MOYLAN, Judge.

The appellee, Jerry Lee Wilson, was indicted by the Baltimore County Grand Jury for the possession of cocaine with intent to distribute and for related narcotics and conspiracy offenses. He filed a pretrial motion to suppress physical evidence on the ground that it had been obtained in violation of his Fourth Amendment right to be secure from unreasonable searches and seizures. On November 23, 1994, Judge Thomas J. Bollinger conducted a suppression hearing and reserved his decision on the motion. On January 10, 1995, Judge Bollinger granted the appellee's motion to suppress the evidence. Under the provisions of Md. Code (1995 Repl. Vol.), § 12-302(c)(3) of the Cts. Jud. Proc. Article, the State has appealed that suppression order.

The Issue

The single issue before us is very narrow. When a police officer makes a routine traffic stop, does his automatic right to order the driver to exit the vehicle, a procedure deemed to be constitutionally reasonable by Pennsylvania v. Mimms, 434 U.S. 106, 98 S.Ct. 330, 54 L.Ed.2d 331 (1977), also extend to passengers in the stopped vehicle?

The Factual Background

The only witness to testify at the suppression hearing was Trooper David Hughes of the Maryland State Police. At approximately 7:30 P.M. on June 8, 1994, Trooper Hughes observed a white 1994 Nissan Maxima driving southbound on I-95 at what appeared to be a high rate of speed. The trooper pulled into the lane behind the Maxima and "paced" it for approximately one mile. It was going 64 miles per hour in a 55 miles-per-hour zone. He also observed that there was no regular license tag on the front or rear of the car, except that on the back there was "a paper tag kind of hanging half off, half on that said Enterprise Rent-A-Car." Officer Hughes activated his lights and siren, but the Maxima continued to drive, with Trooper Hughes behind it, for approximately one-and-one-half miles before finally stopping in Baltimore City.

Both during the pursuit and then in approaching the Maxima after it had stopped, Trooper Hughes observed that the car had three occupants. During the pursuit, the two passengers had turned and looked at him several times and had on several occasions ducked below the sight level and then reappeared.

As Trooper Hughes started to approach the Maxima on foot, he saw that the driver had spontaneously exited the vehicle. The trooper directed the driver to step back toward him and the two met at a point between their respective vehicles. Trooper Hughes advised the driver, a Mr. McNichol, why he had been stopped and asked McNichol for his license and registration card. McNichol explained that he was coming from Connecticut and going toward South Carolina. He produced a valid Connecticut driver's license. McNichol further indicated that the rental papers for the car were in the vehicle. It was at that point that Trooper Hughes instructed McNichol to return to the vehicle to retrieve the rental documents. McNichol got in the vehicle and sat in the driver's seat.

Throughout the initial encounter, Trooper Hughes had observed that McNichol was extremely nervous. He appeared at times to be trembling and answered every question with a question. Trooper Hughes had also observed that the front seat passenger, the appellee Jerry Lee Wilson, was sweating and extremely nervous.

It was after McNichol had reentered the car that Trooper Hughes ordered Wilson out of it. As Wilson complied with the trooper's direction to walk back closer to the police vehicle, what appeared to be (and, indeed, turned out to be) crack cocaine fell to the ground. Trooper Hughes drew his weapon and placed Wilson under arrest. When Trooper Hughes was asked why he had directed Wilson to exit the vehicle, he replied:

Well, due to the movement in the vehicle I thought possibly there could be a handgun in the vehicle. I had concern for my safety. At that time when Mr. McNichol went back to the car, I asked Mr. Wilson to step out, that is my whole purpose of not approaching the vehicle, by myself, with three occupants in the vehicle, I wanted each one out at a time to speak to each individual, for my safety.

The single issue before Judge Bollinger was whether Trooper Hughes violated Wilson's Fourth Amendment right against having his person seized unreasonably when he ordered Wilson to step out of the vehicle.

A False Trail

Both at appellate argument and in appellate brief, the State urged, as an alternative rationale, that Trooper Hughes had articulable or particularized suspicion under Terry v. Ohio, 392 U.S. 1, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968) and Sibron v. New York, 392 U.S. 40, 88 S.Ct. 1889, 20 L.Ed.2d 917 (1968), either to "stop" Wilson for questioning or to "frisk" Wilson for weapons. Under either set of circumstances, the minimal seizure of Wilson's person occasioned by ordering him from the car would have been reasonable. We reject the State's alternative rationale, however, for several reasons.

In the first place, the State never urged such a basis for the exit-order at the suppression hearing. Judge Bollinger was not called upon to rule whether there was articulable suspicion either for a Terry "stop" or for a Terry "frisk" and, indeed, he made no such rulings.

With respect to this alternative rationale, the State is even more bereft. Articulable suspicion, for either a stop or a frisk, requires not simply the external circumstances that would justify such particularized suspicion. It requires, in addition, that the officer purporting to act on the basis of such suspicion actually articulate such a purpose and such a basis for action. Gibbs v. State, 18 Md. App. 230, 239-42, 306 A.2d 587 (1973).

What must be articulated to justify a Terry "stop" is particularized suspicion that a crime has occurred, is then occurring, or is about to occur. A "stop," unlike a "frisk," is crime-related, not weapon-related. The societal purpose served by a Terry "stop" is the prevention or detection of crime. The justification for a "stop," therefore, must be framed and phrased in terms of suspected crime. It is, moreover, the officer who must do the articulating, not the Attorney General by way of appellate afterthought. Trooper Hughes articulated nothing with respect to any crime that he suspected Wilson of being involved in. The absence of an articulated basis for a Terry "stop" is as absolute here as it was in Gibbs v. State:

Officer Stewart, in the case at bar, articulated absolutely nothing as to what crime or type of crime he reasonably suspected the appellant of having engaged in, of then engaging in, or of being about to engage in. (Emphasis supplied.)

