Last Halloween the highest court in the land literally went to the dogs, as the Supreme Court heard argument in cases involving two Florida narcotics-detection dogs, Aldo and Franky. In Florida v. Harris, Aldo finished best in show, with the Court holding for all practical purposes that a trained drug dog’s positive alert creates probable cause to search. The Court put Franky on a shorter leash, however, concluding in Florida v. Jardines that a dog conducts a search under the Fourth Amendment when it physically intrudes on a suspect’s property for the purpose of collecting evidence.
Both decisions had the effect of articulating bright-line rules, thereby deviating from precedent that favored a more amorphous standard considering all the surrounding circumstances. Totality-of-the-circumstances inquiries can be messy and unpredictable, but if done fairly they can also lead to sensible outcomes. By contrast, purportedly clear rules tend to oversimplify, creating an inherent risk of overinclusion or underinclusion. The recent rulings in the drug-dog cases illustrate these dangers, as Harris exhibits overconfidence in the accuracy of drug-dog alerts while Jardines threatens to underprotect less privileged socioeconomic classes.
I. Aldo and Probable Cause
On two separate occasions, Clayton Harris had the misfortune of being pulled over by Officer William Wheetley for a minor traffic offense. Both times, the officer’s drug-detection dog, Aldo, alerted to the door handle on the driver’s side of Harris’s truck. Although the officer’s subsequent searches of the truck did not reveal any drugs the dog was trained to detect, the first uncovered materials used in manufacturing methamphetamine. Harris was arrested following that first search for possessing pseudoephedrine, a key ingredient of methamphetamine, and he later admitted that he regularly used and produced the narcotic. The Florida Supreme Court ruled that Aldo’s positive alert was insufficient to create probable cause to search the truck absent specific details about the dog’s training regimen and performance in the field, including any alerts that had not led to the discovery of narcotics.
In an opinion written by Justice Kagan, a unanimous Supreme Court reversed the Florida court. The Supreme Court’s decision was not particularly surprising, given its ruling thirty years earlier in Illinois v. Gates redefining probable cause using a totality-of-the-circumstances approach. Describing the lower court’s opinion in Harris as creating “a strict evidentiary checklist,” the Supreme Court noted that its concept of probable cause, by contrast, “rejected rigid rules, bright-line tests, and mechanistic inquiries in favor of a more flexible, all-things-considered approach.”
More problematic was the Court’s sweeping rule that a drug dog’s positive alert is enough to create a presumption of probable cause so long as the dog either “recently and successfully completed a training program” or was certified by a “bona fide organization.” Rather than a totality-of-the-circumstances approach, this statement resembles the bright-line rules the Court’s probable cause jurisprudence has recently avoided.
Admittedly, the Court qualified its holding with the parenthetical “subject to any conflicting evidence,” and then went on to require that defendants be given an opportunity to contest a drug dog’s reliability either by cross-examining the dog’s handler or presenting their own witnesses. This caveat is presumably what enabled the Court to claim its opinion called for an examination of “all the facts surrounding a dog’s alert,” just like “every inquiry into probable cause.”
For all practical purposes, however, an alert by a trained dog will now lead to a finding of probable cause because the defense is not likely to have access to information necessary to challenge a dog’s reliability. Details about training programs the dog and its handler completed are in the hands of the government, and a defendant who was not on the scene during the dog sniff cannot know whether the dog was cued by its handler or working under “unfamiliar conditions.” In addition, evidence concerning the dog’s performance in the field may not even be available. In Harris, for example, records were kept only of Aldo’s alerts that actually led to an arrest. Even where such information exists, some courts have denied defense requests for discovery of a dog’s records, and in Harris, the Solicitor General took the position that such evidence should “typically” not be discoverable.
Even if field records are turned over to the defendant, the Court’s ruling in Harris suggested that this data is not entitled to much weight in probable cause determinations. Although at one point the Court’s opinion cryptically mentioned that field records “may sometimes be relevant,” it devoted a paragraph to defending the view that “in most cases” field performance has “relatively limited import.” In addition to providing little information about false negatives, the Court thought field data may “markedly overstate . . . real false positives” if a dog alerts where drugs are too well concealed or too small to be discovered. Though these hypotheticals seem somewhat unrealistic, the Court expressed a distinct preference for evidence of a dog’s performance in “controlled testing environments.” But a controlled classroom provides an imperfect sense of a dog’s accuracy as well. It may not have the same distractions or complications present in the field, and may generate skewed results absent double-blind testing where neither the handler nor the dog knows the location of the drugs. Moreover, the “practical and common-sensical” nature of Gates’s totality-of-the-circumstances test suggests that the dog’s actual track record under similar circumstances ought to be considered for whatever light it can shed on the probable cause inquiry. Therefore, the largely meaningless opportunity Harris affords defendants to rebut the prosecution’s showing of probable cause does not diminish the bright-line nature of the Court’s ruling.
