(03-923) 543 U.S. 405 (2005)
207 Ill. 2d 504, 802 N. E. 2d 202, vacated and remanded.
[ Stevens ]
[ Souter ]
[ Ginsburg ]
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version

Souter, J., dissenting


No. 03—923



[January 24, 2005]

Justice Souter, dissenting.

I would hold that using the dog for the purposes of determining the presence of marijuana in the car’s trunk was a search unauthorized as an incident of the speeding stop and unjustified on any other ground. I would accordingly affirm the judgment of the Supreme Court of Illinois, and I respectfully dissent.

In United States v. Place, 462 U.S. 696 (1983), we categorized the sniff of the narcotics-seeking dog as “sui generis” under the Fourth Amendment and held it was not a search. Id., at 707. The classification rests not only upon the limited nature of the intrusion, but on a further premise that experience has shown to be untenable, the assumption that trained sniffing dogs do not err. What we have learned about the fallibility of dogs in the years since Place was decided would itself be reason to call for reconsidering Place’s decision against treating the intentional use of a trained dog as a search. The portent of this very case, however, adds insistence to the call, for an uncritical adherence to Place would render the Fourth Amendment indifferent to suspicionless and indiscriminate sweeps of cars in parking garages and pedestrians on sidewalks; if a sniff is not preceded by a seizure subject to Fourth Amendment notice, it escapes Fourth Amendment review entirely unless it is treated as a search. We should not wait for these developments to occur before rethinking Place’s analysis, which invites such untoward consequences.1

At the heart both of Place and the Court’s opinion today is the proposition that sniffs by a trained dog are sui generis because a reaction by the dog in going alert is a response to nothing but the presence of contraband.2 See ibid. (“[T]he sniff discloses only the presence or absence of narcotics, a contraband item”); ante, at 3—4 (assuming “that a canine sniff by a well-trained narcotics dog will only reveal ‘the presence or absence of narcotics, a contraband item’ ” (quoting Place, supra, at 707)). Hence, the argument goes, because the sniff can only reveal the presence of items devoid of any legal use, the sniff “does not implicate legitimate privacy interests” and is not to be treated as a search. Ante, at 4.

The infallible dog, however, is a creature of legal fiction. Although the Supreme Court of Illinois did not get into the sniffing averages of drug dogs, their supposed infallibility is belied by judicial opinions describing well-trained animals sniffing and alerting with less than perfect accuracy, whether owing to errors by their handlers, the limitations of the dogs themselves, or even the pervasive contamination of currency by cocaine. See, e.g., United States v. Kennedy, 131 F.3d 1371, 1378 (CA10 1997) (describing a dog that had a 71% accuracy rate); United States v. Scarborough, 128 F.3d 1373, 1378, n. 3 (CA10 1997) (describing a dog that erroneously alerted 4 times out of 19 while working for the postal service and 8% of the time over its entire career); United States v. Limares, 269 F.3d 794, 797 (CA7 2001) (accepting as reliable a dog that gave false positives between 7 and 38% of the time); Laime v. State, 347 Ark. 142, 159, 60 S. W. 3d 464, 476 (2001) (speaking of a dog that made between 10 and 50 errors); United States v. $242,484.00, 351 F.3d 499, 511 (CA11 2003) (noting that because as much as 80% of all currency in circulation contains drug residue, a dog alert “is of little value”), vacated on other grounds by rehearing en banc, 357 F.3d 1225 (CA11 2004); United States v. Carr, 25 F.3d 1194, 1214—1217 (CA3 1994) (Becker, J., concurring in part and dissenting in part) (“[A] substantial portion of United States currency … is tainted with sufficient traces of controlled substances to cause a trained canine to alert to their presence”). Indeed, a study cited by Illinois in this case for the proposition that dog sniffs are “generally reliable” shows that dogs in artificial testing situations return false positives anywhere from 12.5 to 60% of the time, depending on the length of the search. See Reply Brief for Petitioner 13; K. Garner et al., Duty Cycle of the Detector Dog: A Baseline Study 12 (Apr. 2001) (prepared under Federal Aviation Administration grant by the Institute for Biological Detection Systems of Auburn University). In practical terms, the evidence is clear that the dog that alerts hundreds of times will be wrong dozens of times.