18 Md. App. at 241, 306 A.2d 587.

With respect to a possible justification for the exit-order based on the notion that Trooper Hughes was somehow undertaking a Terry "frisk," the overarching fact is that Trooper Hughes never remotely articulated having entertained any such purpose. He did, to be sure, express some fear that "possibly there could be a handgun in the vehicle." The ostensible purpose for ordering Wilson out of the car, however, was to take Wilson out of proximity to such a possible weapon rather than to frisk him for a weapon. Whatever Trooper Hughes was doing when he ordered Wilson out of the car, it was not in furtherance of any intent to conduct a frisk. In the absence of such a purpose, whether there might, in the abstract, have been a constitutional basis for a frisk is immaterial.

If these were not impediments enough, the State urges us to exercise our own independent constitutional appraisal on a de novo basis. Such appellate latitude is not available to us. Although we may exercise de novo review with respect to mixed questions of law and fact, we are enjoined to extend the more deferential "clearly erroneous" standard of review to the findings of the trial judge on purely factual questions. Aiken v. State, 101 Md. App. 557, 563, 647 A.2d 1229 (1994), cert. denied, 337 Md. 89, 651 A.2d 854 (1995); State v. Jones, 103 Md. App. 548, 606-11, 653 A.2d 1040 (1995). Judge Bollinger found as a first-level fact that Trooper Hughes possessed no fear that Wilson was armed. Judge Bollinger's conclusion was not that the external circumstances did not add up to articulable suspicion. That, indeed, would have been a finding on a mixed question of law and fact and would be subject to de novo review. Judge Bollinger's finding, by way of contrast, was that Trooper Hughes did not even possess such a suspicion. That is a finding of pure fact that can be overturned only if clearly erroneous.

In part because Trooper Hughes did not order all three individuals out of the Maxima initially, and in part because McNichol was freely permitted to reenter the vehicle, Judge Bollinger concluded that Trooper Hughes's action was not a preventive or preemptive measure intended to neutralize the risk of harm from offensive weapons. His conclusion was not that there was no basis for a reasonable suspicion that Wilson was armed and dangerous, but rather that Trooper Hughes entertained no such suspicion, reasonable or unreasonable:

In this case the officer's experience is 13 months as a trooper and this Court finds that when the officer allowed the driver of the vehicle to return to the car to obtain the rental documents he could not have had a reasonable suspicion that the person was armed and dangerous; and, therefore, any future intrusion into the right of the occupants of the car are violative of one's Fourth Amendment proscription of unreasonable searches and seizures.

As first-level fact finding, inevitably influenced by Judge Bollinger's observation of Trooper Hughes's demeanor and manner of testifying and by Judge Bollinger's assessment of Trooper Hughes's credibility, this is the type of thing that the trial judge is far more competent to weigh and to decide than we would be on the basis of a cold record. This, we hold, is a finding that was not clearly erroneous.

It is as important, perhaps, to note what we do not hold as it is to note what we do. Had Judge Bollinger found that Trooper Hughes had reason to fear that Wilson might be in possession of a weapon and had ordered Wilson out of the vehicle in order to frisk him for a weapon, we are not holding that such a conclusion and such an action on the trooper's part would have been unreasonable. That, however, is not the decision before us for review. Judge Bollinger found, to the contrary, that the trooper had no such fear and we simply hold that Judge Bollinger was not clearly erroneous in so finding.

Pennsylvania v. Mimms

If the State's argument that it was reasonable for Trooper Hughes to order Wilson out of the car, therefore, is somehow to fly, it must lift off from the doctrinal launching pad of Pennsylvania v. Mimms and not from that of Terry v. Ohio.

It is treacherous to attempt to extract too much meaning from Pennsylvania v. Mimms. It is a six-page, unsigned opinion explaining a summary disposition reached without benefit of oral argument. One must be careful not to read more meaning into such an opinion than the authors ever intended to put there. The simple holding of Pennsylvania v. Mimms is that whenever a police officer stops a vehicle for a traffic violation, even of the most minor variety, it is reasonable for the officer to order the driver to alight from the vehicle before the officer proceeds to inspect the driver's license and registration card, to question the driver, to observe the driver with respect to sobriety, to issue a traffic citation, or to take other appropriate action.

In the Mimms case itself, two Philadelphia police officers had observed Harry Mimms driving an automobile with an expired license plate. They stopped the car for the purpose of issuing a traffic citation. One of the officers approached the vehicle and asked Mimms to step out of the car and to produce his owner's card and operator's license. When Mimms did so, one of the officers noticed a large bulge under his sports jacket. Fearing that the bulge might be a weapon, the officer frisked Mimms and discovered in his waistband a .38 caliber revolver loaded with five rounds of ammunition. Mimms was arrested and convicted for carrying a concealed weapon and for carrying a firearm without a license.

In its opinion, the Supreme Court made it clear that its focus was very narrow. There was no question about the propriety of the initial stop of the vehicle. 434 U.S. at 109, 98 S.Ct. at 332. Nor was there any question about the propriety of the ultimate frisk of Mimms for weapons, because the observation of the bulge furnished articulable or particularized suspicion that a weapon might be present, to wit, an independent basis for that further intrusion. The only issue before the Supreme Court was the propriety of the intermediate act of ordering Mimms to step out of the vehicle after it had been stopped:

[W]e need presently deal only with the narrow question of whether the order to get out of the car, issued after the driver was lawfully detained, was reasonable and thus permissible under the Fourth Amendment. This inquiry must therefore focus not on the intrusion resulting from the request to stop the vehicle or from the later "pat-down," but on the incremental intrusion resulting from the request to get out of the car once the vehicle was lawfully stopped.

434 U.S. at 109, 98 S.Ct. at 332.

After balancing the societal interest of protecting the officer from harm against the intrusion into the Fourth Amendment interests of the driver, the Court concluded that it is reasonable for an officer to order a driver to alight from the vehicle whenever the vehicle is lawfully stopped for a traffic violation. The holding was very clear that that prerogative is automatic under such circumstances and that no individualized or particularized suspicion that the driver might be armed is involved. The police practice under review in Mimms was always to order the driver to alight from the vehicle:

The State freely concedes the officer had no reason to suspect foul play from the particular driver at the time of the stop, there having been nothing unusual or suspicious about his behavior. It was apparently his practice to order all drivers out of their vehicles as a matter of course whenever they had been stopped for a traffic violation. (Emphasis supplied.)

