant’s favor did not collaterally estop the State from relitigating that same search and seizure as a predicate for submitting evidence at a subsequent trial of another offense. At the trial for armed robbery, the constitutional propriety of an antecedent search for evidence was a mere evidentiary fact and not an issue of ultimate fact.
The appellant, Michael Butler, satisfies this requirement of the collateral estoppel defense. He claims that his conviction at the first trial as an accessory-after-the-fact established that he did not aid and abet Kent Tilghman as a principal in the second degree but only came to the aid of Tilghman after the fact of the murderous attacks. Whether the appellant can persuade us that the jury necessarily found this as a matter of fact is yet to be determined. For immediate purposes for this aspect of the analysis, however, the “fact” urged upon us is indeed an issue of ultimate fact.
Upon the retrial of the appellant for the murderous assault upon Sharrell Hudson, the State is relying upon the theory not that the appellant was the triggerman but that the appellant was a principal in the second degree, aiding and abetting Kent Tilghman. The arguable fact that the appellant did not aid and abet Kent Tilghman is, by definition, an issue of ultimate fact, the relitigation of which the appellant seeks to have estopped. On this aspect of collateral estoppel, the appellant qualifies.
*538 What Is “A Valid and Final Judgment”?
One of the requirements is that there have been an earlier determination in a defendant’s favor made by a “valid and final judgment.” A “valid and final judgment” contemplates that “the earlier proceeding must have ended with a final judgment or ‘final determination’ of the issue.” Bowling v. State, 298 Md. 396, 402, 470 A.2d 797 (1984); Cousins v. State, 277 Md. 383, 398, 354 A.2d 825 (1976). “[N]o estoppel can be based upon an informal probation-like disposition which involved neither verdict nor judgment.” LaFave and Israel, Criminal Procedure (1985), § 17.4 at 676.
The entry in a criminal case of a nol pros by the State, for instance, has been held by us in Mitchell v. State, 44 Md.App. 451, 456-460, 409 A.2d 260 (1979), to be “a premature termination [that] does not suffice to trigger the ‘constitutionalized’ doctrine of collateral estoppel.” 44 Md.App. at 460, 409 A.2d 260. Judge (now Chief Judge) Wilner analyzed this inadequacy of the nol pros as a trigger at 44 Md.App. 459-460, 409 A.2d 260:
“We are aware of no instance in which the doctrine has been held triggered by a nol pros.
As the Supreme Court pointed out in Ashe, collateral estoppel is a doctrine that arose in the civil, rather than the criminal, law; and it is therefore not inappropriate to look at how the doctrine has been viewed and applied in civil cases. The touchstone of the doctrine in the civil law is the requirement that the earlier determination — that which forms the basis of the estoppel — be a judgment on the merits; and in that context, it has generally been held that a judgment for the defendant entered upon a voluntary nonsuit, nolle prosequi, or non prosequitur is not a judgment on the merits and does not bar a subsequent action.
The reason for this is evident from the very nature of a nonsuit, a voluntary dismissal. It terminates the action prior to a decision on the merits. The case is ended without the adjudication of disputed facts necessary to *539render a judgment on the merits. This is equally true with a nol pros entered in a criminal case. Although if entered after jeopardy attaches and without the defendant’s consent, it serves as an acquittal and, by virtue of traditional double jeopardy principles bars reprosecution for the same offense, it nevertheless, like its cousin the nonsuit, terminates the cause without an adjudication of disputed facts. The defendant is deemed acquitted and the imputation of guilt is removed, but there has been no adjudication that he did not commit the offense alleged.” (emphasis in original) (citations omitted).
In Stevens v. State, 34 Md.App. 164, 169-171, 366 A.2d 414 (1976), we held that a determination of Probation Before Verdict was not the type of valid and final judgment that could activate collateral estoppel. In Perkins v. State, 26 Md.App. 526, 528-529, 339 A.2d 360 (1975), we held that the starting point for collateral estoppel was “an issue previously determined by a valid and final judgment” and that “[a] preliminary hearing is obviously not a final judgment in any sense.” Findings and rulings made at a suppression hearing do not qualify as valid and final judgments. Cook v. State, 281 Md. 665, 668-670, 381 A.2d 671 (1978) (“[I]t is widely agreed that exclusionary orders resemble mere evidentiary rulings, interlocutory in nature and hence are intrinsically non-final.” Id. at 670, 381 A.2d 671).
In Ohio v. Johnson, 467 U.S. 493, 500 n. 9, 104 S.Ct. 2536, 2541 n. 9, 81 L.Ed.2d 425, 434 n. 9 (1984), the Supreme Court observed that:
“[T]he taking of a guilty plea is not the same as an adjudication on the merits after full trial, such as took place in Ashe v. Swenson.”
Hoag v. New Jersey, 356 U.S. 464, 470, 78 S.Ct. 829, 834, 2 L.Ed.2d 913, 919 (1958) (quoting Restatement, Judgments, § 68(1)), captured the essence of this requirement of an actual decision on the ultimate merits with its use of the phrase “actually litigated”:
*540“A common statement of the rule of collateral estoppel is that ‘where a question of fact essential to the judgment is actually litigated and determined by a valid and final judgment, the determination is conclusive between the parties in a subsequent action on a different cause of action.’ ” (emphasis supplied).