Additionally, the Court offered no justification for creating a drug-dog exception to the Gates totality-of-the-circumstances analysis. Although the Solicitor General’s amicus brief pleaded for a bright-line test to provide guidance to law enforcement, the Government did not distinguish drug-dog alerts from any other probable cause inquiry. Gates justified the decision to abandon the rigid two-pronged definition of probable cause the Court had previously applied to informants on the grounds that tips come in “many shapes and sizes,” and here too the reliability of drug-dog alerts varies. No national or uniform standards govern the programs used to train drug-detection dogs. Moreover, the accuracy of an alert depends on the particular dog’s abilities as well as the conditions under which it is working.
In addition to deviating from thirty years of probable cause precedent, the opinion in Harris glossed over concerns about drug dogs’ reliability—an issue that has surfaced in the literature and that was addressed by the court below, in the Supreme Court briefs, and at oral argument. Though the scientific understanding of drug-detection dogs is still developing, research has shown that a significant percentage of positive alerts do not lead to the discovery of narcotics. In many cases, errors result from miscommunications between the handler and the dog—sometimes even from the handler’s subconscious cues, such as prolonging the dog’s exposure to places where the handler suspects narcotics are hidden. Moreover, without some indication of the base rate of criminality—i.e., the odds of drug activity—an accurate assessment of a dog’s reliability cannot be made.
Furthermore, as the Court pointed out in Harris, a dog “recognizes an odor, not a drug.” As a result, a positive alert may occur when the dog detects the smell of either a chemical ingredient of narcotics that is present in legal products, or an innocuous item associated with drugs or the dog’s training. A dog’s detection of such items using its sense of smell should be treated no differently than a police officer’s visual observation of them, which is not necessarily sufficient in and of itself to give rise to probable cause.
The dog may also give a positive signal that yields no evidence of drugs where it detects residual odor. The Harris opinion did address this issue, denying that an alert to residual odor constitutes a false positive. Rather, the Court argued, a dog “should alert” to a scent “even if the substance is gone.” And the Court rejected as irrelevant what it called “the mere chance” that drugs might no longer be present, instead concluding without support or analysis that there is still “a fair probability” of discovering “either drugs or evidence of a drug crime (like the precursor chemicals in Harris’s truck).”
Of course, the residual-odor theory can be proffered—and cannot be refuted—for any unproductive search, and in fact was the Court’s “likely” explanation for Aldo’s two positive alerts to Harris’s vehicle. Even more important, probable cause to search requires an assessment of the odds that evidence is currently located in the place to be searched, not that it was there at some indeterminate time in the past. Officer Wheetley could not say for how long Aldo could still detect residual odor, and estimates of dogs’ capacity to smell residual odors vary depending on, among other things, the particular drug at issue. Admittedly, as the Court pointed out, a police officer who smells the distinctive odor of marijuana might likewise be detecting residual odor. But a dog, unlike an officer, cannot communicate the strength of a detected scent or its level of certainty that the odor is actually present. Paradoxically, then, a dog’s much more sensitive nose is a less reliable indicator that drugs are still on the scene than the human sense of smell.
A defensible argument can be made both for defining probable cause by a totality-of-the-circumstances inquiry and for creating bright-line rules, or at least “presumptive rules for . . . recurring . . . issues.” Although the Court’s Fourth Amendment jurisprudence is notoriously inconsistent on the subject of bright lines, at the very least it should be able to stick to one approach on any given Fourth Amendment issue. Moreover, fairness demands a two-way street, and the Court should not reject only bright-line rules that make it harder to prove probable cause while endorsing ones that facilitate the prosecution’s case. Harris’s overconfidence in the accuracy of drug dogs and its disregard for the distinct possibility of error further suggest that the Court acted prematurely by creating a drug-dog exception to the totality-of-the-circumstances test. Just like many fixed rules create problems of overinclusion or underinclusion, the one that effectively emerges from the decision in Harris is overinclusive and threatens to close off debate concerning an investigative tool whose reliability depends on a technical understanding that is still evolving.
II. Franky and Fourth Amendment “Searches”
After receiving an uncorroborated tip that Joelis Jardines was growing marijuana in his house, a detective and his narcotics-detection dog, Franky, walked up the driveway to Jardines’s front porch. When the dog alerted positively at the base of the front door, a search warrant for the home was issued and a number of marijuana plants were seized. In a five-to-four decision written by Justice Scalia, the Court determined that the police had conducted a “search” within the meaning of the Fourth Amendment.
The majority relied on last year’s decision in United States v. Jones, in which the Court held that law enforcement officials effected a Fourth Amendment search when they installed a GPS device on a suspect’s car and tracked the car’s movements for four weeks. In an opinion also authored by Justice Scalia, the Jones majority made the surprising announcement that the Katz reasonable-expectation-of-privacy definition of a search that the Court had consistently applied for forty-five years merely “added to,” but did not “substitute for, the common-law trespassory test.” Reasoning that the Fourth Amendment should at least protect the same “‘degree of privacy’” recognized at the time of the framing, the Jones majority relied on common-law notions of trespass to find that government officials conduct a search when they “obtain information by physically intruding on a constitutionally protected area.” Accordingly, Justice Scalia, a persistent critic of Katz, saw no need to analyze GPS tracking under the reasonable-expectation-of-privacy test.
