concurring.
I agree that the circuit court did not err in denying the motion to suppress after finding that the purpose of the stop had not been completed at the time K-9 Major was deployed. Yet, I write separately to address two aspects of the majority’s opinion that give me pause.
I. Standard of Review
The first pertains to our Court’s standard of review. In outlining the applicable standard of review, the majority states:
In reviewing a circuit court’s denial of a motion to suppress evidence, we conduct a de novo review based on the totality of the circumstances, reviewing findings of historical facts for clear error and determining whether those facts give rise to reasonable suspicion or probably cause, giving due weight to inferences drawn by the circuit court and proper defer-enee to the circuit court’s findings. E.g., Menne v. State, 2012 Ark. 37, 386 S.W.3d 451.
A finding is clearly erroneous, even if there is evidence to support it, when the appellate court, after review of the entire evidence, is left with the definite and firm conviction that a mistake has been made. E.g., Lee v. State, 2009 Ark. 255, 308 S.W.3d 596. We defer to the superiority of the circuit court to evaluate the credibility of witnesses who testify at a suppression hearing. E.g., Cockrell v. State, 2010 Ark. 258, 370 S.W.3d 197. We reverse only if the circuit court’s ruling is clearly against the preponderance of the evidence. Ritter v. State, 2011 Ark. 427, 385 S.W.3d 740.
While that formulation is virtually identical to what appears in scores of search and seizure opinions that this Court has issued over the past two decades, it amounts to a conflation of multiple tests that is, at best, ambiguous and, at worst, self-contradictory. It needs to be re-tooled.
The lineage of this standard of review can be traced to at least two separate lines of | ¡^authority that ultimately merged following Davis v. State, 351 Ark. 406, 94 S.W.3d 892 (2003). One line germinated following a concurrence by Justice Fogle-man, who expressed dismay over the then-existing lack of any discernible standard of review for suppression cases. Vault v. State, 256 Ark. 343, 345-48, 507 S.W.2d 111 (1974). After summarizing the prevailing language on appellate review, he lamented: “I now confess I do not know what this means.” Id. at 345, 507 S.W.2d 111.
Justice Fogleman’s plea for a new standard was promptly answered in Degler v. State, 257 Ark. 388, 517 S.W.2d 515 (1974), wherein the Court unveiled the “clearly against the preponderance of the evidence test.”1 This formulation was derived from a review standard that had been employed for chancery court appeals and was apparently viewed as a close cousin to the “clearly erroneous”' standard that the United States Supreme Court had championed in United States v. U.S. Gypsum Co., 333 U.S. 364, 68 S.Ct. 525, 92 L.Ed. 746 (1948). This language, and alternative forms of it, see, e.g., Cook v. State, 293 Ark. 103, 105, 732 S.W.2d 462 (1987); Findley v. State, 300 Ark. 265, 269, 778 S.W.2d 624 (1989) (“if we find the evidence to preponderate against the findings of the trial court, then it is our duty to reverse.”), periodically crop up in subsequent decisions leading into the twentieth century. See, e.g., Laime v. State, 347 Ark. 142, 153, 60 S.W.3d 464 (2001).
But by 2003, we found it necessary in the wake of Ornelas v. United States, 517 U.S. 690, 116 S.Ct. 1657, 134 L.Ed.2d 911 (1996), to develop the second line of authority for reviewing suppression cases. Thus, in Davis v. State, 351 Ark. 406, 94 S.W.3d 892 (2003), we jettisoned other language that had ^infiltrated our standard of review, see, e.g., Laime, 347 Ark. at 152, 60 S.W.3d 464 (2001)(requiring us to “view the evidence in the light most favorable to the State”), and reformulated it to read:
Our standard is that we conduct a de novo review based on the totality of the circumstances, reviewing findings of historical facts for clear error and determining whether those facts give rise to reasonable suspicion or probable cause, giving due weight to inferences drawn by the trial court.
Davis, 351 Ark. at 413, 94 S.W.3d 892 (2003) (citing Ornelas). Although this language is, at first blush somewhat paradoxical, the Davis test properly charts the multi-step analysis that an appellate court must conduct as it evaluates the mixed questions of law and fact that underlie a proper analysis of reasonable suspicion and probable cause. Significantly, the “clearly against the preponderance of the evidence” language was not mentioned as having any role in the analysis.
But by 2007, such “clearly against the preponderance of the evidence” verbiage inexplicably started re-appearing as a sort-of legal coda to the Davis search and seizure standard of review. See, e.g., Yarbrough v. State, 370 Ark. 31, 36, 257 S.W.3d 50 (2007)(citing Laime). That approach has continued, albeit inconsistently, up through the present. E.g., Menne v. State, 2012 Ark. 37, at 5, 386 S.W.3d 451. But see State v. Thompson, 2010 Ark. 294, at 5, 377 S.W.3d 207 (omitting such language in reversing the grant of a motion to suppress).