Thus, such adjudicatory phenomena as the entry of a nol pros, the awarding of probation before verdict, the determination in a preliminary hearing that no probable cause exists to hold a defendant, the ruling on a constitutional issue at a suppression hearing, and the taking of a guilty plea are not valid and final judgments within the particular contemplation of collateral estoppel law. Some of these adjudicatory phenomena, to be sure, may have binding legal effect in the context of res judicata or traditional double jeopardy law. They do not, however, have binding fact-finding effect within the very different context of collateral estoppel law.
Once again, the appellant here qualifies. The triggering event he relies upon was the jury verdict of guilty on the charge of being an accessory after the fact to murder, finalized by the sentence. That was “a valid and final judgment.”
There is, to be sure, a bit of “reverse English” on the triggering event here, for the appellant is relying, atypically, upon a conviction rather than an acquittal. Because the triggering event for collateral estoppel is so frequently an earlier acquittal, much of the case law has fallen into the easy linguistic habit of referring to the necessary activating event as an “acquittal.” See, e.g., Powers v. State, 285 Md. 269, 283-284, 401 A.2d 1031 (1979) (emphasis supplied). (“Thus, the primary purpose of the doctrine of collateral estoppel is to protect an accused from the unfairness of being required to relitigate an issue which has once been determined in his favor by a verdict of acquittal”) Such references, however, are nothing more than careless language.
*541The earlier ruling, as a matter of course, must be considered by wily counsel to be “a ruling in the defendant’s favor” now or he would not invoke it. It need not necessarily have been in the defendant’s favor when made, however. “In one’s favor,” moreover, is a relative notion that may shift with the tides. One seeks to make the best of even a bad situation. A verdict of guilty of manslaughter, for instance, is not in one’s favor, compared to going free. It is most definitely in one’s favor, however, compared to a verdict of guilty of first-degree murder. Thus, the appellant now seeks to find, as he may, some present utility in his earlier conviction as an accessory after the fact. Judge Adkins addressed just such a situation in Robinson v. State, 307 Md. 738, 742, 517 A.2d 94 (1986):
“The State counters by contending that in the criminal context a defendant may raise the bar of collateral estoppel only when an ‘issue of ultimate fact’ has been determined in her favor by an acquittal. Since Robinson was convicted of assault with intent to disable, says the State, she cannot raise collateral estoppel.
We reject the State’s position. Although both Ashe and Powers v. State, 285 Md. 269, 401 A.2d 1031 involved acquittals, the language we have just quoted from the former case makes it clear that the critical consideration is whether ‘an issue of ultimate fact’ has been determined in favor of a defendant. The process by which that determination is made, whether by acquittal or conviction, is not critical.” (emphasis in original).
“... Between the Same Parties”
For the first adjudication to have, via collateral estoppel principles, a binding fact-finding effect upon a projected second adjudication, there must, as a matter of course, be an identity of parties between the two adjudications. This is why Standefer v. United States, 447 U.S. 10, 100 S.Ct. 1999, 64 L.Ed.2d 689 (1980), and Carbaugh v. State, 294 Md. 323, 329-330, 449 A.2d 1153 (1982), declined to let a defendant, charged with aiding and abetting, invoke an *542earlier acquittal of the first-degree principle as a preclusive event. Different defendants were parties in the respective prosecutions. This is why Bailey v. State, 303 Md. 650, 660-662, 496 A.2d 665 (1985), refused to permit a defendant to interpose his acquittal in one state as a bar to his subsequent prosecution in another state for precisely the same criminal episode. Different states were parties in the respective prosecutions.
In the present case, however, this necessary identity of parties poses no problem. In both the first adjudication and the projected second adjudication, the parties were immutably the State of Maryland and the appellant.
What Is A Factual Determination?
In assessing whether a factual determination in an earlier litigation should have a foreclosing effect upon the projected relitigation of the same factual issue, several procedural incidents must be considered.
A. The Respective Burdens of Persuasion:
For an earlier litigation of a fact effectively to preclude the relitigation of that fact, it is necessary that the respective burdens of persuasion be in proper alignment. If a defendant only prevailed on an earlier occasion because the State was shouldering a heavier burden of persuasion, that victory will not have decided in advance a projected relitigation requiring only a lesser burden of persuasion. This is why in a series of cases involving acquittals for an underlying crime, the government was not precluded from relitigating the happening of the crime in subsequent forfeiture proceedings. That the first jury was not 85% persuaded that the defendant perpetrated the underlying crime does not imply necessarily that they were not 51% persuaded. That first determination, therefore, did not settle in the defendant’s favor the factual issue that was submitted to the second jury. Helvering v. Mitchell, 303 U.S. 391, 58 S.Ct. 630, 82 L.Ed. 917 (1938); One Lot Emerald Cut Stones v. United States, 409 U.S. 232, 93 S.Ct. 489, 34 L.Ed.2d 438 (1972); United States v. One Assortment of 89 *543Firearms, 465 U.S. 354, 104 S.Ct. 1099, 79 L.Ed.2d 361 (1984). In United States v. One Assortment of 89 Firearms, the Supreme Court explained, at 465 U.S. at 361-362, 104 S.Ct. at 1104:
“[The acquittal did] not prove that the defendant is innocent; it merely proves the existence of a reasonable doubt as to his guilt____ [T]he jury verdict in the criminal action did not negate the possibility that a preponderance of the evidence could show that [the defendant] was engaged in an unlicensed firearms business____ It is clear that the difference in the relative burdens of proof in the criminal and civil actions precludes the application of the doctrine of collateral estoppel.”