Following in Jones’s footsteps—though without ever using the word trespass—the majority in Jardines held that the police committed an “unlicensed physical intrusion” on “a constitutionally protected area” by “gathering information” in the curtilage of Jardines’s home. The Court acknowledged that not every entry onto private property constitutes a search given that permission to enter “may be implied from the habits of the country.” Therefore, the Court explained, police may “approach a home and knock,” just like any solicitor, delivery person, or other member of the public, but “the background social norms that invite a visitor to the front door do not invite him there to conduct a search.”
Even though Justice Scalia’s opinion again declined to apply the reasonable-expectation-of-privacy test, three of the five Justices in the majority joined a separate opinion written by Justice Kagan, which described Jardines as an “‘easy cas[e] . . .’ twice over” because the dog also effected a search under Katz. The police violated Jardines’s reasonable expectation of privacy, the concurrence concluded, because Franky “reveal[ed] within the confines of a home what they could not otherwise have found there.”
The four dissenters, in an opinion written by Justice Alito, disagreed on both counts. On the Jones test, the dissent argued that under an implied license by custom, trespass law allows even “unwelcome visitors”—including the police and their dogs—“to use a walkway to approach the front door of a house and to remain there for a brief time.” On the Katz question, the dissenting opinion thought there was no reasonable expectation of privacy in “odors emanating from a house [that] may be detected from locations that are open to the public.”
Although adherents of the Jones test defend it as a way of avoiding the “thorny” and “‘vexing’” issues created by Katz, the elements of the Jardines definition each create line-drawing issues of their own. On the most mundane level, the Court might be right that Jardines’s front porch is the “classic exemplar” of the curtilage of a home. But it was on less solid ground when it claimed that the curtilage is “generally ‘clearly marked’” and “‘easily understood.’” In fact, the Solicitor General’s office argued that Jardines’s front porch and walkway “may be considered outside the curtilage.” And even the Court admitted that the multi-factor definition of “curtilage” articulated in United States v. Dunn was not a “finely tuned formula” that can be “mechanically applied” to “yield a ‘correct’ answer to all extent-of-curtilage questions.” The Court proved quite prescient on that point, as the Dunn factors have spawned a tremendous amount of litigation.
Second, an implied license by custom can vary depending on the norms of a particular location, contrary to the Court’s traditional insistence that the contours of the Fourth Amendment do not “vary from place to place and from time to time.” Moreover, permission to enter property is subject to the complete discretion of the property owner, and individuals can therefore opt out of the traditional customs. Thus, the implied invitation to “the Nation’s Girl Scouts and trick-or-treaters”—and to the police—presumably can be revoked by a locked front gate, or a sign that says “No Solicitors,” “No Dogs,” or even “No Police.” As a result, Jardines puts police officers and courts in the unenviable position of trying to “guess . . . whether landowners . . . erected fences sufficiently high [or] posted a sufficient number of warning signs.”
The final element of the Jardines test turns on a police officer’s subjective motivation for entering the property. In an effort to remain faithful to its two-year-old decision in Kentucky v. King, the Jardines majority acknowledged that police do not conduct a search by approaching a home and knocking “in order to speak with the occupant, because all are invited to do that.” The seeds of this intent inquiry can be found in Jones, where the Court observed that a trespass alone does not qualify as a Fourth Amendment search unless it involves “an attempt to find something or to obtain information.” Yet Jones’s casual references to law enforcement objectives did not attract much attention, perhaps because it was hard to imagine any other plausible explanation for attaching a GPS device to Jones’s vehicle. Therefore, Justice Breyer can hardly be faulted for expressing surprise during oral argument in Jardines when Justice Scalia suggested that law enforcement intent is relevant to the Jones analysis.
The likely source of Justice Breyer’s confusion is, as Justice Scalia himself noted in his 2011 majority opinion in Ashcroft v. al-Kidd, the Court’s “almost uniform reject[ion of] invitations to probe subjective intent” in Fourth Amendment cases because the Amendment “regulates conduct rather than thoughts.” According to Justice Scalia, the one “unusual” and “narrow” exception to this practice comes in the administrative inspection and special needs cases, which recognize an exception to the warrant requirement only for searches conducted for a proper regulatory purpose. Thus, the Court previously considered subjective intent “irrelevant” in defining a Fourth Amendment search under Katz, on the grounds that the key is “the objective effect of [the officers’] actions,” not that they may have been “‘motivated by a law enforcement purpose.’”
In explaining why a police officer’s motive is nonetheless an element of Jardines’s definition of a search, Justice Scalia reasoned that law enforcement intent is only irrelevant in Fourth Amendment cases involving searches that are “objectively reasonable.” In contrast, Jardines turned on “whether the officer’s conduct was an objectively reasonable search.” But this is the first time the Court has considered an officer’s purpose when analyzing the “search” half of that equation, as opposed to the “objectively reasonable” inquiry at issue in the administrative inspection cases.