434 U.S. at 109-10, 98 S.Ct. at 332.

The police prerogative vis-a-vis a stopped driver is plenary and automatic. Once the lawfulness of the initial stop of the vehicle is established, nothing more need be shown on a case-by-case basis to justify the incremental intrusion of ordering the driver out of the vehicle. In the case now before us, Trooper Hughes had made a lawful stop of the Nissan Maxima on I-95. He was, thereby, automatically entitled to order the driver out of the car. It turned out, coincidentally, to be unnecessary because the driver had already alighted spontaneously.

The appellant Wilson, however, was not the driver but only one of the passengers. Did or did not the automatic police prerogative extend to him?

Mimms Did Not Deal With Passengers

The Mimms opinion was completely silent on the police prerogative, if any, vis-a-vis a passenger. There had coincidentally been a second occupant of Harry Mimms's vehicle who, it turned out, was carrying a .32 caliber revolver. Once that narrative fact was mentioned, however, the armed passenger dropped totally from sight and the opinion does not further allude to him even obliquely. The Supreme Court, indeed, went out of its way to disclaim any consideration of the rights or vulnerabilities of passengers:

The State does not, and need not, go so far as to suggest that an officer may frisk the occupants of any car stopped for a traffic violation. Rather, it only argues that it is permissible to order the driver out of the car.

434 U.S. at 110 n. 5, 98 S.Ct. at 332 n. 5.

The State argues, however, that the automatic police prerogative vis-a-vis a driver should be equally automatic vis-a-vis any other occupant of a stopped vehicle. The State actually makes two arguments in support of this proposition, one based on legal authority and the other based on the inherent logic of drawing the parallel.

A. The Argument Based on Legal Authority:

The State does not simply argue that the principle of Pennsylvania v. Mimms should be extended to passengers; it argues that the Mimms principle already has been extended to passengers. It cites as the extending authorities the Supreme Court opinions in Rakas v. Illinois, 439 U.S. 128, 99 S.Ct. 421, 58 L.Ed.2d 387 (1978) and Michigan v. Long, 463 U.S. 1032, 103 S.Ct. 3469, 77 L.Ed.2d 1201 (1983), and our own opinion in Derricott v. State, 84 Md. App. 192, 578 A.2d 791 (1990), rev'd on other grounds, 327 Md. 582, 611 A.2d 592 (1992).

What significance, if any, has Michigan v. Long, Rakas v. Illinois, or our own Derricott v. State in resolving the issue now before us. If nothing else, they afford an occasion for serious philosophic reflection on the present state of stare decisis. In looking to the case law as a source of legal authority, there seems to be an almost epidemic trend among lawyers generally — civil and criminal, on the prosecution side and the defense side of the trial table alike, and even among some judges — to forget the most elemental ABC's of first-semester, first-year Legal Method. Everything said by a judge or by a court is not of equal weight. Were the three cases that we are about to examine offering us three thoroughly considered holdings, their authority would stand on undeniable bedrock. Were they three well considered, well researched, and well analyzed instances of even deliberate and conscious dicta, they would still be standing on firm, persuasive ground. If, on further examination, however, they turn out to be no more than three careless, casual, and passing instances of the most obiter of dicta, they should represent no more to us than the glib blandishments of a snake oil salesman, no matter who wrote them.

In Michigan v. Long, the Supreme Court was not remotely dealing with a Pennsylvania v. Mimms situation. It was a case, rather, where there was undeniable articulable suspicion that an individual might be armed and dangerous. The only issue before the Court was whether a Terry "frisk" could extend geographically from the suspect himself into the passenger compartment of the automobile from which he had just alighted. Pennsylvania v. Mimms was cited only to illustrate that an automobile can be the source of a dangerous weapon. The reference to Mimms consisted of the following:

In Pennsylvania v. Mimms, we held that police may order persons out of an automobile during a stop for a traffic violation, and may frisk those persons for weapons if there is a reasonable belief that they are armed and dangerous. Our decision rested in part on the "inordinate risk confronting an officer as he approaches a person seated in an automobile." (Citation omitted.) (Emphasis supplied.)

463 U.S. at 1047-48, 103 S.Ct. at 3480.

The Mimms principle itself was not being considered, let alone some more subtle ultra- Mimms distinction between driver and passenger. The only significance that could conceivably be derived from this quotation for present purposes is that twice the word "persons" appeared in the plural rather than in the singular. Clearly, there was involved no consideration of the police prerogative vis-a-vis a passenger as opposed to the prerogative vis-a-vis a driver, for in Michigan v. Long there was only one suspect involved — the driver himself. The casual use of the plural rather than the singular in this content does not in any way support the doctrinal weight the State seeks to place upon it. Every word that is uttered in a legal opinion is not legal authority.

The other two instances illustrate even more pointedly why the fundamental distinction between holdings and dicta should be rigorously maintained. In the Anglo-American common law tradition of judge-made law, it is, when we are meticulously precise about the matter, the decision of an appellate court that has, within the appropriate jurisdictional framework, binding precedential authority, and not necessarily the opinion announcing the decision. Generally speaking, however, when it comes to the actual holding of a case, the decision of a court and the opinion of the court announcing the decision are coterminous.

The precedential weight of a holding is predicated in large measure on its status as the deliberate and considered judgment of an entire collegiate court, including the opinion writer, on the issue before it that must be decided. Each member of an appellate court peers in with painstaking scrutiny not only on the decision itself but on the framing of the holding that announces the decision. The articulation of the holding passes through a stern editorial process that insists that every "i" be dotted, every "t" be crossed, every word be carefully chosen, and every far-flung repercussion be sagely anticipated. A holding, therefore, has earned the authoritative weight we give it.