Even where the trial sequence is not a criminal trial followed by a civil trial but, rather, a criminal trial followed by another criminal trial, the alignment of the burdens of persuasion will still be the controlling factor. In Dowling v. United States, 493 U.S. 342, 110 S.Ct. 668, 107 L.Ed.2d 708 (1990), Dowling was acquitted of the armed robbery of Vena Henry because the jury was not persuaded beyond a reasonable doubt that he had committed the crime. When evidence as to that robbery, however, was offered to a second jury trying Dowling for the robbery of a bank, that evidence was offered only as “other crimes” evidence, which the jury was permitted to consider as an evidentiary fact if it could “reasonably conclude that the act occurred and that the defendant was the actor.” Under the circumstances, the earlier failure of the State to satisfy the higher burden of persuasion as to a fact did not estop the State from relitigating that fact under a lesser burden of persuasion. The Supreme Court explained, at 493 U.S. 349, 110 S.Ct. 672:
“Our decision is consistent with other cases where we have held that an acquittal in a criminal case does not preclude the Government from relitigating an issue when it is presented in a subsequent action governed by a lower standard of proof.”
*544To restate the same principle from the vantage point of the defendant’s effective obligation rather than from that of the State’s obligation, if the defendant may have prevailed in the first litigation because he was only required to generate a little bit of doubt, that does not necessarily establish that he would still have prevailed if he had been required, as he will be in the second litigation, to generate a lot more doubt.
The appellant in this case has no problem with burdens of persuasion. The State had the same burden of persuasion in proving him guilty beyond a reasonable doubt of accessoryship after the fact to the murder of Sherman Chenault as it will have in retrying him for assault with intent to murder Sharrell Hudson.
B. The Allocation of the Burden of Proof:
The moving party has the burden of proof. A party seeking to estop an opposing party from litigating a fact bears the burden of showing entitlement to that estoppel. In Dowling v. United States, the challenged evidence as to the robbery of Vena Henry, notwithstanding Dowling’s acquittal on that charge, was “admissible because Dowling did not demonstrate that his acquittal in his first trial represented a jury determination that he was not one of the men who entered Ms. Henry’s home.” 493 U.S. at 350, 110 S.Ct. at 673 (emphasis supplied). The Supreme Court expressly stated that in ascertaining, if possible, why a first jury reached the verdict it did and what it may have decided, the burden of proof is upon the defendant:
“The Courts of Appeals have unanimously placed the burden on the defendant to demonstrate that the issue whose relitigation he seeks to foreclose was actually decided in the first proceeding. We see no reason to depart from the majority rule in this case.” (citations omitted).
Id. In Holloway v. State, 14 Md.App. 703, 711, 288 A.2d 652 (1972), Judge Orth was equally emphatic in speaking for this Court:
*545“[T]he rule of collateral estoppel in criminal cases requires the court ... to determine whether the trier of fact could have grounded its verdict upon an issue other than that which the defendant seeks to foreclose from consideration____ [A] Holloway [the defendant] was the moving party in the motion to dismiss, it was encumbent upon him to produce proper evidence sufficient for the court to rule.” (emphasis supplied).
When, therefore, the circumstances themselves are too ambiguous or the record itself too inadequate to permit an accurate assessment of what factual findings the jury must have made, it is the defendant, as moving party, who necessarily loses. In Holloway v. State, we held the defendant not to be entitled to collateral estoppel largely because of the inadequacy of the record:
“On the state of the record before us we cannot say as a matter of constitutional fact, see Ashe v. Swenson, supra [397 U.S.], at 443 [90 S.Ct. at 1194], that the prosecution of Holloway under indictment 752 on the charge of assaulting Billy Anderson was barred under the doctrine of collateral estoppel____ [W]e cannot ascertain that the verdicts were rendered without alternative ultimate facts outstanding. That is, we are unable to conclude from the record that a rational trier of fact could not have grounded its verdict upon an issue other than that which Holloway seeks to foreclose from consideration ... The short of it is that Holloway did not produce sufficient evidence to support his motion to dismiss and such evidence did not otherwise get into the record.”
14 Md.App. at 715-716, 288 A.2d 652.
C. What Must the Jury Have Decided?
The ultimate fact that the appellant seeks to establish is that he did not, as a principal in the second degree, aid and abet Kent Tilghman in the armed robbery that eventuated in the shooting of the two victims. His authority for the existence of such a fact rests entirely upon his conviction as accessory after the fact to the murder. Even
*546accepting, purely arguendo, that State v. Hawkins, 326 Md. 270, 604 A.2d 489 (1992), reversing of Hawkins v. State, 87 Md.App. 195, 589 A.2d 524 (1991), will not retroactively impact on his situation, we note that the appellant looks to our Hawkins v. State to distill from it a proposition of law. The proposition that guilt of accessoryship after the fact precludes guilt as a principal is something that, legally and logically, may have legal impact. Within the context of a collection of related offenses constituting the “same offense” or involving the same victim, such legal impact may be significant. Involved is the avoidance of inconsistent verdicts because of mutual exclusivity.
When dealing with a different victim and with a different offense, however, only collateral estoppel is arguably involved. Collateral estoppel, unlike res judicata, is not triggered by a verdict’s legal impact but only by a jury’s factual determination. The burden is upon the appellant to prove that the jury necessarily decided, as a matter of fact, that he was not a principal. The law, emanating from the conviction as an accessory, may establish that the appellant could not have been a second-degree principal in the murder of Sherman Chenault. For whatever happened in that case to have any collateral impact upon the charges involving Sharrell Hudson, however, what the law may establish does not matter. The only thing that matters is what the jury necessarily decided.