This final element of the Jardines definition is not only illustrative of the fluctuation between objective and subjective standards running through the Court’s Fourth Amendment rulings, but is also likely to pose difficult line-drawing questions for the courts. Does any use of a drug dog on the curtilage constitute a search, in light of the Jardines majority’s observation that “a stranger” is not expected “to explore the curtilage of the home with trained drug dogs”? Or would the case have come out the other way if the officers had also knocked on the door while Franky was sniffing around, given other language in the opinion suggesting that police exceed the scope of their implied invitation only if they “enter . . . in order to do nothing but conduct a search”? If so, is the sequencing of the knock and dog sniff important? Or does even a preliminary knock exceed the scope of the implied license because the Court focused elsewhere on the inappropriateness of a visitor conducting a search “before saying hello and asking permission”? On the facts before it in Jardines, the Court thought the officers’ “behavior objectively reveal[ed] a purpose to conduct a search.” But in other cases, law enforcement conduct will be more equivocal—especially if police are trained to steer clear of Jardines—and courts will be forced to conduct a subjective inquiry into the officers’ state of mind.
Even if lower courts prove capable of administering the Supreme Court’s new test and discerning law enforcement’s motivations, the majority did not explain why homeowners impliedly invite police onto their property to conduct investigations that involve questioning but not searching. In Kentucky v. King, for example, police had witnessed an apparent drug deal outside an apartment building, and they were knocking on the door of the apartment they thought the suspect had entered. The officers in King were certainly trying to “find something” or “obtain information” as required by Jones. But the Jardines opinion at some points (though not others) subtly tweaked that language to focus on the narrower law enforcement purpose of “conduct[ing] a search.” That modification of Jones adds an element of circularity to the definition of a search. Furthermore, if “[t]he scope of a license” can be restricted to “a specific purpose,” it seems unreasonable to assume a homeowner would impliedly invite the police onto her property to use any investigative technique designed to generate evidence of her guilt—or to approach the front door for any reason other than to solicit donations for a police association charity, respond to a call for help, or look for witnesses or assistance solving a crime committed by someone else.
The inconsistencies and confusion generated by the majority’s approach are not the most troubling aspects of Jardines. The Court’s Fourth Amendment jurisprudence has traditionally—and appropriately—accorded the same level of protection to every “castle,” no matter how grand or humble. Yet the focus on property rights threatens to privilege those of higher socioeconomic status. The Court considered the front porch of Jardines’s single-family residence part of the protected curtilage of a home, but it is difficult to see what “constitutionally protected area” would have been invaded if Franky had alerted from the public walkway outside a row house or trailer home or from the hallway of an apartment building or residential hotel that could be accessed by anyone. Likewise, Jardines presumably means that police conduct a search when they enter the curtilage to search the garbage of those who pay a premium to have it collected from the backyard, but not necessarily when the garbage is found on the curb.
The Katz approach endorsed by the three concurring Justices does not share the same defect of underinclusiveness because it does not turn on where a drug dog happens to be standing. It does, however, create problems of its own. The reasonable-expectation-of-privacy analysis is hardly a model of clarity either, though it offers an alternative to a physical-intrusion test that is no less outdated today than it was when Katz was decided.
Nevertheless, Justice Kagan’s conclusion in Jardines—that the dog sniff violated the defendant’s reasonable expectation of privacy—is difficult to reconcile with the Court’s precedents. The concurring opinion claimed that the thermal imaging decision in Kyllo v. United States “already resolved” the case by holding that police effect a search when they “‘use a device that is not in general public use, to explore details of the home that would previously have been unknowable without physical intrusion.’” Several years after Kyllo, however, Illinois v. Caballes distinguished thermal imagers from narcotics-detection dogs in refusing to find that a dog’s positive alert to a car during a lawful traffic stop constituted a search. Justice Kagan’s concurrence in Jardines in turn attempted to distinguish Caballes on the grounds that the Court has routinely found a lower expectation of privacy in cars than in homes. But, as reflected in the precedents Justice Kagan cited, the Court has used that rationale to create exceptions to the warrant requirement and has never denied that car interiors carry a sufficient expectation of privacy to meet the Katz test at issue in Jardines.
Moreover, the distinction between homes and vehicles was not the salient factor for the Caballes majority. Rather, the Caballes Court claimed that its conclusion was “entirely consistent” with Kyllo because the critical fact in Kyllo was the thermal imager’s ability to “detect lawful activity,” whereas a drug dog that “only reveals the possession of contraband ‘compromises no legitimate privacy interest.’” For this point, Caballes relied on dicta in United States v. Place, where the Court described a dog sniff as a sui generis investigative tool that does not effect a search because it “discloses only the presence or absence of narcotics” and therefore “does not expose noncontraband items that otherwise would remain hidden from public view.” Jardines may have been at the intersection of Place/Caballes and Kyllo, but it was clearly an overstatement to say the issue had already been decided.