The holding, however, consists of no more than a few sentences or, at most, a few paragraphs, generally located near the end of what may be a twenty or thirty-page opinion. When it comes to the composition of the opinion leading up to the holding, the collegiate editorial reins are, although not totally relaxed, far looser. There is a wide stylistic range within which the opinion writer may freely express a particular legal philosophy; a special analytic approach to problem solving; possibly idiosyncratic reactions to certain arguments; or, above all, an individualistic writing personality. The active collegiate participation in the formulation of a holding retreats to gentle stylistic suggestion when it comes to the writer's modality of expression. It is not uncommon for a panel member to subscribe to an opinion, notwithstanding an occasional wince of pain or smile of indulgent tolerance along the way. The point is that everything said in an opinion — the dicta — is not entitled to the same weight as is the holding of the Court.

Well-considered dicta, of course, is sometimes very good and, therefore, of significant persuasive weight. That is a far cry, however, from giving persuasive weight to every hurried word that may appear in the course of an opinion. The opinion-writing process is frequently a high volume production line and it should be obvious that every syllable is not chiseled in marble. Inadvertent mistakes inevitably creep in. This is unremarkable, unless we seize upon the mistakes, extend them a precedential weight to which they are not entitled, and then follow them as false gods. The citations to Rakas v. Illinois and Derricott v. State at least flirt with just such precedential heresy.

The citation to Rakas is to a footnote in a concurring opinion. The issue for decision in Rakas had nothing to do with the police prerogative to order a driver or a passenger to alight from an automobile. It dealt exclusively with the standing of a "mere passenger" to raise a Fourth Amendment challenge to the search of someone else's automobile. The footnote in the concurring opinion was responding to the dissent in the case and was making the point that an automobile generally, with no distinction even considered between driver and passenger, enjoys a lesser expectation of privacy than do other places, such as homes. Pennsylvania v. Mimms was used simply to illustrate that point:

Last Term, this Court determined in Pennsylvania v. Mimms that passengers in automobiles have no Fourth Amendment right not to be ordered from their vehicle, once a proper stop is made. (Citation omitted.)

439 U.S. at 155 n. 4, 99 S.Ct. at 436 n. 4 (concurring opinion by Powell, J.). The deceptively broad reference to "passengers" was simply wrong. It should, more carefully, have been a more limited reference to "drivers." That distinction, however, was not remotely involved in the discussion and should not be given a significance that was never intended.

The same sort of inadvertent mistake infected our opinion in Derricott v. State. As we worked our way toward the critical issue actually before us in that case, we described, almost as passing landscape, the unquestioned legitimacy of an initial traffic stop and then the unquestioned legitimacy of ordering the stopped driver to alight from his vehicle. Clearly, we were not dealing with any distinction between a driver and a passenger, because there was no passenger in the Derricott vehicle. No one other than the driver was involved. Nonetheless, our characterization of Pennsylvania v. Mimms did read:

Pennsylvania v. Mimms established unequivocally that when the police have legitimately stopped an automobile, for a traffic offense or for any other reason, they are automatically entitled to order the driver and/or any of the passengers to alight from the vehicle. (Citation omitted.) (Emphasis supplied.)

84 Md. App. at 197-98, 578 A.2d 791. The overly broad inclusion of the phrase "and/or any of the passengers" was simply wrong. Pennsylvania v. Mimms did not stand for so broad a proposition.

The writer of the concurring opinion in Rakas and the writer of this Court's opinion in Derricott did exactly the same thing, as opinion writers sometimes do. On an immaterial issue and casually in passing, they characterized Pennsylvania v. Mimms as they remembered it, but they remembered it wrongly. They did not put down a pen in mid-sentence and run to the United States Reports for the obvious reason that the passing reference was not in any way the focus of consideration. To be sure, a judge said the words. The words, however, in such a context, should be given the weight they would be given if the judge had said them in a law review article or in a newspaper column or in a talk to the Kiwanis Club. In this same regard, see also Harris v. State, 81 Md. App. 247, 300, 567 A.2d 476 (1989), rev'd on other grounds, Harris v. State, 324 Md. 490, 597 A.2d 956 (1991).

In summation, stare decisis is ill served if readers hang slavishly on every casual or hurried word as if it had bubbled from the earth at Delphi. Obiter dicta, if noticed at all, should be taken with a large grain of salt.

B. The Argument Based on Logic:

In urging that the Mimms-based police prerogative vis-a-vis a stopped driver should be extended to cover passengers in a stopped vehicle as well, the State is on much firmer ground as it makes its argument based on the inherent logic of the parallel. In deciding Mimms as it did, the Supreme Court engaged in a balancing exercise. It measured the governmental or societal interest that would be furthered by the prerogative against the privacy interest that would be compromised by it. On balance, it deemed the prerogative, vis-a-vis a driver, to be reasonable.

The same general factors would go into the weighing process of whether it would be reasonable automatically to subject a passenger to the same police procedure. We accept the State's invitation to engage in such a weighing exercise. We point out, as we do so, that we are assessing the reasonableness of subjecting a passenger to the automatic police prerogative of ordering him out of an automobile, just as automatically as in the case of the driver, and not to such a prerogative when conditioned on some sort of particularized or individualized suspicion.

On the societal-interest pan of the balance scale, the Supreme Court placed two factors. The heavy factor was the interest in protecting the officer from harm at the hands of the driver. A secondary factor was the interest in protecting the officer from the risk of harm from oncoming traffic. In referring to that secondary factor, the Court observed:

The hazard of accidental injury from passing traffic to an officer standing on the driver's side of the vehicle may also be appreciable in some situations. Rather than conversing while standing exposed to moving traffic, the officer prudently may prefer to ask the driver of the vehicle to step out of the car and off onto the shoulder of the road where the inquiry may be pursued with greater safety to both.

Mimms, 434 U.S. at 111, 98 S.Ct. at 333.