Ashe v. Swenson, at 397 U.S. at 444, 90 S.Ct. at 1194, described the process of trying to read the collective mind of the jury:
“Where a previous judgment of acquittal was based upon a general verdict, as is usually the case, this approach requires a court to ‘examine the record of a prior proceeding, taking into account the pleadings, evidence, charge, and other relevant matter, and conclude whether a rational jury could have grounded its verdict upon an issue other than that which the defendant seeks to foreclose from consideration.’ The inquiry ‘must be set in a prac*547tical frame and viewed with an eye to all the circumstances of the proceedings.’ ” (footnote omitted).
Looking at the entire trial record, the suggestion that this jury must have concluded that the appellant borrowed a car to follow another car all the way to Columbia in the wee hours of the morning simply as an unwitting chauffeur is untenable. The appellant cannot persuade us of that if there are any rational alternative explanations for the package of verdicts. There are.
Since all of the State’s evidence pointed toward the appellant’s being an aider and abettor rather than the trigger-man, the jury was obviously troubled with the problem of the murderous mens rea. The acquittal on the charge of the second-degree murder of Chenault, coupled with the hung jury on the first-degree murder of Chenault, strongly implies that the jury might not have believed that the appellant actually intended that anyone should be killed rather than simply robbed. The verdicts also imply that the jury may have had more trouble with the subtler notion that a participant in an underlying felony can nonetheless be responsible for a resultant murder even if he did not intend it. Trouble with the concept of attributing a mens rea from one intended crime to another unintended crime is, whatever the law may say, grist for mill of the layman’s mind.
This jury, because of the subtle attributive mens rea problem, did not “determine” any “issue of ultimate fact” by any “valid and final judgment” as to whether the appellant participated as an aider and abettor in Kent Tilghman’s unquestioned crimes. Hung juries on six of eight charges — both the first-degree murder and the handgun charge involving Chenault and all four charges involving Sharrell Hudson — attest to that absence of any factual determination as to the “issue of ultimate fact” that the appellant claims was decided.
The appellant seeks to dismiss the jury’s obvious irresolution on six of eight charges by citing to a comment in *548Powers v. State, 285 Md. 269, 285, 401 A.2d 1031 (1979), repeated in Ferrell v. State, 318 Md. 235, 249, 567 A.2d 937 (1990), that “In Maryland, a mistrial is equivalent to no trial at all.” The statement was taken out of context from Cook v. State, 281 Md. 665, 671, 381 A.2d 671 (1978), which simply stated that a mistrial is not a final determination and, therefore, decides no question of fact. A nonverdict, to be sure, does not affirmatively establish one of the opposing poles for a set of inconsistent verdicts. The historic fact of these six mistrials, however, in the total context of this particular case lends substantial circumstantial insight into what this jury probably did and probably did not decide. It is the appellant, of course, who must persuade us that this jury almost certainly decided that the appellant did not aid and abet in the crimes of August 11-12, 1989. The six mistrials are strong factual evidence that they decided no such thing. Some jurors obviously did think that the appellant was a participant or there would have been six additional acquittals instead of six mistrials. The only thing we can say with certainty that the jury did determine was that the appellant had no specific intent to kill Sherman Chenault.
The comment from Powers, moreover, has been thoroughly repudiated by Wooten-Bey v. State, 308 Md. 534, 520 A.2d 1090 (1987).13 At a single trial, a jury acquitted the *549defendant of premeditated murder but was hung on the theory of felony-murder. The appellant pleaded his acquittal as a bar to his retrial for the “same offense” of murder, albeit under a different theory of guilt. Had there been sequential prosecutions (as distinct from sequential trials), the defendant’s position would have been unassailable. If the retrial following mistrial were but a continuation of a single original jeopardy, on the other hand, the plea in bar of double jeopardy would be unavailing. Mauk v. State, 91 Md.App. 456, 605 A.2d 157 (1992). The defendant sought to avoid the impact of continuing jeopardy by invoking Powers for the proposition that “a mistrial is equivalent to no trial at all.” 285 Md. at 285, 401 A.2d 1031. The Court of Appeals, through Judge Orth, totally repudiated the idea:
“He claims that ‘[t]he effect of the mistrial as to felony murder was the same as if it had never been before the jury,’ looking to Powers v. State, 285 Md. 269, 285, 401 A.2d 1031 (1979) and Cook v. State, 281 Md. 665, 671, 381 A.2d 671 (1978)____ He maintains that since he was, in effect, not prosecuted for felony murder by reason of the mistrial, his acquittal on premeditated murder precludes, under the general rule, a second prosecution on the theory of felony murder for the same homicide. The Court of Special Appeals declined to accept this logic. Its view was that, for double jeopardy purposes, ‘the declaration of a mistrial for manifest necessity is the equivalent *550of a reversal of a conviction on appeal.’ Wooten-Bey, 67 McLApp. at 610, 508 A.2d 1010____
We agree.”
308 Md. at 542, 520 A.2d 1090. The Court of Appeals completely affirmed our decision of Wooten-Bey v. State, 67 Md.App. 606, 508 A.2d 1010 (1986), wherein Judge Adkins reasoned that the entitlement of a case to go forward to retrial following a mistrial, free from extraneous interference, should be analyzed just as if it were a conviction reversed upon appeal:
“The appellant ... seems to suggest that since a mistrial is equivalent to no trial at all, Cook v. State, 281 Md. 665, 671, 381 A.2d 671 (1978), he was therefore not prosecuted on the theory of felony murder ... We decline to accept that logic. In our view, the declaration of a mistrial for manifest necessity is the equivalent of a reversal of a conviction on appeal, for double jeopardy purposes.” (emphasis supplied).