This point was not lost on the Jardines dissenters, who observed that the Kagan concurrence made “a very similar, if not identical argument” to one that was raised by Justice Souter’s dissent in Caballes and obviously failed to persuade a majority of that Court. But Justice Alito’s dissenting opinion in Jardines did not elaborate further, and therefore did not expressly endorse either Place’s characterization of dogs as sui generis or the notion that there is no reasonable expectation of privacy in contraband. Rather, the dissent relied primarily on a “plain view” rationale, that one lacks a reasonable expectation of privacy in anything police agents, including dogs, can detect with their own senses. But given that the canine nose is so much more sensitive than its human counterpart—and that here it was trained on a home—this argument required the dissent to deal with Kyllo. Justice Alito made an unpersuasive effort to dismiss Kyllo as “a decision about the use of new technology” and the dangers of “‘advancing technology.’” Dogs, the dissent argued, are neither new nor a form of technology. But, again, that was not the Caballes Court’s understanding of Kyllo. Caballes distinguished dogs from thermal imagers because they do not reveal any lawful activity, not because they are more analogous to human noses than technological devices. Moreover, a dog may fall somewhere along the continuum between a person and a mechanical instrument, but relying on the plain view doctrine to compare a police officer’s sense of smell with something that is between 10,000 and 100,000 times more sensitive is “unsound.” Thus, the dissent was no more successful than the concurrence in trying to reconcile its position with both Kyllo and the Place line of cases.
The Jardines majority, of course, avoided that difficult task by skirting the Katz analysis altogether. Rather, Justice Scalia purported to adopt a bright-line rule—that a physical intrusion into a constitutionally protected area to perform a search constitutes a Fourth Amendment search. But like many attempts to forge a clear rule, the Court’s test turns out to be fuzzier than it first appears and is underinclusive by affording less protection to certain types of residences.
The Katz reasonable-expectation-of-privacy analysis may be amorphous and subject to varying interpretations, but it is a commonsense standard just like Gates’s totality-of-the-circumstances concept of probable cause. The problem is not necessarily the Katz test itself, but the Court’s application of it in ways that defy reasonable expectations, such as the refusal to find a reasonable expectation of privacy whenever a suspect is in public or turns information over to a third party. Like other inquiries that consider all the surrounding circumstances, Katz could lead to sensible results and avoid the underinclusion problem plaguing Jardines if the Court were to truly analyze what reasonable people expect.
III. The Relationship Between Aldo and Franky
Although the rulings in Harris and Jardines are each subject to criticism in their own right, so too is the relationship between the two decisions. It was presumably no coincidence that both cases were argued on the same day, but the tension between them was left largely unexplored in either opinion.
The majority’s property-based ruling in Jardines incentivizes police to keep their dogs off private property, thereby diminishing the dogs’ reliability. A dog’s accuracy depends on its proximity to the target, and there is little scientific data measuring how reliably a dog can detect narcotics even from outside a home much less from beyond the curtilage. As Jardines moves dogs farther away from a suspect’s property, the probability of error increases and Harris’s confidence that a positive alert gives rise to probable cause becomes even more questionable.
As discussed above, none of the opinions in Jardines reiterated the argument initially made in Place—that dog sniffs are binary searches that disclose only the presence or absence of contraband. In fact, Jardines’s only mention of Place came in Justice Scalia’s opinion, which cited the case for the proposition that the State had “support in our case law” for a Katz argument the majority then declined to reach. But both the Kagan concurrence and the Alito dissent did discuss Caballes, with the concurring Justices distinguishing it and the dissenters relying on it. None of the Justices seemed willing to repudiate Caballes’s holding that the dog sniff of a properly stopped vehicle is not a search, even though both opinions ignored the decision’s central rationale. Caballes did not turn on the lesser expectation of privacy in vehicles or on the analogy to the human sense of smell. Rather, Caballes was fundamentally derived from Place.
To be sure, there is good reason to push Place off center stage. First, Place’s sui generis argument is of questionable pedigree—whether a dog sniff effects a Fourth Amendment search was not briefed or argued in either the Supreme Court or the court below, and yet the Court resolved it in two paragraphs of dicta unsullied by a single supporting citation. Nevertheless, the dicta took on a life of its own, with the Court repeating it in later opinions—and ultimately Caballes—that relied on Place (and then each other) for support. On a more substantive level, the definition of a search should not turn on what the police uncover, for the Fourth Amendment is designed to protect the guilty as well as the innocent. As Justice Kennedy observed several times during the oral argument in Jardines, the no-reasonable-expectation-of-privacy-in-contraband mantra is “circular” reasoning. (Ironically, he then chose to join the opinion that strayed the least from Place and Caballes, Justice Alito’s dissent.) Finally, as an empirical matter, scientific understanding of drug-detection dogs has advanced in the thirty years since Place was decided such that the sui generis label no longer seems accurate. Justice Souter forcefully made this point in his lone dissent in Caballes, observing that “[t]he infallible dog . . . is a creature of legal fiction.” While his argument fell on deaf ears, the current Court’s apparent inclination to snub Place and refashion Caballes may signal a welcome willingness to rethink the wisdom of the Place dicta.
