Last week, in Part I of this series, I discussed whether having a drug dog sniff a vehicle is a search if the drug dog might alert upon smelling hemp, a substance that is legal to possess. Today’s post focuses on what may be an even more significant question: if a dog alerts, does the alert provide probable cause to search?
The accuracy of drug dogs, and so the significance of a dog’s alert, has long been a subject of dispute in our criminal courts. The Supreme Court mostly put that debate to bed in Florida v. Harris, 568 U.S. 237 (2013), where it held that “[i]f a bona fide organization has certified a dog after testing his reliability in a controlled setting, a court can presume (subject to any conflicting evidence offered) that the dog’s alert provides probable cause to search.” Of course, the caveat about “conflicting evidence” is an important one for defenders, who may want to dig into a dog’s training and field performance records. But for present purposes, the more material point is that Harris didn’t address the impact of legal hemp. Does its holding still apply to dogs trained to alert to marijuana, and who may now wrongly alert upon smelling legal hemp?
“Sniff plus” is definitely sufficient. In some cases, it isn’t necessary to answer this question head on. For example, when there are other incriminating circumstances beyond the dog’s alert, the court may find that the “sniff plus” provides probable cause. Consider the recent case of State v. Walters, __ N.C. App. __ (2022). There, an officer stopped the defendant to serve an outstanding drug warrant. The officer had recently seized suspected drugs from the defendant in another incident, so the officer had a drug dog sniff the defendant’s vehicle. The dog alerted, which led to a search of the vehicle and the discovery of drugs. The court of appeals ruled that the outstanding warrant and the prior seizure, “[i]n addition to the positive indication by the dog,” provided probable cause notwithstanding the possibility that a dog could alert to legal hemp. In short, “sniff plus.” See also State v. Teague, __ N.C. App. __, 879 S.E.2d 881 (2022) (finding probable cause to search a parcel based on dog’s alert and other suspicious indicators, such as the method of packaging, source location, and lack of a valid telephone number for the recipient).
Things are hazier when there is no “plus.” We don’t have a recent North Carolina case on point where there is no “plus.” Courts in other jurisdictions have mostly found that dog alerts by themselves still provide probable cause despite the possibility of a false positive based on the dog’s detection of hemp. See United States v. Hayes, 2020 WL 4034309 (E.D. Tenn. Feb. 21, 2020) (finding that a drug dog alert provides probable cause to search a vehicle despite the legality of hemp because probable cause requires only a fair probability, not a certainty, and the fact “that the smell could have been hemp does not change the fact that it also could be . . . marijuana”); United States v. Deluca, 2022 WL 3451394 (10th Cir. Aug. 18, 2022) (unpublished) (“It is undisputed that Maverick was trained to alert on marijuana, heroin, methamphetamine, and cocaine. If hemp was added to this list of four controlled substances, Maverick’s alert on a car would still give rise to a high probability that a controlled substance is in the car as four of the five substances that Maverick could detect are illegal. Thus, we find that the officers had probable cause to search the car regardless of whether Maverick was trained to alert on legal hemp.”).
In jurisdictions where marijuana – not just hemp – is legal under state law, courts have been more skeptical of the value of dog sniffs. See State v. McKnight, 446 P.3d 397 (Colo. 2019) (holding that “a sniff from a drug-detection dog that is trained to alert to marijuana constitutes a search under . . . the Colorado Constitution because that sniff can detect lawful activity, namely the legal possession of up to one ounce of marijuana by adults,” and that a dog’s alert doesn’t provide probable cause to search a vehicle). Cf. Attorney General Alliance, Cannabis Law Deskbook §11:11 (“Although inconsistencies exist across jurisdictions, the growing trend among states that have legalized adult-use markets is that the “odor alone” of cannabis no longer constitutes probable cause to search a vehicle. The same is not true in states with medical-only markets or with decriminalization laws.”); Matter of T.T., 479 P.3d 598 (Or. Ct. App. 2021) (stating, in a case involving an officer rather than a dog smelling marijuana, that “given the legality of an adult possessing . . . marijuana in Oregon, the smell of marijuana in a car in which an adult is present is no longer remarkable, and, by itself, does not give rise to reasonable suspicion that it is being unlawfully possessed,” but relying on the occupants’ unusual travel pattern and other facts to support a detention for purposes of a drug investigation). But see People v. Vansteen, 2020 WL 772512 (Cal. Ct. App. Feb. 18, 2020) (unpublished) (“Though some possession of cannabis is legal . . . the possession of cannabis can still be illegal [for example, if beyond statutory limits]. Further, all the other substances Aero is trained to detect, methamphetamine, cocaine, heroin and ecstasy, are illegal to possess in any quantity. Because Aero’s alert raised a probability or substantial chance that defendant’s backpack contained contraband, probable cause existed for the officer’s search.”).
The idea that an alert alone doesn’t provide probable cause in a state where marijuana is legal makes sense to me given the relative prevalence of marijuana compared to other drugs. Federal data suggest that marijuana is used by four times as many people as all other illegal drugs combined. However, as one of the dissenters in McKnight noted, possession of marijuana remains illegal under federal law, which may provide a basis for finding probable cause in some cases.
Few if any of the cases in states where hemp is legal but marijuana is not have involved the presentation of evidence about the relative prevalence of hemp, marijuana, and other substances. It seems to me that such evidence could be highly pertinent. As a hypothetical example, if there were evidence that 50% of the population regularly use hemp products, 10% regularly use marijuana, and 1% regularly use cocaine or methamphetamine, a dog’s alert by itself would seem to be of doubtful value since the statistically most likely explanation for the alert would be the lawful possession of hemp. By contrast, if marijuana is much more prevalent than hemp, then a dog’s alert would be highly probative: sure, it could be alerting on hemp, but it probably is alerting on marijuana.
Practice suggestions. Existing case law in other states where marijuana remains illegal tends to support the idea that a dog alert provides probable cause to search a vehicle. But this is still an emerging issue and not one that is settled in North Carolina. Therefore, unless and until “legacy” drug dogs are replaced with new dogs not trained to alert on marijuana, or further court opinions clarify the law, the value of a dog’s alert is an appropriate issue for defense attorneys to raise and litigate. Note that the issue is pertinent in any case in which a drug dog’s alert provides probable cause to search, irrespective of which substance is ultimately found.
Attorneys who choose to raise this issue may want to confirm whether the dog was trained to alert on marijuana, what properties of marijuana the dog detects (if known), and whether the dog has had any training or experience with regard to hemp. Furthermore, as noted above, offering evidence about the prevalence of hemp and hemp products may bolster the argument that it is not merely possible that a dog would alert to hemp but that it is sufficiently likely to undermine probable cause.
By contrast, prosecutors may want to introduce evidence that marijuana is very prevalent, even more so than hemp. I suspect that many drug enforcement officers, including K-9 officers, may have experience concerning the relative prevalence of the two substances in their communities.