67 Md.App. at 610, 508 A.2d 1010.14 If the appellant here had been convicted of assault with intent to murder on *551Sharrell Hudson and that conviction had been reversed on appeal for some evidentiary error, he could hardly claim that the State is collaterally estopped from retrying him.
How then do we explain the jury’s conviction of the appellant as an accessory after the fact to the murder of Sherman Chenault? Does it not imply that the jury must have decided that the appellant did not aid and abet Kent Tilghman in any of the criminal activities while they were still in progress but only enlisted in the enterprise after the crimes had been completed? Of course not! What this jury obviously concluded was that although it had some doubt about whether aiding and abetting in an armed robbery necessarily meant that the aider and abettor was also guilty of several unintended (by him) shootings growing out of the armed robbery, it had no doubt whatsoever that the appellant, at the very least, drove the getaway car and qualified, therefore, as an accessory after the fact.
The jury, as a group of laymen, would have no independent knowledge that the Maryland common law at the time treated principalship and accessoryship after the fact as mutually exclusive. Common sense would dictate no such notion. To the laymen, a criminal might easily be guilty both of 1) participating in a crime and then 2) driving the getaway car. Each could well qualify as a separate crime.
The definition given, during jury instructions, on accessoryship after the fact, clearly suggested that the driver of a getaway car would be guilty of accessoryship:
“Now, the defendant is also charged with being an accessory after the fact to the murder of Sherman Joseph Chenault. In this case, an accessory after the fact is a person [who] with knowledge that a crime has been committed assists the offender by transporting him from the scene of the crime. In order to be guilty of the crime: (1) the defendant must have become associated with the felony after its commission; (2) the defendant *552must have provided assistance with knowledge that the felony had been completed; (3) the assistance must have been given to one known to be the felon; and, (4) the assistance must have been given in order to hinder the felon’s apprehension by transporting him from the scene.” (emphasis supplied).
Did that instruction also tell the jury that before it could find the appellant guilty of accessoryship, it must affirmatively find that he had not participated in the crimes themselves? It did not. There was, to be sure, a passing reference to the fact that an accessory is one who has “become associated with the felony after its completion.” One might infer from those words that the accessory may not also have been a principal, but the preclusion of that additional possibility is not mandated by the words. It does not defy logic to suggest that a defendant might become associated with another in the commission of a crime and then again become associated with that other in the coverup phase.15 No instruction was given to the effect that it was necessary that the jury find the appellant not to have been a principal in order to convict him of accessoryship. The mutual exclusivity of the two crimes was never mentioned. The appellant, moreover, never requested such an instruction. The closing arguments of the prosecutor and defense counsel did not mention the subject. When the jury was subsequently instructed about rendering its various verdicts, there was no suggestion that if they found the appellant guilty of the murder of Sherman Chenault, they should not then even consider the accessory charge.
*553The reason for this lack of attention is clear. The courtroom was focused upon a first-degree murder charge. An eighth count, charging accessoryship after the fact, was peripheral in the extreme. In argument and in instructions, it was simply not a matter of significant concern as a large package of more major charges went to the jury. All of this current brouhaha is an appellate afterthought, arising only out of the bizarre configuration of verdicts.
When a judge sits as a jury, we look, for collateral estoppel triggering purposes, at what he probably actually decided as a matter of fact, even if that decision was based upon a mistake of law. State v. Anderson, 320 Md. 17, 32, 575 A.2d 1227 (1990). A fortiori, we do not demand of jurors the legal mastery of Justice Cardozo and the logical consistency of Mr. Spock. We only want to know what they probably really did, right or wrong.
Any law to the contrary notwithstanding, a jury determination that the appellant was “at the very least, the driver of the getaway car” is a far cry from the very different determination that the appellant was “to the exclusion of all other criminality, the driver of the getaway car.” To foreclose the appellant’s assertion of a collateral estoppel claim here, it is not, of course, necessary that we conclude that the former determination was made. It is enough to raise the possibility that it might have been made.
It is the appellant’s burden to persuade us to the contrary. He has not done so. We feel as did the Supreme Court in Dowling v. United States, 493 U.S. 342, 352, 110 S.Ct. 668, 674, 107 L.Ed.2d 708, 708-720 (1990):
“There are any number of possible explanations for the jury’s acquittal verdict at Dowling’s first trial. As the record stands, there is nothing at all that persuasively indicates that the question of identity was at issue and was determined in Dowling’s favor at the prior trial ...
*554[E]ven if we were to apply the Double Jeopardy Clause this case, we would conclude that petitioner has failed satisfy his burden.” 2 2
JUDGMENT AFFIRMED; COSTS TO BE PAID BY APPELLANT.
. Bartkus v. Illinois, 359 U.S. 121, 79 S.Ct. 676, 3 L.Ed.2d 684 (1959) and progeny. See also Bailey v. State, 303 Md. 650, 660-662, 496 A.2d 665 (1985); Evans v. State, 301 Md. 45, 481 A.2d 1135 (1984); Bell v. State, 22 Md.App. 496, 509-510, 323 A.2d 677 (1974).
. North Carolina v. Pearce, 395 U.S. 711, 89 S.Ct. 2072, 23 L.Ed.2d 656 (1969) and progeny.