But that puts Jardines on a collision course with Harris. If a dog sniff is not viewed as a sui generis investigative tool that discloses only the presence of contraband, it becomes harder to defend Harris’s conclusion that an alert from a trained dog generally gives rise to probable cause.
Contrary to the sentiments expressed in the popular song, then, the dog days are far from over, and Harris and Jardines have not yet put the Fourth Amendment issues surrounding drug-detection dogs to bed. Not only is there an inherent tension between the two opinions, but they both generate rigid rules in place of the more commonsense totality-of-the-circumstances standards favored in the Court’s precedents. By deliberately sidestepping the Katz reasonable-expectation-of-privacy analysis in Jardines, and effectively ignoring the Gates totality-of-the-circumstances concept of probable cause in Harris, both opinions essentially draw bright lines that, like many rules, lack precision and thus suffer from underinclusiveness or overinclusiveness.
* Professor and Polisher Family Distinguished Faculty Scholar, Penn State Dickinson School of Law.
 Harris, 133 S. Ct. at 1055, 1056.
 Id. at 1057.
 Id. at 1058.
 Id. at 1057–58 (suggesting ways a defendant might challenge a dog’s reliability).
 Id. at 1054.
 Brief for the United States as Amicus Curiae Supporting Petitioner at 20, Harris, 133 S. Ct. 1050 (No. 11-817) (link); see also Transcript of Oral Argument at 23, Harris, 133 S. Ct. 1050 (No. 11-817) (citing the “burden . . . on law enforcement”) (link).
 Harris, 133 S. Ct. at 1056, 1057.
 Id. at 1057.
 Id. The Court’s suggestion that an unproductive alert might also result from the residual odor of drugs is discussed infra at notes 31–38 and accompanying text.
 See Kenneth Furton et al., The Scientific Working Group on Dog and Orthogonal Detector Guidelines (SWGDOG) 136 (2010); see also id. at 138 (recommending that dogs’ field-performance records be kept) (link).
 Harris, 133 S. Ct. at 1055; see also Illinois v. Gates, 462 U.S. 213, 238 (1983).
 See Max Minzner, Putting Probability Back into Probable Cause, 87 Tex. L. Rev. 913, 913–15 (2009) (analogizing probable cause to baseball batting averages and calling for consideration of the success rates of individual officers and their units in assessing probable cause) (link); cf. Erica Goldberg, Getting Beyond Intuition in the Probable Cause Inquiry, Lewis & Clark L. Rev. (forthcoming) (manuscript at 31), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2116389 (suggesting that statistical evidence be used in cases, like those involving drug dogs, where probable cause is based on “a device . . . that has essentially replaced the police officer”) (link).
 See Brief for the United States, supra note 12, at 27–28.
 Gates, 462 U.S. at 232.
 See William S. Helton, Overview of Scent Detection Work: Issues and Opportunities, in Canine Ergonomics: The Science of Working Dogs 83, 85 (William S. Helton ed., 2009) [hereinafter Canine Ergonomics].
 See Harris v. State, 71 So. 3d 756, 768–69 (Fla. 2011), rev’d, 133 S. Ct. 1050 (2013); Brief for Respondent at 33–36, Harris, 133 S. Ct. 1050 (No. 11-817) (link); Transcript of Oral Argument, supra note 12, at 13–15.
 See Canine Ergonomics, supra note 22, at x (“The scientific analysis of working dogs is emerging but underdeveloped . . . .”); Furton et al., supra note 16, at 10 (noting that there is a “limited amount of reliable scientific information” and “limited peer reviewed research on error rates”).
 See, e.g., Brief of Amici Curiae: The National Ass’n of Criminal Defense Lawyers et al. at 6–9, Harris, 133 S. Ct. 1050 (No. 11-817) (citing various studies showing success rates of 20%, 26%, and 44%) (link); Lisa Lit et al., Handler Beliefs Affect Scent Detection Dog Outcomes, 14 Animal Cognition 387, 390 (2011) (finding that only one out of eighteen dogs did not falsely alert in a double-blind test) (link).
 See William S. Helton, Conclusion: Working Dogs and the Future, in Canine Ergonomics, supra note 22, at 325, 326; Lit et al., supra note 25, at 390–91 (finding in double-blind study that dogs were more likely to erroneously alert in places the handler had falsely been told contained contraband).
 See Myers, supra note 11, at 13–16 (using Bayes’s Theorem to explain that a positive alert by a dog with a 90%–95% accuracy rate indicates the presence of drugs in only approximately 16% of cases in a population where only one in fifty people are in possession of drugs).
 Harris, 133 S. Ct. at 1056 n.2.
 See Myers, supra note 11, at 4 (citing baggies and air freshener by way of example).