Assuming that stopped vehicles generally pull to the right-hand curb of the street or the right-hand shoulder of the road and that it is the driver's side that is thereby exposed to traffic, this factor obviously does not come into play when considering the officer's position vis-a-vis a passenger. In a routine traffic stop, there would, indeed, be few occasions when an officer would even have any interest in standing outside the passenger's window and talking to a passenger. Since this societal interest, however, was clearly little more than a makeweight, its absence from the balance in the case of a passenger detracts little from the governmental interest being weighed.

The heavy and predominant societal interest that persuaded the Supreme Court to decide Pennsylvania v. Mimms as it did was the concern for minimizing the risk of harm to an officer from an armed and dangerous driver:

Establishing a face-to-face confrontation diminishes the possibility, otherwise substantial, that the driver can make unobserved movements; this, in turn, reduces the likelihood that the officer will be the victim of an assault.

We think it too plain for argument that the State's proffered justification — the safety of the officer — is both legitimate and weighty. "Certainly it would be unreasonable to require that police officers take unnecessary risks in the performance of their duties." And we have specifically recognized the inordinate risk confronting an officer as he approaches a person seated in an automobile. "According to one study, approximately 30% of police shootings occurred when a police officer approached a suspect seated in an automobile." We are aware that not all these assaults occur when issuing traffic summons, but we have before expressly declined to accept the argument that traffic violations necessarily involve less danger to officers than other types of confrontations. Indeed, it appears "that a significant percentage of murders of police officers occurs when the officers are making traffic stops." (Footnote and citations omitted.) (Emphasis supplied.)

434 U.S. at 110, 98 S.Ct. at 333.

Assuming the worst-case scenario, as the Supreme Court did with a driver and as we now do with a passenger, it is self-evident that an armed and dangerous passenger poses just as great a threat to an officer as does an armed and dangerous driver. In looking, therefore, exclusively at the "benefit" side of this cost-benefit analysis, we conclude that the societal benefit or societal interest is just as great when considering protecting an officer from a passenger as it is when protecting an officer from a driver.

A balancing process, however, is an exercise in relativity. A cost-benefit analysis requires that we compare the social "costs" to driver and passenger, respectively, just as we have compared the respective benefits of protecting the officer from each of them. It is here that the parallel urged by the State falters. It is here that a significantly heavier weight on the "cost" side of the scale in the case of a mere passenger produces, in our judgment, a different result.

In Mimms, the Supreme Court made it very clear that the social cost in terms of interfering with a citizen's liberty interest was a possibly counter-balancing factor that had to be considered:

The touchstone of our analysis under the Fourth Amendment is always "the reasonableness in all the circumstances of the particular governmental invasion of a citizen's personal security." Reasonableness, of course, depends "on a balance between the public interest and the individual's right to personal security free from arbitrary interference by law officers." (Citation omitted.) (Emphasis supplied.)

434 U.S. at 108-09, 98 S.Ct. at 332.

Apparently critical to the Supreme Court's decision that a police officer's prerogative to order a stopped driver out of a car could be automatic rather than conditioned on some showing of particularized suspicion was its conclusion that the social "cost" in terms of interfering with the driver's liberty interest was minimal, if not nugatory. Not only is the automobile itself stopped for a traffic infraction, but the driver is subject to further detention on an in personam basis. The driver is not permitted to walk away from the scene and disappear into the sunset. The driver is required to furnish the officer with evidence of a valid operator's license and valid registration for the vehicle, subject to penalties for failure to do so. The driver is further detained while the officer checks out the identifying data over a police radio. The driver is subject to receiving a traffic ticket or citation, if not, indeed, to custodial arrest for the infraction. All this is well settled law before the Pennsylvania v. Mimms considerations even come into play. With the detention of the driver already an undisputed fait accompli, the small incremental intrusion of having him alight from the vehicle while being detained was considered by the Supreme Court to be absolutely de minimis:

Against this important interest we are asked to weigh the intrusion into the driver's personal liberty occasioned not by the initial stop of the vehicle, which was admittedly justified, but by the order to get out of the car. We think this additional intrusion can only be described as de minimis. The driver is being asked to expose to view very little more of his person than is already exposed. The police have already lawfully decided that the driver shall be briefly detained; the only question is whether he shall spend that period sitting in the driver's seat of his car or standing alongside it. Not only is the insistence of the police on the latter choice not a "serious intrusion upon the sanctity of the person," but it hardly rises to the level of a "`petty indignity.'" What is at most a mere inconvenience cannot prevail when balanced against legitimate concerns for the officer's safety. (Citation and footnote omitted.) (Emphasis supplied.)

434 U.S. at 111, 98 S.Ct. at 333.

A police interference with the presumptively unfettered liberty of a mere passenger, by contrast, cannot so easily be dismissed as "de minimis" or as something "hardly ris[ing] to the level of a petty indignity." The passenger has not committed any wrongdoing, even at the level of a traffic infraction. The passenger may not be issued a traffic ticket or citation, let alone subjected to custodial arrest. The passenger is not required to furnish identification or any other documentation. Once the automobile has slowed down sufficiently for the passenger safely to alight, moreover, the passenger is subject to no mandatory detention whatsoever. The passenger is presumptively free to abandon the driver to the clutches of the law and to hail a cab.

Under the circumstances, a police intrusion on the Fourth Amendment interest of the passenger is not so minimal or merely incremental as to be of no more than slight significance in the weighing operation. To order a passenger to stay in the car or to get out of the car is the imposition of detention per se, and not a mere shift of the location of already established detention.

The interest in officer safety, to be sure, continues to loom large, and this is admittedly a close question. All warrantless activity, however, is presumptively invalid, Katz v. United States, 389 U.S. 347, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967); Coolidge v. New Hampshire, 403 U.S. 443, 91 S.Ct. 2022, 29 L.Ed.2d 564 (1971), and the burden is on the State to persuade us that it is reasonable. The State has not persuaded us that the automatic police prerogative, under Pennsylvania v. Mimms, to order a driver out of a vehicle should be extended to a passenger — on an automatic basis. That is the only issue before us in this case. Because Trooper Hughes's right to order Wilson out of the car was not automatic, the ruling of Judge Bollinger that the evidence be suppressed is affirmed.