. Justice Brennan waged a dogged but ultimately unsuccessful campaign to have compulsory joinder, using the "same transaction” test, included as yet a sixth distinct species within the genus "double jeopardy,” but was never able to muster more than two additional votes. Ashe v. Swenson, 397 U.S. 436, 448-461, 90 S.Ct. 1189, 1196— 1203, 25 L.Ed.2d 469, 478-484 (1970) (concurring opinion by Brennan, J.); Simpson v. Florida, 403 U.S. 384, 387, 91 S.Ct. 1801, 1803, 29 L.Ed.2d 549, 553 (1971) (concurring opinion by Brennan, J.); Harris v. Washington, 404 U.S. 55, 57, 92 S.Ct. 183, 184, 30 L.Ed.2d 212, 215 (1971) (concurring opinion by Brennan, J.).
. For an articulate distinction between the broadly foreclosing effect of res judicata as a matter of law and the far narrower foreclosing effect of collateral estoppel as a matter of fact, see MPC, Inc. v. Kenny, 279 Md. 29, 32-33, 367 A.2d 486 (1977). Double jeopardy, incidentally, has frequently been referred to as "res judicata in prison grey."
. At the time of the proceedings now under review, Kent Tilghman was still at large.
. Judge Sybert ordered a stay in the circuit court proceedings pending the outcome of this appeal. Neal v. State, 272 Md. 323, 326, 322 A.2d 887 (1974); Pulley v. State, 287 Md. 406, 414-415, 412 A.2d 1244 (1980).
. See, however, State v. Hawkins, 326 Md. 270, 604 A.2d 489 (1992), which prospectively changed the common law of Maryland so that now "it is no longer an element of the crime of accessoryship after the fact that the accessory may not be a principle in either degree, in the commission of the substantive felony.” As with most prospective changes, this would apply to all cases still on direct review. The change, however, had not yet been announced when the decision was made in the circuit court to dismiss Counts 1 and 3 rather than to permit a retrial on those charges, a decision which is not on review.
. The cavalier fashion ("decision by slogan") in which this fifth species — collateral estoppel — was summarily dragooned into the double jeopardy fold by judicial fiat is to be contrasted with the careless fashion in which the fourth species — mistrial/retrial—was indiscernibly slipped into the double jeopardy fold through historic accident or analytic oversight.
*529The scholarly dissenting opinion of Justice Powell in Crist v. Bretz, 437 U.S. 28, 40-53, 98 S.Ct. 2156, 2163-70, 57 L.Ed.2d 24, 34-42 (1978), traces step by step “this rather unreflective incorporation of a common-law rule of jury practice into the guarantee against double jeopardy.” 437 U.S. at 46, 98 S.Ct. at 2166. With full agreement on this point by the majority, Justice Powell pointed out that it was clear that “in the early years of our national history the constitutional guarantee against double jeopardy was restricted to cases in which there had been a complete trial — culminating in acquittal or conviction.” 437 U.S. at 40, 98 S.Ct. at 2163. He further observed that “[t]he pleas of autrefois acquit and autrefois convicf embodied “a res judicata policy” at which "the Double Jeopardy Clause was directed.” 437 U.S. at 41, 98 S.Ct. at 2164.
By contrast, "there existed a separate rule of English practice" out of which developed our modern law of mistrial/retrial. "[Tjhis rule arose as an aspect of jury practice, rather than as an element of the guarantee against double jeopardy.” Id. Even while affording defendants the benefit of what came to be called Lord Coke’s rule, the English judges scrupulously "refused to import the rule into the realm of pleas in bar, and it was the latter which informed the framing of the Double Jeopardy Clause.” 437 U.S. at 43, 98 S.Ct. at 2165.
The Supreme Court first applied Lord Coke's rule in the 1824 case of United States v. Perez, 9 Wheat. 579, 6 L.Ed. 165 (1824). From a variety of statements made both before and after the Perez decision, it was clear that Justice Story, who authored the opinion, as well as Justice Washington, did not look upon Perez as an aspect of double jeopardy law. “As both Justices Washington and Story believed that the Double Jeopardy Clause embraced only actual acquittal and conviction, they must have viewed Perez as an independent rule barring needless discharges of the jury." 437 U.S. at 44-45, 98 S.Ct. at 2165. In the wake of Perez, however, and "[throughout the 19th century ... many state courts began to blend the rule against needless discharges of juries into the guarantee against double jeopardy contained in the Federal and State Constitutions.” 437 U.S. at 45, 98 S.Ct. at 2166. Justice Powell pointed out that the state courts "did so — with no apparent awareness of the novelty of their action — under the rubric of the Double Jeopardy Clause.” 437 U.S. at 46, 98 S.Ct. at 2166.
Wade v. Hunter, 336 U.S. 684, 69 S.Ct. 834, 93 L.Ed. 974 (1949), appears to have been, in terms of its fundamental conceptualization, nothing more than a perpetuation of the uncritical categorizing that had been going on in many of the state courts. "It was after more than a century of development in state courts that the ‘defendant’s valued right to have his trial completed by a particular tribunal’ appeared in the decisions of this Court for the first time, also without analysis, as an element of the Double Jeopardy Clause.” 437 U.S. at 47, 98 S.Ct. at 2166-67. Justice Powell summed up the historic process as one in which the Supreme Court had “accepted almost without articulated thought the doctrine that the Double Jeopardy Clause protected against needless discharge of the jury.” Id.