Regardless of its pedigree, the “clearly against the preponderance of the evidence” clause muddles what is already a somewhat-challenging2 .test. Fusing the preponderance of the evidence” language (which our civil juries are asked to employ in civil trials) with the additional adverb “clearly” creates a hybrid that admits of no easy interpretation. Does it lomean “clearly against the greater weight of the evidence?” If so, then that seems to fall short of the “clearly erroneous standard.” And if it is simply another way of saying “clearly erroneous” then it is confusingly redundant.
But beyond those inquiries, a more pressing question remains: to which part of “the circuit court’s ruling” does it apply? Does it apply strictly to findings of fact or does it also apply to conclusions about the existence of probable cause or reasonable suspicion? If applied to the latter, then how can it be squared with an appellate court’s mission to conduct a de novo review? Taken in context, I (like Justice Fogleman in Vault), do not know what our standard means.
Given such deficiencies inherent in the “clearly against the preponderance of the evidence” language, I respectfully submit that it is time to permanently retire the phrase from our search-and-seizure parlance and to return to the language that Davis announced, though with a slight clarifying elaboration. Such elaboration is merited since the threshold test articulated- — that we will conduct a de novo review — may seem inconsistent with subsequent language stating that we will review “findings of historical facts for clear error and determin[e] whether those facts give rise to reasonable suspicion or probable cause, giving due weight to inferences drawn by the trial court.” How can these two seemingly-contradictory standards coexist for the same core issue?
Fortunately, the two clauses can be harmonized. The explanation lies in the fact that the two standards apply to different aspects of the analytical process that an appellate court must employ when reviewing probably cause and reasonable suspicion rulings. Circuit courts and appellate courts addressing search and seizure issues seldom have the luxury of facing |2fiundisputed fact patterns. In general, it is the trial judges who sit in a far better position than we to evaluate the testimony because “they are there”; we are not. For that reason, we must indulge a healthy dose of deference to the circuit court’s superior position to determine the credibility of the witnesses at suppression hearings and the weight to be accorded to their testimony. Cockrell v. State, 2010 Ark. 258, at 10, 370 S.W.3d 197; Flanagan v. State, 368 Ark. 143, 154, 243 S.W.3d 866 (2006). In a related vein, we must give due deference to the circuit court’s findings of historical facts — for example, what events led up to the stop or search, what statements were (or were not) made during the course of the seizure, whether the detainee was Mirandized or, as in the case at bar, whether a dog handler reasonably inferred that his dog actually alerted to illegal drugs. Accordingly, the circuit court’s findings of historical facts may not be overturned unless the appellate court, after review of the entire evidence, is left with a definite and firm conviction that a mistake has been made. E.g., Lee v. State, 2009 Ark. 255, at 4, 308 S.W.3d 596.
Yet such deference evaporates when assessing whether the facts give rise to a determination of reasonable suspicion or probable cause. As the United States Supreme Court pointed out in Ornelas, “as a general matter determinations of reasonable suspicion and probable cause should be reviewed de novo on appeal.” 517 U.S. at 699, 116 S.Ct. 1657. That is because such determinations ultimately involve mixed questions of law and fact. Only by exercising de novo review can appellate courts appropriately unify precedent which, in turn, can enable law enforcement officers to make better and more-informed decisions in the field. See id. at 697-98, 116 S.Ct. 1657; United States v. Arvizu, 534 U.S. 266, 275, 122 S.Ct. 744, 151 L.Ed.2d 740 (2002). Thus, it is up to the appellate court to reach the legal conclusion whether, under the totality of the circumstances and | ^considering the historical facts as addressed by the circuit court, a reasonably-objective police officer either would have reasonable suspicion that the person stopped has committed, or is about to commit, a criminal act, or would have probable cause to believe that, upon search, contraband or evidence of criminal activity will be found.
Distilled to its essence, the appellate court must apply a clearly erroneous standard over the circuit court’s historical factual findings3 and then employ a de novo review as it determines whether those facts — taken in totality — indeed constitute reasonable suspicion or probable cause.
This interlude into the ramifications of standard of review sets the stage for what is a crucial issue in this case. I agree that, under State v. Thompson, 2010 Ark. 294, 377 S.W.3d 207, a positive canine sniff can provide probable cause for the subsequent vehicle search. Yet in this instance, it appears that K-9 Major was not actually deployed until slightly more than 15 minutes after the traffic stop had commenced. It is well-established that, after routine checks and related paperwork are completed during the course of a lawful traffic stop, | ¡^“continued detention of the driver can become unreasonable” unless the officer has developed adequate reason to suspect that criminal activity is afoot. Sims v. State, 356 Ark. 507, 514, 157 S.W.3d 530 (2004). Mr. Jackson argues that Trooper Behnke had sufficient time to complete the ticket-writing process and that he had developed no additional reasonable suspicion during the course of the traffic stop to justify Jackson’s continued detention. Thus, the question is: had the initial stop ended by the time Trooper Behnke developed probable cause to initiate a warrant-less search? This turns on whether the purpose of the initial stop had ended by the time K-9 Major performed his sniff.