Of a like mind on this issue is Professor Wayne LaFave, Search Seizure: A Treatise on the Fourth Amendment, § 9.4(a) at 514-15 (2d ed. 1987):

The state court in Mimms relied upon the earlier decision in Commonwealth v. Pollard [ 450 Pa. 138, 299 A.2d 233 (1973)], holding that when a vehicle is stopped for a traffic violation the police may not, by virtue of the stop alone, order the passengers in the vehicle to get out of the car. Although the issue presented by a Pollard-type fact situation is by no means an easy one, it would seem that Pollard was correctly decided and ought to be distinguished from Mimms. For one thing, it may fairly be said that "an operator's expectation of privacy differs from that of an occupant of a vehicle detained for a traffic violation," as the driver is detained for some violation by him or the car he is driving, while the detention of the passengers is no more than an inevitable incident of the stopping of the car. But more important, the potential danger to police engaged in traffic enforcement could be adequately met if the police allowed passengers to remain in the stopped vehicle and instead had the driver accompany them to the police vehicle while the citation is prepared. (Footnotes omitted.)

Our holding in that case was clearly limited to passengers occupying a vehicle. ("Further, as was previously noted, appellant was not the driver of the automobile.") The majority in the case at bar ignores this distinction, and thus completely overlooks the question left open in Pollard of whether an operator's expectation of privacy differs from that of an occupant of a vehicle detained for a traffic violation. (Emphasis in original.) (Citation omitted.)

In State v. Becker, 458 N.W.2d 604, 607 (Iowa 1990), the Supreme Court of Iowa very articulately contrasted the situation of a driver, following a traffic stop, with the situation of a passenger:

Nothing in either Eis [ State v. Eis, 348 N.W.2d 224 (Iowa 1984)] or Mimms suggests that the privacy rights of the driver and passenger are the same at that stage. The pronouncements in Mimms concerning enhancement of the officer's safety were within the context of action which might be taken against a driver known to the officer to have violated the traffic laws. A person in that position is, in many states, including Iowa, technically subject to full custodial arrest. To the extent the officer elects to temporarily pursue a lesser intrusion, he has the right to condition that election on certain aspects of detention and search which are conducive to the officer's safety.

The situation of the passenger, on the other hand, is entirely different. The fact that the driver was speeding authorizes the officer to stop the vehicle in which the passenger is riding. The resulting intrusion on the passenger which flows from the initial stop is an unavoidable consequence of action justifiably taken against the driver. Further intrusion on the passenger is not justified, however, unless some articulable suspicion exists concerning a violation of law by that person. (Citations omitted.)

See also Johnson v. State, 601 S.W.2d 326, 327-28 (Tenn.Crim. App. 1980); State v. Williams, 366 So.2d 1369 (La. 1978), overruled by State v. Landry, 588 So.2d 345 (La. 1991).

Indeed, in New York v. Class, 475 U.S. 106, 117-18, 106 S.Ct. 960, 967-68, 89 L.Ed.2d 81, 92-93 (1986), the Supreme Court strongly suggested that its vesting a police officer with the automatic prerogative of ordering a driver out of a stopped car rested on three factors, one of which was the focusing of suspicion of wrongdoing specifically on the driver:

In Pennsylvania v. Mimms, the officers had personally observed the seized individual in the commission of a traffic offense before requesting that he exit his vehicle. . . . All three of the factors involved in Mimms and [ Michigan v.] Summers [ 452 U.S. 692, 101 S.Ct. 2587, 69 L.Ed.2d 340 (1981)] are present in this case: the safety of the officers was served by the governmental intrusion; the intrusion was minimal; and the search stemmed from some probable cause focusing suspicion on the individual affected by the search. (Citations omitted.) (Emphasis supplied.)

By contrast, a traffic stop per se does not, without more, focus suspicion on a passenger as either a wrongdoer or as a possible source of danger to the officer.

A number of state courts, to be sure, have gone in the opposite direction on this issue and have extended the automatic police prerogative to the passenger as well as the driver. Most have gone through the formal exercise of balancing the governmental interest in police safety against the degree of intrusion with the right of the passenger, although the analysis, generally speaking, has been very skimpy. State v. Landry, 588 So.2d 345, 347 (La. 1991); State v. Gilberts, 497 N.W.2d 93, 95-96 (N.D. 1993); State v. Ferrise, 269 N.W.2d 888, 890-91 (Minn. 1978); Warr v. State, 580 N.E.2d 265, 267 (Ind. App. 1991); People v. Salvator, 236 Ill. App.3d 824, 177 Ill.Dec. 58, 68, 602 N.E.2d 953, 963 (1992); People v. Martinez, 187 Mich. App. 160, 466 N.W.2d 380, 383-84 (1991), vacated on other grounds, People v. Martinez, 439 Mich. 986, 483 N.W.2d 868 (Mich. 1992). A Florida appellate court reached the same conclusion without the benefit of any apparent analysis. Doctor v. State, 573 So.2d 157, 159 (Fla.App. 1991).

Several decisions from the New York Court of Appeals leave New York's position somewhat in doubt. People v. Robinson, 74 N.Y.2d 773, 545 N.Y.S.2d 90, 91, 543 N.E.2d 733, 734 (1989), stated, without elaboration, that the Pennsylvania v. Mimms prerogative extended to passengers as well as to drivers. Its primary reliance, however, was on People v. McLaurin, 70 N.Y.2d 779, 521 N.Y.S.2d 218, 219-20 515 N.E.2d 904, 905-06 (1987), a case that explicitly refrained from holding that the prerogative was automatic and that based the propriety of the police order in that case on specific facts articulated by the police officer which warranted heightened apprehension on his part:

We need not now resolve whether the Mimms rationale would always justify an officer in ordering any or all of the passengers out of a lawfully stopped car, for this record reveals that the police conduct at issue here was predicated not only on the traffic violation and the perceived threat of danger from the driver, but also on the combination of additional factors that warranted apprehension as to the defendant passenger as well. (Emphasis supplied.)