The majority opinion in Crist v. Bretz simply shrugged its shoulders with the resigned observation that it was too late to do anything about an error so deeply rooted:
*530"In fact, a close reading of the short opinion in that case [Perez] could support the view that the Court was not purporting to decide a constitutional question, but simply settling a problem arising in the administration of federal criminal justice. But to cast such a new light on Perez at this late date would be of academic interest only." (emphasis supplied).
437 U.S. at 34 n. 10, 98 S.Ct. at 2166 n. 10. Maryland, to the commendable contrary, has steadfastly refused to treat Lord Coke’s rule or the mistrial/retrial doctrine as part of double jeopardy law, adhering rather to the principle that the common law maxim, embedded in the constitutional protection:
"[Ijs properly interpreted by the authorized exposition established at its adoption. ‘At common law it meant nothing more than that where there had been a final verdict either of acquittal or conviction, on an adequate indictment, the defendant could not be a second time placed in jeopardy for the particular offense."
Hoffman v. State, 20 Md. 425, 434 (1863) (quoting Wharton’s Criminal Law, 147); Kearney v. State, 48 Md. 16, 27 (1877); Anderson v. State, 86 Md. 479, 482, 38 A. 937 (1897); Boone v. State, 3 Md.App. 11, 23-28, 237 A.2d 787 (1968); West v. State, 52 Md.App. 624, 627-628, 451 A.2d 1228 (1982). Kyle v. State, 6 Md.App. 159, 250 A.2d 314 (1969), was decided before Benton v. Maryland, 395 U.S. 784, 89 S.Ct. 2056, 23 L.Ed.2d 707 (1969), imposed the federal law of double jeopardy upon the states and represents, therefore, a pure view of the Maryland common law of double jeopardy uncontaminated by extraneous influences. It held flatly, at 6 Md.App. at 161, 250 A.2d 314, that the entire mistrial/retrial phenomenon was not a part of the Maryland, or the English, common law of double jeopardy:
"It is clear ... that no verdicts were rendered at the first trial and the present status of the law in Maryland, following the common law, is that no man is in jeopardy until verdict is rendered."
We pointed out that the distinct question of whether the common law procedural rule had been violated had not been raised upon the appeal. 6 Md.App. at 162 n. 2, 250 A.2d 314. Citing Kyle v. State, with approval, the Court of Appeals recognized in Cornish v. State, 272 Md. 312, 316 n. 2, 322 A.2d 880 (1974), that although the mistrial/retrial phenomenon must now be considered under the federal mandate of Benton v. Maryland, it never has been a part of the Maryland common law of double jeopardy:
“Prior to the decision in Benton v. Maryland, supra, the double jeopardy prohibition was applicable to Maryland prosecutions only as a common law principle. The view in this state was that, under the common law’s double jeopardy prohibition, jeopardy did not attach until the rendition of a verdict and that, therefore, a retrial following the declaration of a mistrial did not give rise to a double jeopardy problem." (emphasis supplied).
. One other case, Standefer v. United States, 447 U.S. 10, 100 S.Ct. 1999, 64 L.Ed.2d 689 (1980), declined to extend to criminal cases the Supreme Court’s very recent and very limited doctrine of nonmutual collateral estoppel. Blonder-Tongue Laboratories, Inc. v. University of Illinois Foundation, 402 U.S. 313, 91 S.Ct. 1434, 28 L.Ed.2d 788 (1971); Parklane Hosiery Co. v. Shore, 439 U.S. 322, 99 S.Ct. 645, 58 L.Ed.2d 552 (1979). That limited exemption from the mutuality requirement of collateral estoppel “promoted judicial economy and conserved private resources without unfairness to the litigant against whom estoppel was invoked." 447 U.S. at 21, 100 S.Ct. at 2007. Standefer pointed out, that by contrast, the “criminal case involves 'competing policy considerations’ that outweigh the economy concerns that undergird the estoppel doctrine.” 447 U.S. at 25, 100 S.Ct. at 2008. As a result, it did not permit the defendant there to utilize the prior acquittal of a principal so as to estop collaterally his subsequent trial as an aider and abettor. Nonmutual collateral estoppel, whatever its future elsewhere, will not apply to criminal cases.
In this regard, see Carbaugh v. State, 294 Md. 323, 329-330, 449 A.2d 1153 (1982) (Disposition of State’s case against one defendant will not collaterally estop relitigation of a common factual issue in a subsequent prosecution of a different defendant because of the nonmutuality of the parties in the two cases). For a variation on the theme of nonmutuality, see Bailey v. State, 303 Md. 650, 660-662, 496 A.2d 665 (1985) (Collateral estoppel will not bar prosecution against the same defendant for the same criminal episode by two different sovereigns because of the nonmutuality of the parties in the two cases).
. Both cases build upon the earlier decision of Helvering v. Mitchell, 303 U.S. 391, 58 S.Ct. 630, 82 L.Ed. 917 (1938), decided when collateral estoppel was still a law unto itself and had not yet been reclassified as an aspect of constitutional double jeopardy law by Ashe v. Swenson.
. Quite aside from the impact that differing burdens of persuasion may have on collateral estoppel issues, some other interesting questions, as yet unresolved, potentially rear their heads in cases of criminal prosecutions and civil forfeiture proceedings arising out of the same criminal episode. Since the forfeiture proceeding is an in rent proceeding against the chattel itself and not nominally against its owner, it may well be argued that collateral estoppel is inapplicable between the two proceedings because the opposing parties in the criminal action and in the civil action are not the same.