II. Determining the End Point of the Traffic Stop
In addressing the question of when a traffic stop legitimately ends, the majority properly recognizes that “there is no bright-line rule, as it would likely create an inflexibility that is at odds with Rule 3.1” of the Arkansas Rules of Criminal Procedure. Yet the majority then observes: “While we may not have a bright-line rule for when a stop is legitimately completed, our case law has consistently held that a stop is not concluded when the officer has not returned the license, paperwork or ticket.” (citing Yarbrough v. State, 370 Ark. 31, 257 S.W.3d 50 (2007)).4
This gives rise to my second point of concern with the majority opinion. I agree that Yarbrough and Sims v. State, 356 Ark. 507, 157 S.W.3d 530 (2004), both underscore that the traffic stop normally has not ended until the paperwork has been completed. See also Menne v. State, 2012 Ark. 37, at 5-6, 386 S.W.3d 451. But whether construed as a “bright-line rule” |29or merely as a “consistent holding,” I do not read any of these decisions to support the notion that, just because the paperwork has not been returned, the stop cannot be challenged as unduly protracted. While such a rule might add predictability and simplicity to the legal task, it could provide a fecund ground for mischief. To be sure, while the return of paperwork may be a presumptively-appropriate starting point to gauge the justifiable duration of an initial traffic stop, Menne, Yar-brough, and Sims should not be construed as creating some sort of safe harbor that immunizes an investigating officer’s purposeful delay from judicial scrutiny. The State ultimately conceded as much in oral argument.
Thus, absent independent reasonable suspicion to detain a motorist, no officer should be empowered to delay completion or delivery of the traffic stop paperwork simply to continue trolling for probable cause or, in a case such as this one, to buy time so that the canine cavalry can arrive to save the day. Illinois v. Caballes, 543 U.S. 405, 407, 125 S.Ct. 834, 160 L.Ed.2d 842 (2005).
In the case at bar, Trooper Behnke claimed that he had cultivated reasonable suspicion to prolong the stop during the course of the traffic stop. Among other key “indicators” he noted that he had seen a road atlas, fast food wrappers, and a single small suitcase inside the pickup. The Trooper also noted that, according to its rental agreement, the pickup had been due to be returned the previous day. But such “evidence” proves too little. Indeed, if fast food wrappers and a road atlas (or GPS device) constitute legitimate touchstones for criminal activity, then the highways of Arkansas are hopelessly teeming with felons. His observation that the rental contract had recently expired might have added more vitality to his suspicions but for the fact Trooper Behnke never suggested that he felt that the vehicle was indeed stolen. Nor did he otherwise persuasively link the vehicle’s status per se as a rental to | .^potential evidence of criminality. Cf. Dowty v. State, 363 Ark. 1, 13, 210 S.W.3d 850 (2005) (officer asserted that drug dealers routinely use rental cars to avoid seizure of their personal vehicles upon arrest).
Were these the universe of Trooper Behnke’s justification for initiating the warrantless search, then in conducting our de novo review for probable cause, I would deem the search unconstitutional. See Lilley v. State, 362 Ark. 436, 442-44, 208 S.W.3d 785 (2005). But those did not comprise Trooper Behnke’s full menu of criminal “indicators.” In addition to pointing to the occupants’ demeanor and delayed answers to questions, Trooper Behnke relied upon the drug sniff that K-9 Major undertook beginning roughly 15 minutes 44 seconds after he had stopped the pickup.
It is here that our duty to give deference to the trial court’s determination of historical facts must take center stage. For the circuit court’s decision that the Trooper truly was still waiting for a return of the criminal history check from ACIC is pivotal.5 While the record on that point is not crystal clear, such a factual determination requires our deference. As such, I cannot say that I have a definite and firm conviction that the circuit court made a mistake.
Accordingly, I agree that the deployment of K-9 Major took place during the course of the initial stop. If K-9 Major was “reliable” and if Trooper Behnke had adequate basis to conclude that he indeed had alerted, then Trooper Behnke had probable cause to conduct the subsequent warrantless search. The majority found no basis to conclude that the circuit court was clearly erroneous in reaching these findings of historical fact. And based on the record before us, I must agree.
|S1III. The Canine Sniff Reliability /Probable Cause Framework
One final observation regarding the evi-dentiary thicket surrounding drug sniff cases bears mentioning. The Court of Appeals dispatched with the challenge to the drug dog’s reliability by stating “Trooper Behnke provided both his and his dog’s training and certification records. That is all that is required to establish the dog’s reliability.” 2012 Ark.App. 508, at 5, 2012 WL 4194657. That somewhat oversimplifies the analysis where, as here, the defendant challenged reliability. To be sure, particularly in light of Florida v. Harris, 568 U.S.-, 133 S.Ct. 1050, 185 L.Ed.2d 61 (2013), the State’s threshold burden is to establish either that a bona fide organization certified the K-9 after testing the dog in a controlled setting or that the dog recently and successfully completed a training program that evaluated his proficiency in locating drugs. If the State fails on both of these inquiries, then no probable cause based on the canine sniff can normally exist because the officer could not have reasonably relied on the dog’s alert.