Once the dog’s fallibility is recognized, however, that ends the justification claimed in Place for treating the sniff as sui generis under the Fourth Amendment: the sniff alert does not necessarily signal hidden contraband, and opening the container or enclosed space whose emanations the dog has sensed will not necessarily reveal contraband or any other evidence of crime. This is not, of course, to deny that a dog’s reaction may provide reasonable suspicion, or probable cause, to search the container or enclosure; the Fourth Amendment does not demand certainty of success to justify a search for evidence or contraband. The point is simply that the sniff and alert cannot claim the certainty that Place assumed, both in treating the deliberate use of sniffing dogs as sui generis and then taking that characterization as a reason to say they are not searches subject to Fourth Amendment scrutiny. And when that aura of uniqueness disappears, there is no basis in Place’s reasoning, and no good reason otherwise, to ignore the actual function that dog sniffs perform. They are conducted to obtain information about the contents of private spaces beyond anything that human senses could perceive, even when conventionally enhanced. The information is not provided by independent third parties beyond the reach of constitutional limitations, but gathered by the government’s own officers in order to justify searches of the traditional sort, which may or may not reveal evidence of crime but will disclose anything meant to be kept private in the area searched. Thus in practice the government’s use of a trained narcotics dog functions as a limited search to reveal undisclosed facts about private enclosures, to be used to justify a further and complete search of the enclosed area. And given the fallibility of the dog, the sniff is the first step in a process that may disclose “intimate details” without revealing contraband, just as a thermal-imaging device might do, as described in Kyllo v. United States, 533 U.S. 27 (2001).3

It makes sense, then, to treat a sniff as the search that it amounts to in practice, and to rely on the body of our Fourth Amendment cases, including Kyllo, in deciding whether such a search is reasonable. As a general proposition, using a dog to sniff for drugs is subject to the rule that the object of enforcing criminal laws does not, without more, justify suspicionless Fourth Amendment intrusions. See Indianapolis v. Edmond, 531 U.S. 32, 41—42 (2000). Since the police claim to have had no particular suspicion that Caballes was violating any drug law,4 this sniff search must stand or fall on its being ancillary to the traffic stop that led up to it. It is true that the police had probable cause to stop the car for an offense committed in the officer’s presence, which Caballes concedes could have justified his arrest. See Brief for Respondent 31. There is no occasion to consider authority incident to arrest, however, see Knowles v. Iowa, 525 U.S. 113 (1998), for the police did nothing more than detain Caballes long enough to check his record and write a ticket. As a consequence, the reasonableness of the search must be assessed in relation to the actual delay the police chose to impose, and as Justice Ginsburg points out in her opinion, post, at 3—4, the Fourth Amendment consequences of stopping for a traffic citation are settled law.

In Berkemer v. McCarty, 468 U.S. 420, 439—440 (1984), followed in Knowles, supra, at 488, we held that the analogue of the common traffic stop was the limited detention for investigation authorized by Terry v. Ohio, 392 U.S. 1 (1968). While Terry authorized a restricted incidental search for weapons when reasonable suspicion warrants such a safety measure, id., at 25—26, the Court took care to keep a Terry stop from automatically becoming a foot in the door for all investigatory purposes; the permissible intrusion was bounded by the justification for the detention, id., at 29—30.5 Although facts disclosed by enquiry within this limit might give grounds to go further, the government could not otherwise take advantage of a suspect’s immobility to search for evidence unrelated to the reason for the detention. That has to be the rule unless Terry is going to become an open-sesame for general searches, and that rule requires holding that the police do not have reasonable grounds to conduct sniff searches for drugs simply because they have stopped someone to receive a ticket for a highway offense. Since the police had no indication of illegal activity beyond the speed of the car in this case, the sniff search should be held unreasonable under the Fourth Amendment and its fruits should be suppressed.

Nothing in the case relied upon by the Court, United States v. Jacobsen, 466 U.S. 109 (1984), unsettled the limit of reasonable enquiry adopted in Terry. In Jacobsen, the Court found that no Fourth Amendment search occurred when federal agents analyzed powder they had already lawfully obtained. The Court noted that because the test could only reveal whether the powder was cocaine, the owner had no legitimate privacy interest at stake. 466 U.S., at 123. As already explained, however, the use of a sniffing dog in cases like this is significantly different and properly treated as a search that does indeed implicate Fourth Amendment protection.

In Jacobsen, once the powder was analyzed, that was effectively the end of the matter: either the powder was cocaine, a fact the owner had no legitimate interest in concealing, or it was not cocaine, in which case the test revealed nothing about the powder or anything else that was not already legitimately obvious to the police. But in the case of the dog sniff, the dog does not smell the disclosed contraband; it smells a closed container. An affirmative reaction therefore does not identify a substance the police already legitimately possess, but informs the police instead merely of a reasonable chance of finding contraband they have yet to put their hands on. The police will then open the container and discover whatever lies within, be it marijuana or the owner’s private papers. Thus, while Jacobsen could rely on the assumption that the enquiry in question would either show with certainty that a known substance was contraband or would reveal nothing more, both the certainty and the limit on disclosure that may follow are missing when the dog sniffs the car.6