Virginia also appeared initially to be in the automatic extension of the police prerogative camp, but has since backed away from that position. In Bethea v. Commonwealth, 14 Va. App. 474, 419 S.E.2d 249, 250-51 (1992), the Court of Appeals of Virginia appeared to extend the prerogative to passengers automatically. On appeal, however, the Supreme Court of Virginia, in Bethea v. Commonwealth, 245 Va. 416, 429 S.E.2d 211, 213 (1993), expressly refrained from any automatic extension and rested its affirmance of the officer's actions on his ability to "point to specific and articulable facts" that justified his order for the passenger to alight from the vehicle:

Nevertheless, we need not determine whether the de minimis rationale utilized in Mimms is applicable to a passenger in a vehicle when the initial vehicle stop is predicated solely on matters pertaining to the driver. The facts of this case only require the application of the more general principle that Fourth Amendment interests are not violated when a police officer can "point to specific and articulable facts which, taken together with rational inferences from those facts, reasonably warrant that intrusion." (Emphasis supplied.)

What We Are Not Holding

By way of conscious, considered, and deliberate dicta, we hasten to point out what we are not holding. We are not holding that a police officer, as a matter of personal safety, may not order a suspect passenger out of a stopped automobile as readily as he may frisk a suspect on the street or in an automobile, and perhaps even more readily. We are simply holding that the prerogative is not automatic, but requires, for justification, some individualized or particularized suspicion — just as in the case of a frisk for weapons. Where officer safety is concerned, the reasonableness threshold is low, but there is a threshold.

Of all the opinions we have found dealing with the distinction between the vulnerability of a driver to an order to alight from a car and the vulnerability of a passenger to the same order, far and away the most thoughtful and sensitive analysis is that of the Supreme Court of New Jersey, in State v. Smith, 134 N.J. 599, 637 A.2d 158 (1994). The New Jersey Supreme Court recognized, as we have done, that the danger posed to an officer is just as great at the hands of a passenger as at the hands of a driver. It is on the "cost" side of the balance scale that a difference is found. With respect to the greater intrusion that occurs in the case of a passenger, the Smith opinion reasoned:

We turn now to the second prong of the Mimms analysis and weigh the intrusion into the passenger's liberty occasioned by the trooper's order to a passenger to get out of the car as a routine safety precaution. Ordering a passenger to leave the vehicle is distinguishable from ordering the driver to get out of the vehicle because the passenger has not engaged in the culpable conduct that resulted in the vehicle's stop. Although the State's interest in safety remains the same whether the driver or the passenger is involved, requiring a passenger to alight from a car in the course of a routine traffic stop represents a greater intrusion on a passenger's liberty than the same requirement does on a driver's liberty. With respect to the passenger, the only justification for the intrusion on the passenger's privacy is the untimely association with the driver on the day the driver is observed committing a traffic violation. Because the passenger has not engaged in culpable conduct, the passenger has a legitimate expectation that no further inconvenience will be occasioned by any intrusions beyond the delay caused by the lawful stop. The intrusion on the passenger's privacy, therefore, is greater than it is on the driver's privacy. (Emphasis supplied.)

637 A.2d at 165-66.

The Smith Court recognized that although the intrusion on a passenger was not a major intrusion, it was enough to require some specific justification under the Fourth Amendment:

[T]he primary concern of the Mimms rule — officer protection — remains the same whether a passenger or a driver is involved.

. . . Applying the balancing test to passengers, however, we conclude that the scale tips against a per se rule that a passenger may always be ordered out of a vehicle lawfully stopped for a routine traffic violation. Courts have long held that some quantum of individualized suspicion is a prerequisite to a constitutional search and seizure. Although we do not think that a passenger being routinely asked to step out of a lawfully detained vehicle suffers a major intrusion, the request nevertheless amounts to an intrusion. Therefore, we do not think reasonable the proposition that auto passengers should be routinely ordered to get out of their cars after ordinary traffic stops.

Hence, we conclude that the Mimms per se rule should not be applied automatically to passengers. (Citations omitted.)

637 A.2d at 166.

In turning its attention to the relatively low threshold of justification required to order a passenger out of a car, the Smith opinion reasoned that that justification might well be less than, and different from, the articulable suspicion that a crime was occurring, so as to justify a Terry "stop," or the articulable suspicion that the passenger was armed and dangerous, so as to justify a Terry "frisk":

We do recognize, however, that instances will surface in which police officers, with less than a reasonable suspicion that a passenger is engaged in criminal activity or is armed and dangerous, may reasonably order a passenger to step out of the car.

637 A.2d at 166.

What must be articulated to justify ordering a passenger out of a car are specific and articulable facts that would support a "heightened caution" that the traffic stop was fraught with danger:

Although the per se rule under Mimms permits an officer to order the driver out of a vehicle incident to a lawful stop for a traffic violation, we decline to extend that per se rule to passengers. Instead, we determine that an officer must be able to point to specific and articulable facts that would warrant heightened caution to justify ordering the occupants to step out of a vehicle detained for a traffic violation. (Emphasis supplied.)

637 A.2d at 167.

The Smith Court made it clear, moreover, that a reasonable sensing of the need for "heightened caution," although constituting particularized suspicion, does not necessarily have to rise to the level of articulable suspicion necessary for a Terry "frisk":

Although the requirements for ordering a passenger from a vehicle are more stringent than those for ordering a driver out under the Mimms per se rule, the standard that justifies an order to a passenger to step out of a vehicle does not rise to the Terry standard that must be met for a protective pat-down. We adopt this lesser standard because of the need to protect police officers and because of the minimal intrusion the requirement to exit the car imposes on the passenger.