On the other hand, if the State meets its threshold burden, then the defendant still has several avenues through which to attack the canine sniff as a basis for probable cause. First, the defendant may prove that the certification or training program was too lax or employed faulty methods.6 Second, the defendant may seek to establish that, though certified, the dog did not perform adequately in either the training/certification program or in field performance. Such proof may undermine the dog’s reliability, which could vitiate the Inexistence of probable cause. Third, should the defendant fail to convince the trial judge of such inadequate performance, the defendant may then seek to challenge the dog handler’s credentials by proving that the handler was not properly trained to correctly interpret the dog’s actions. To complete the analysis, the trial judge should then determine whether the handler’s account of the dog’s alert was indeed credible.7
In the case at bar, the State offered up proof that K-9 Major had been certified and that he and Trooper Behnke had completed training programs that evaluated the dog’s proficiency. The State offered the video of the stop, as captured by the camera in the Trooper Behnke’s car and had the Trooper comment on what was shown therein. The defendant did not mount a strong challenge to the certification or training programs and instead focused his primary challenge on K-9 Major’s track record in the field — specifically his roughly 14% false positive rate — and attacked Trooper Behnke’s conclusions through cross-examination and through his own K-9 expert witness. But in the final analysis, the trial judge weighed the proof, and concluded that Trooper Behnke had a proper basis for interpreting K-9 Major’s labored breathing and approach to the pickup’s window as evidence of an alert, which in turn, provided Trooper Behnke with probable cause to search the vehicle. Since the trial court was able to watch and listen to the testimony of Trooper Behnke and the Defendant’s K-9 expert, I cannot say that our review of the record compels a different conclusion.
. The Court announced: "We now set the issue at rest by stating explicitly that in each case we will make an independent determination based upon the totality of the circumstances and that the trial judge’s findings of voluntariness will not be set aside unless it is clearly against the preponderance of the evidence." 257 Ark. at 392, 517 S.W.2d 515.
. See United States v. Arvizu, 534 U.S. 266, 278, 122 S.Ct. 744, 151 L.Ed.2d 740 (2002) (Scalia, J., concurring).
. It may be that, at some point, certain types of evidence and the manner in which they are preserved in the record may render obsolete the very justification for deference to the trial court. Thus, with the proliferation of "dash cam” and surveillance video evidence, an appellate court may arguably be in as good a position as the trial court to evaluate certain historical facts, assuming no corresponding authenticity or credibility-related issues need to be addressed. Cf. Scott v. Harris, 550 U.S. 372, 127 S.Ct. 1769, 167 L.Ed.2d 686 (2007) (reaching vastly different interpretations from those of the district court and Court of Appeals upon review of "dash cam” videotape depicting police chase of motorist). To the extent that an appellate court's analysis of probable cause or reasonable suspicion rests in part on interpretations of unchallenged video or audio evidence then arguably the justification for deference to he trial court's "better position” to evaluate facts tends to fade. And yet, there are countervailing notions rooted in the very structure of our trial court — appellate court system that suggest that fact-finding lies most appropriately in the trial courts' domain. In this case, the video was in my opinion inconclusive. Given the fact that the trial judge was both able to review it and to hear live testimony about it, I feel compelled to defer to the circuit court’s factual findings and related inferences on this point.
. The State invokes this principle as justification for the somewhat tardy drug sniff, arguing that "[t]he legitimate purpose of a traffic stop does not end ... until the officer delivers the vehicle’s documentation, along with a ticket or warning, to the driver.” Brief at 10.
. Implicit in that decision is the recognition that Trooper Behnke did not drag his feet on the paperwork in order to initiate a delayed canine sniff.
. For example, a defendant may seek to prove that the dog was not subjected to sufficiently diverse environments or that blind testing (in order to prevent cueing) was not employed by the handler. If the defendant succeeds in establishing that the program was too lax or riddled with faulty methods, then, again, the Officer could not have reasonably relied on the dog's alert.
. This determination may rest on such issues as whether the Officer cued the dog (consciously or unconsciously), whether the team was working in unfamiliar conditions, or whether the handler’s account is belied by video, audio, internal inconsistency, or other credible testimony.