The Court today does not go so far as to say explicitly that sniff searches by dogs trained to sense contraband always get a free pass under the Fourth Amendment, since it reserves judgment on the constitutional significance of sniffs assumed to be more intrusive than a dog’s walk around a stopped car, ante, at 4. For this reason, I do not take the Court’s reliance on Jacobsen as actually signaling recognition of a broad authority to conduct suspicionless sniffs for drugs in any parked car, about which Justice Ginsburg is rightly concerned, post, at 5—6, or on the person of any pedestrian minding his own business on a sidewalk. But the Court’s stated reasoning provides no apparent stopping point short of such excesses. For the sake of providing a workable framework to analyze cases on facts like these, which are certain to come along, I would treat the dog sniff as the familiar search it is in fact, subject to scrutiny under the Fourth Amendment.7


1. I also join Justice Ginsburg’s dissent, post, p. ___. Without directly reexamining the soundness of the Court’s analysis of government dog sniffs in Place, she demonstrates that investigation into a matter beyond the subject of the traffic stop here offends the rule in Terry v. Ohio, 392 U.S. 1 (1968), the analysis I, too, adopt.

2. Another proffered justification for sui generis status is that a dog sniff is a particularly nonintrusive procedure. United States v. Place, 462 U.S. 696, 707 (1983). I agree with Justice Ginsburg that the introduction of a dog to a traffic stop (let alone an encounter with someone walking down the street) can in fact be quite intrusive. Post, at 4—5 (dissenting opinion).

3. Kyllo was concerned with whether a search occurred when the police used a thermal-imaging device on a house to detect heat emanations associated with high-powered marijuana-growing lamps. In concluding that using the device was a search, the Court stressed that the “Government [may not] us[e] a device … to explore details of the home that would previously have been unknowable without physical intrusion.” 533 U.S., at 40. Any difference between the dwelling in Kyllo and the trunk of the car here may go to the issue of the reasonableness of the respective searches, but it has no bearing on the question of search or no search. Nor is it significant that Kyllo’s imaging device would disclose personal details immediately, whereas they would be revealed only in the further step of opening the enclosed space following the dog’s alert reaction; in practical terms the same values protected by the Fourth Amendment are at stake in each case. The justifications required by the Fourth Amendment may or may not differ as between the two practices, but if constitutional scrutiny is in order for the imager, it is in order for the dog.

4. Despite the remarkable fact that the police pulled over a car for going 71 miles an hour on I—80, the State maintains that excessive speed was the only reason for the stop, and the case comes to us on that assumption.

5. Thus, in Place itself, the Government officials had independent grounds to suspect that the luggage in question contained contraband before they employed the dog sniff. 462 U.S., at 698 (describing how Place had acted suspiciously in line at the airport and had labeled his luggage with inconsistent and fictional addresses).

6. It would also be error to claim that some variant of the plain-view doctrine excuses the lack of justification for the dog sniff in this case. When an officer observes an object left by its owner in plain view, no search occurs because the owner has exhibited “no intention to keep [the object] to himself.” Katz v. United States, 389 U.S. 347, 361 (1967) (Harlan, J., concurring). In contrast, when an individual conceals his possessions from the world, he has grounds to expect some degree of privacy. While plain view may be enhanced somewhat by technology, see, e.g., Dow Chemical Co. v. United States, 476 U.S. 227 (1986) (allowing for aerial surveillance of an industrial complex), there are limits. As Kyllo v. United States, 533 U.S. 27, 33 (2001), explained in treating the thermal-imaging device as outside the plain-view doctrine, “[w]e have previously reserved judgment as to how much technological enhancement of ordinary perception” turns mere observation into a Fourth Amendment search. While Kyllo laid special emphasis on the heightened privacy expectations that surround the home, closed car trunks are accorded some level of privacy protection. See, e.g., New York v. Belton, 453 U.S. 454, 460, n. 4 (1981) (holding that even a search incident to arrest in a vehicle does not itself permit a search of the trunk). As a result, if Fourth Amendment protections are to have meaning in the face of superhuman, yet fallible, techniques like the use of trained dogs, those techniques must be justified on the basis of their reasonableness, lest everything be deemed in plain view.

7. I should take care myself to reserve judgment about a possible case significantly unlike this one. All of us are concerned not to prejudge a claim of authority to detect explosives and dangerous chemical or biological weapons that might be carried by a terrorist who prompts no individualized suspicion. Suffice it to say here that what is a reasonable search depends in part on demonstrated risk. Unreasonable sniff searches for marijuana are not necessarily unreasonable sniff searches for destructive or deadly material if suicide bombs are a societal risk.