To support an order to a passenger to alight from a vehicle stopped for a traffic violation, therefore, the officer need not point to specific facts that the occupants are "armed and dangerous." Rather, the officer need point only to some fact or facts in the totality of the circumstances that would create in a police officer a heightened awareness of danger that would warrant an objectively reasonable officer in securing the scene in a more effective manner by ordering the passenger to alight from the car. (Emphasis supplied.)

637 A.2d at 167.

As it upheld the officer's ordering of the defendant-passenger out of the car, the Smith opinion provided an illustrative checklist of those types of observations that, although possibly falling short of articulable suspicion for a Terry "frisk," would justify the exercise of heightened caution in ordering a passenger out of a car:

Had the vehicle pulled to the side of the road without incident, the trooper would have had no basis for requiring the passenger to step out of the car. Here, however, Trooper Cacina witnessed the apparent passing of objects between the front and back seats. Gacina testified that while following the car as it pulled to the shoulder, he "did not have a view of the most important thing that [he] needed to have a view of; and that is [the occupants'] hands."

Thus the unusual movements, the early morning hour, and a largely deserted Turnpike are facts that warrant proceeding with extra caution in handling the occupants of the vehicle. The suspicious behavior by the occupants permitted the officer to exercise increased care to secure the scene even though the order that the passenger step out of the vehicle involved some intrusion on the passenger. That intrusion on the passenger's privacy interest is justified, however, because the suspicious movements in the car warranted a reasonably prudent officer's belief that the occupants of the car might be armed. Trooper Cacina thought that he would best be able to control the scene if each of the passengers stepped out of the vehicle and remained in complete view of the officers. Those circumstances were sufficient to justify Trooper Cacina's exercise of heightened caution in ordering Muhammad to step out of the car. (Emphasis supplied.)

637 A.2d at 168.

In People v. McLaurin, 70 N.Y.2d 779, 521 N.Y.S.2d 218, 515 N.E.2d 904 (1987), the New York Court of Appeals found of significance in this regard the facts 1) that the area where the car was stopped was abandoned, 2) that the hour was late, and 3) that the car was moving slowly with no lights on:

Here, the area was abandoned, the hour was late, and the fact that the vehicle — having already violated the speed limit — was moving slowly with no lights on was sufficiently peculiar to awaken suspicion as to the activity of the occupants. Under these circumstances, the officer was justified in acting to secure the safety of both officers by imposing upon defendant the de minimis intrusion of ordering him — a person whose progress had already been impeded as a necessary incident to the lawful stop of the vehicle — to step out while his partner investigated the driver's credentials. (Citation omitted.) (Emphasis supplied.)

521 N.Y.S.2d at 220, 515 N.E.2d at 906.

In Bethea v. Commonwealth, 245 Va. 416, 429 S.E.2d 211 (1993), the Supreme Court of Virginia discussed the "specific and articulable facts" that caused the stopping officer in that case to be "reasonably concerned for his safety":

The totality of the circumstances we consider here included a traffic stop in a high-crime area; similar traffic stops by the same officers two days earlier in the same neighborhood where weapons were discovered in a car; Bethea's actions immediately prior to the stop; Warren's 22 years of experience and his statements that Bethea's actions "startled" and "scared" him; and Warren's concern that Bethea might have weapons in the car. These circumstances constitute "specific and articulable facts" which show that Warren was reasonably concerned for his safety and believed that Bethea might have had access to weapons with which to assault him. These facts justified the intrusion on Bethea's Fourth Amendment rights that occurred when Warren asked him to get out of the car.

The fact that Warren did not immediately pat down or frisk Bethea does not belie Warren's concern for his own safety, as Bethea contends. Warren stated that he felt safer with Bethea outside the vehicle, not only because of Bethea's behavior, but also because of his concern that there might be weapons in the car. (Citation omitted.) (Emphasis supplied.)

429 S.E.2d at 213-14. See also State v. Reynolds, 753 S.W.2d 1, 2 (Mo. App. 1988).

In Search Seizure: A Treatise on the Fourth Amendment, § 9.4(a) at 516 (2d ed. 1987), Professor Wayne LaFave suggests two factors that may establish heightened suspicion that a passenger may be dangerous. One is suspicious movement on the part of the passenger. Another is where the stop is not for a routine traffic infraction but for more serious criminal activity, which may, ipso facto, be reasonable grounds to take preventive action with respect to the passengers as well as the driver:

But where the car has been stopped because of a suspicion of more serious criminal activity involving use of that vehicle, there may be greater reason to require the passengers to alight. In United States v. Hensley, where a car was stopped because of a flyer from another police department indicating the driver was wanted for investigation of robbery and should be considered armed and dangerous, the Court summarily concluded that the officer's conduct in having both the driver and his passenger step out of the car was "well within the permissible range." Even if the suspected offense is not serious, ordering a passenger out of the car for purposes of self-protection may sometimes be justified by the suspicious movements of the passenger. (Footnotes omitted.) (Emphasis supplied.)

See also United States v. Hensley, 469 U.S. 221, 224, 105 S.Ct. 675, 677, 83 L.Ed.2d 604, 609 (1985).

The issue of whether these are appropriate guidelines, however, is not before us. Judge Bollinger made no ruling that the circumstances in this case would not be sufficient, as a matter of constitutional law, to justify a sense of apprehension on the part of Trooper Hughes and appropriate preventive action by him. We are, therefore, intimating nothing as to whether, under appropriate circumstances, we would follow the reasoning of a case such as State v. Smith.

We are not holding, therefore, that the observations of Trooper Hughes in our case might not, under a set of guidelines similar to those used by State v. Smith, have constituted a reasonable justification for a sense of heightened caution or apprehension on the trooper's part and for his order for Wilson to alight from the car as a result thereof. No ruling in that regard was ever made or is before us for review. Judge Bollinger's finding, which was not clearly erroneous, was that Trooper Hughes did not act out of any such sense of apprehension. His suppression ruling, therefore, based on his findings of fact, was not in error.

ORDER TO SUPPRESS EVIDENCE AFFIRMED; COSTS TO BE PAID BY BALTIMORE COUNTY .