 See 2 Wayne R. LaFave, Search and Seizure: A Treatise on the Fourth Amendment § 3.6(b), at 388–98 (5th ed. 2012) (citing conflicting cases).
 Harris, 133 S. Ct. at 1056 n.2 (emphasis added).
 Id. at 1059 (justifying this conclusion on the grounds that Harris admitted he used and manufactured methamphetamine “on a regular basis”).
 See Harris v. State, 71 So. 3d 756, 773–74 (Fla. 2011), rev’d, 133 S. Ct. 1050 (2013).
 See Brief of Amici Curiae Fourth Amendment Scholars in Support of Respondent at 28–30, Harris, 133 S. Ct. 1050 (No. 11-817) (citing sources reporting that the chemical ingredients of cocaine and heroin can be detected for two hours and three days respectively) (link).
 Harris, 133 S. Ct. at 1056 n.2.
 See Canine Ergonomics, supra note 22, at 87.
 Albert W. Alschuler, Bright Line Fever and the Fourth Amendment, 45 U. Pitt. L. Rev. 227, 256 (1984).
 For discussion of the distinction between rules and standards, see generally Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 Duke L.J. 557 (1992) (link); Cass R. Sunstein, Problems with Rules, 83 Calif. L. Rev. 953 (1995) (link).
 Florida v. Jardines, 133 S. Ct. 1409, 1417–18 (2013).
 Jones, 132 S. Ct. at 952 (emphasis omitted). But see 1 LaFave, supra note 30, § 2.1(e), at 593 (observing that Katz “seemed to sound the death knell” for the trespass test).
 Jones, 132 S. Ct. at 950 & n.3 (quoting Kyllo v. United States, 533 U.S. 27, 34 (2001) (link)). But see Orin S. Kerr, The Curious History of Fourth Amendment Searches, Sup. Ct. Rev. (forthcoming) (manuscript at 24) (arguing that “Jones purports to restore a trespass test that never previously existed”), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2154611 (link).
 See Jones, 132 S. Ct. at 950. Five members of the Court did, however, apply the reasonable-expectation-of-privacy analysis. Four Justices, in an opinion written by Justice Alito, were critical of the majority’s novel trespass approach, but agreed that “lengthy monitoring” by means of a GPS device in the course of investigating “most offenses” constitutes a search under Katz. Id. at 964 (Alito, J., concurring in the judgment). Justice Sotomayor also wrote a separate opinion, agreeing with both the majority and the Alito concurrence, and suggesting she might be prepared to find that “unique attributes of GPS surveillance” make “even short-term monitoring” a search under Katz. Id. at 955 (Sotomayor, J., concurring).
 For the suggestion that the Court did so to avoid reversing the Florida Supreme Court’s ruling that no trespass occurred under state tort law, see Bradley Pollina, Florida v. Jardines: Why the Supreme Court Did Not Say “Trespass,” 3 Wake Forest L. Rev. Online 19, 21 (2013), http://wakeforestlawreview.com/wp-content/uploads/2013/05/PollinaArticle1.pdf (link).
 Florida v. Jardines, 133 S. Ct. 1409, 1414, 1415 (2013).
 Id. at 1416.
 See id. at 1417.
 Id. at 1420 (Kagan, J., concurring) (alteration in original) (quoting id. at 1417 (majority opinion)).
 Id. at 1419.
 Id. at 1420, 1422 (Alito, J., dissenting).
 Id. at 1421.
 United States v. Jones, 132 S. Ct. 945, 954 (2012) (quoting id. at 962 (Alito, J., concurring in the judgment)); see also Florida v. Jardines, 133 S. Ct. 1407, 1417 (2013) (claiming that the “property-rights baseline . . . keeps easy cases easy”).
 Jardines, 133 S. Ct. at 1415.
 Id. (quoting Oliver v. United States, 466 U.S. 170, 182 n.12 (1984)).
 Brief for the United States as Amicus Curiae Supporting Petitioner at 24 n.8, Jardines, 133 S. Ct. 1409 (No. 11-564).
 480 U.S. 294, 301 (1987) (listing four factors to be considered in defining curtilage: how far the area in question was from the home, whether it was “within an enclosure surrounding the home,” how it was used, and what precautions were taken to “protect [it] from observation by people passing by”) (link).
 See 1 LaFave, supra note 30, § 2.4(a), at 811–16.
 See Dan B. Dobbs, The Law of Torts § 96, at 220 (2000).
 See Dobbs, supra note 64, § 96, at 220 (noting that homeowners “have an opportunity, at low cost, to express [their] dissent from the custom,” for example, by “post[ing] a sign forbidding salespeople to enter”).
 Florida v. Jardines, 133 S. Ct. 1409, 1415–16 (2013).
 See Brief for the United States, supra note 61, at 24 (admitting that “[t]he analysis may be different if . . . the occupant took steps to bar all visitors”); Transcript of Oral Argument at 10, Jardines, 133 S. Ct. 1409 (No. 11-564) (arguing on behalf of the State that “police officers can walk up the front path, absent a sign or something”) (link).