Even if the owner of the goods subject to forfeiture were deemed to be the real party in interest in the forfeiture case, thus surmounting that first hurdle, other potential problems appear. If the forfeiture case came first and the owner prevailed even where the State had a lower burden of persuasion, a consideration of persuasion levels alone would suggest that the State would be collaterally estopped from going forward with the criminal prosecution where it would carry an even heavier burden of persuasion. On the other hand, collateral estoppel is but an aspect of double jeopardy law and there would not have been, in that sequence, any initial jeopardy. Double jeopardy is a phenomenon peculiar to criminal cases and exposure to property loss in a civil case would not represent the attachment of jeopardy. Absent initial jeopardy, there cannot be, by definition, *534subsequent or double jeopardy. See In re Bennett, 301 Md. 517, 524-525, 483 A.2d 1242 (1984) (“Before there can be double jeopardy there must be jeopardy.”) But see Bowling v. State, 298 Md. 396, 470 A.2d 797 (1984). It is quite possible, of course, that there could be applicable in such a circumstance some other form of issue preclusion or collateral estoppel, distinct and apart from that which has been incorporated into the double jeopardy clause for criminal cases. In that nonconstitutional eventuality, however, it would be important to note that federal precedent, rather than local state law, would no longer be controlling.
A quite distinct problem might involve the mutuality of collateral estoppel. In the civil law, where the concept was born, the estoppel was equally available against either side of the trial table. Its mutuality is limited in a criminal case only because the State may not constitutionally be relieved of its burden of proof. If a defendant were tried and convicted in a criminal case and the civil forfeiture proceeding came afterward, there would seem to be no bar to the mutuality of the estoppel, thereby foreclosing the defendant/owner from controverting the happening of the underlying criminal episode giving rise to the forfeiture proceeding.
. In Simms, we also expressed grave reservations as to whether a sentencing hearing, at least in a noncapital case, is even an appropriate occasion for invoking collateral estoppel.
. Even Powers's core holding that "the doctrine of collateral estoppel applies after a jury, at a single trial, acquits on one count of a multicount indictment and is unable to agree upon a verdict on a related count of the same indictment involving a common issue of ultimate fact,” 285 Md. at 288, 401 A.2d 1031, a question the Supreme Court has never decided, is cast in serious doubt by Ohio v. Johnson, 467 U.S. 493, 104 S.Ct. 2536, 81 L.Ed.2d 425 (1984). Ohio v. Johnson, to be sure, dealt with the former conviction variety of traditional double jeopardy but it did turn its attention to collateral estoppel in an “arguendo" consideration.
Johnson argued that his conviction, upon his plea of guilty, of manslaughter precluded Ohio’s further prosecution of the charge of murder because the respective states of mind were inconsistent and the two crimes, therefore, were mutually exclusive. The Supreme Court accepted the mutual exclusivity, arguendo, but nonetheless rejected Johnson’s collateral estoppel argument on two separate grounds. The second of these was that where the State is not bringing *549successive prosecutions ("successive prosecutions” is a concept distinct from "successive trials”), collateral estoppel does not even apply.
"Respondent also argues that prosecution on the remaining charges is barred by the principles of collateral estoppel enunciated by this Court in Ashe v. Swenson. Even if the two were mutually exclusive crimes, the taking of a guilty plea is not the same as an adjudication on the merits after full trial, such as took place in Ashe v. Swenson. Moreover, in a case such as this, where the State has made no effort to prosecute the charges seriatim, the considerations of double jeopardy protection implicit in the application of collateral estoppel are inapplicable.” (citations omitted) (emphasis supplied).
Ohio v. Johnson, 467 U.S. at 500 n. 9, 104 S.Ct. at 2541 n. 9.
. The continuing jeopardy theory, which limits much of the impact of double jeopardy law to the barring of successive prosecutions (the launching of a new jeopardy) rather than interfering with successive trials (the continuation of original jeopardy), is the prevalent approach in the Supreme Court. As Justice (now Chief Justice) Rehnquist observed in Richardson v. United States, 468 U.S. 317, 325-326, 104 S.Ct. 3081, 3086-87, 82 L.Ed.2d 242, 251 (1984):
"[T]he protection of the Double Jeopardy Clause by its terms applies only if there has been some event, such as an acquittal, which terminates the original jeopardy. Since jeopardy attached here when the jury was sworn, petitioner’s argument necessarily assumes that the judicial declaration of a mistrial was an event which terminated jeopardy in his case and which allowed him to assert a valid claim of double jeopardy.
... [T]he failure of the jury to reach a verdict is not an event which terminates jeopardy____ [A] trial court’s declaration of a mistrial following a hung jury is not an event that terminates the original jeopardy to which petitioner was subjected. The Government, like the defendant, is entitled to resolution of the case by verdict from the jury, and jeopardy does not terminate when the jury is discharged because it is unable to agree." (citations omitted).
*551See also Ohio v. Johnson, 467 U.S. 493, 500-502, 104 S.Ct. 2536, 2541-42, 81 L.Ed.2d 425, 434-435 (1984).
. When looking not to the legal effect of a verdict but to the thought processes of a jury of laymen, common sense is the only sure guide. Listening to a judge drone on for 20 or 30 minutes with dozens of subtle legal definitions, a juror does not seize upon a passing phrase or clause that is not being highlighted, pounce upon it with the ferocity of a law professor and then flesh it out with possible additional content not necessarily conveyed by the words themselves. To suggest that jurors thus react to instructions is not mere fiction; it is fantasy.