 Jardines, 133 S. Ct. at 1416 n.4 (emphasis omitted).
 United States v. Jones, 132 S. Ct. 945, 951 n.5 (2012); see also id. at 949 (concluding that the “installation . . . and . . . use of [a GPS] to monitor” a suspect’s movements qualifies as a search (emphasis added)).
 See Transcript of Oral Argument, supra note 68, at 44 (Justice Scalia chiding Jardines’s attorney for being “wrong not to accept” the relevance of police intent because “our cases support it”); id. at 51 (Justice Breyer noting that the Court’s Fourth Amendment decisions focus on police “behavior” rather than “subjective motive,” and then commenting, “I don’t know what in Jones changed that”).
 Id. at 2083. For a critique of the Court’s failure to analyze al-Kidd under the administrative inspection rubric, see Kit Kinports, Camreta and al-Kidd: The Supreme Court, the Fourth Amendment, and Witnesses, 102 J. Crim. L. & Criminology 283 (2012) (link).
 Florida v. Jardines, 133 S. Ct. 1409, 1416–17 (2013).
 See Kinports, supra note 40, at 77–88.
 Jardines, 133 S. Ct. at 1415 n.2.
 Id. at 1416 n.4 (emphasis added).
 Id. at 1416 (emphasis added). Note that a second officer did knock on Jardines’s door after Franky had alerted and left the scene. See Jardines v. State, 73 So. 3d 34, 37 n.1 (Fla. 2011), aff’d, 133 S. Ct. 1409 (link).
 Jardines, 133 S. Ct. at 1417.
 See Kentucky v. King, 131 S. Ct. 1849, 1854 (2011).
 United States v. Jones, 132 S. Ct. 945, 951 n.5 (2012).
 Jardines, 133 S. Ct. at 1417. But cf. id. at 1414, 1417 (elsewhere using terms more in line with Jones—“obtain[ing]” and “gathering information” and “gather[ing] evidence”).
 Id. at 1416.
 Cf. Illinois v. Lidster, 540 U.S. 419, 423 (2004) (distinguishing checkpoints whose “purpose was not to determine whether a vehicle’s occupants were committing a crime, but to ask vehicle occupants, as members of the public, for their help in providing information about a crime in all likelihood committed by others”) (link).
 Cf. David A. Sklansky, Back to the Future: Kyllo, Katz, and Common Law, 72 Miss. L.J. 143, 185 (2002) (pointing out that the amount of privacy the common law accorded in the eighteenth century “depended strongly on who [you] were”) (link).
 See Katz v. United States, 389 U.S. 347, 362 (1967) (Harlan, J., concurring) (observing that the physical-intrusion test “is, in the present day, bad physics as well as bad law”).
 Florida v. Jardines, 133 S. Ct. 1409, 1419 (2013) (Kagan, J., concurring) (quoting Kyllo v. United States, 533 U.S. 27, 40 (2001)).
 See Jardines, 133 S. Ct. at 1419 n.1 (Kagan, J., concurring).
 See, e.g., New York v. Class, 475 U.S. 106, 114–15 (1986) (concluding that intrusion into a car’s interior constituted a search) (link); Cardwell v. Lewis, 417 U.S. 583, 591–92 & n.8 (1974) (plurality opinion) (noting that only the exterior of the car was inspected there) (link).
 Jardines, 133 S. Ct. at 1424 (Alito, J., dissenting).
 Id. For a possible explanation why the dissent ignored Place, see infra note 110 and accompanying text.
 Jardines, 133 S. Ct. at 1425 (Alito, J., dissenting) (quoting Kyllo v. United States, 533 U.S. 27, 35 (2001)).
 1 LaFave, supra note 30, § 2.2(g), at 692.
 See Christopher Slobogin & Joseph E. Schumacher, Reasonable Expectations of Privacy and Autonomy in Fourth Amendment Cases: An Empirical Look at “Understandings Recognized and Permitted by Society,” 42 Duke L.J. 727, 732 (1993) (concluding based on empirical research that a number of Supreme Court opinions in the Katz line of cases “do not reflect societal understandings”) (link).
 See Sherry F. Colb, What Is a Search? Two Conceptual Flaws in Fourth Amendment Doctrine and Some Hints of a Remedy, 55 Stan. L. Rev. 119, 123 (2002) (criticizing the Court for “equating risk-taking with inviting exposure and equating limited-audience with whole-world self-exposure”).
 Florida v. Jardines, 133 S. Ct. 1409, 1417 (2013).
 Transcript of Oral Argument, supra note 68, at 3, 23–24, 49 (also denying that “all the rules go out the window” when contraband is involved).
 Caballes, 543 U.S. at 411 (Souter, J., dissenting). For a discussion of possible explanations for false alerts, see supra notes 21–38 and accompanying text.
Copyright 2013 Northwestern University
Cite as: 108 Nw. U. L. Rev. Colloquy 64 (2013), http://www.law.northwestern.edu/lawreview/colloquy/2013/7/LRColl2013n7Kinports.pdf.