State v. Sines

379 P.3d 502, 359 Or. 41, 2016 WL 1459146, 2016 Ore. LEXIS 252

• Court: Oregon Supreme Court
• Decided: April 14, 2016
• Docket: CC 06FE1054AB; CA A146025; SC S062493
• Status: Published

Opinion

BALMER, C. J.

This case requires us to consider whether a private citizen’s seizure of criminal evidence was subject to suppression at trial as the fruit of an unlawful government search. Defendant came to the attention of law enforcement after his housekeeper anonymously called the child protective services division of the Department of Human Services (DHS) and said that she suspected that defendant might be sexually abusing his adopted daughter. The housekeeper’s suspicions had been raised after finding an unusual “discharge” on several pairs of the child’s underwear, and she told DHS that she had considered taking a pair for authorities to examine. In response to a question from the housekeeper, the DHS employee who handled the call said that he would be able to connect the housekeeper with someone in law enforcement who could analyze the underwear and confirm or refute her concerns. The DHS employee told the housekeeper several times that he could not tell her to take the victim’s underwear. The next day the housekeeper obtained a pair of the victim’s underwear, and the following day she turned it over to the police. Based on that evidence and other statements by the housekeeper, police obtained a warrant and searched defendant’s house, after which defendant was arrested and charged with a number of sex crimes. Defendant’s motion to suppress the evidence obtained through the search and seizure of the underwear was denied, and he was convicted on four counts of first-degree sexual abuse.

The Court of Appeals reversed, holding that the trial court had erred in denying defendant’s motion to suppress. The court concluded that, although the underwear had been procured by a private person, there was nevertheless sufficient contact between state officials and the private person that the warrantless search and seizure constituted state action, in violation of Article I, section 9, of the Oregon Constitution. State v. Sines, 263 Or App 343, 328 P3d 747 (2014). For the reasons set out below, we reverse the Court of Appeals decision and remand to that court for consideration of other issues raised but not addressed in defendant’s appeal. 1

FACTS

We take the relevant facts from the record and the Court of Appeals opinion, setting them out consistently with the trial court’s explicit and implicit findings. State v. Ehly, 317 Or 66, 75, 854 P2d 421 (1993). We review the trial court’s denial of defendant’s motion to suppress for errors of law.

Early in 2005, defendant and his wife adopted two siblings—T, a young girl, and V, her brother. Approximately one year later, defendant’s wife and biological son moved out of the family residence. Defendant’s housekeeper subsequently began to discover indications of what she thought might be sexual activity between defendant and the then-nine-year-old T.

The housekeeper had observed, among other things, that T was sleeping with defendant in his bedroom and, in the bed, the housekeeper had found a “type of Vaseline stuff’ “[u]p to half way up [defendant’s] sheets,” as well as signs of the substance’s use in the bathroom. Based on her observation of Vaseline-like handprints on the bathroom walls, the housekeeper believed that defendant “had been having sex with somebody in the bathroom area,” despite the fact that defendant’s wife had moved out and defendant had no girlfriend. When the housekeeper, concerned about the possible abuse of T, suggested to defendant “to go get a girlfriend,” he told her “he did not need one, he had T.”

Defendant’s housekeeper also observed a “lot of discharge” in various pairs of T’s underwear, noting that in some, the crotch of the garment had become so stiff that they had to be thrown away. According to the housekeeper, the heavily-stained children’s underwear appeared abnormal in that they did not look as if they had been worn by a child, but rather by a sexually active adult.

In March 2006, after consulting with another employee of defendant who worked in the home and also suspected that defendant was having sex with T, the housekeeper anonymously called a DHS “tip line” regarding the possible abuse. According to the DHS employee who took her call at around noon, the housekeeper appeared to be on the verge of tears, and first asked what the agency could determine from a pair of underwear. The DHS employee testified that he had responded by saying, “Well, there’s a lab here locally that can probably tell a lot. What’s your concern?” 2 The housekeeper then gradually related her observations regarding defendant and T, including the nature and extent of the discharge that she had observed on T’s underwear, and told the DHS employee that she was considering taking a pair from defendant’s house. The DHS employee reiterated several times that he could not tell her to take that kind of action, and that it was her decision. At the hearing on defendant’s motion to suppress, the housekeeper stated that the DHS employee never asked her to get a pair of underwear; she said, “No. Never.” She also testified, “It was my idea.” The DHS employee gave the housekeeper his direct telephone number, expecting, based on their conversation, that she probably would take the underwear. The housekeeper retained her anonymity throughout their conversation, although, she eventally disclosed the names of defendant and defendant’s wife.

Following the housekeeper’s phone call, the DHS employee contacted a deputy at the Deschutes County Sheriffs Office. As a general matter, DHS policy called for safety checks to be conducted within 24 hours after receipt of a call regarding suspected abuse, unless there was good cause for delay. The DHS employee and the deputy sheriff instead decided to assign the case a five-day response time to see whether the housekeeper would take any action. Neither the DHS policy nor the decision to extend the time period was communicated to the housekeeper.

The same day that she talked to DHS, the housekeeper called another employee of defendant who similarly suspected abuse and who was planning to work at defendant’s house the next day. The housekeeper told the other employee, “I’m thinking we need to get something of evidence,” and “I’m thinking underwear.” The other employee said, “I’ll see what I can do.” The following day, while defendant was taking T and her brother to school, the other employee went into the laundry room of defendant’s house and took the first pair of T’s underwear that she saw. She turned the underwear over to the housekeeper after work. The housekeeper then called her DHS contact, who arranged for her to bring the underwear to DHS and the deputy sheriff the next day, which she did.

The child’s underwear was immediately delivered to the Oregon State Police Crime Lab in Bend for testing. When the tests revealed spermatozoa on the garment, authorities obtained and executed a warrant to search defendant’s house. Defendant was arrested at that time, and police seized other evidence, including a nightgown, pajama pants, a bathing suit, and jeans, all belonging to T. Tests conducted on those items revealed additional evidence of spermatozoa and seminal fluid.

PROCEEDINGS BELOW

Defendant was charged with nine counts of first-degree sexual abuse, one count of first-degree rape, and two counts of first-degree sodomy, charges that involved both T and her brother, V. Before trial, as relevant here, defendant moved to suppress

“all evidence, including derivative evidence and statements, obtained through the [housekeeper’s] unlawful and warrantless (a) search of the laundry hamper in his home, (b) seizure of the underwear from the hamper, (c) seizure of the underwear by police and (d) the destruction and testing of the underwear by the Oregon State Crime Lab.”

Following a hearing on that motion, the trial court denied defendant’s motion. As to the initial taking of T’s underwear by defendant’s employees, the court reviewed the evidence at the hearing to determine whether, under the circumstances, either employee had acted “as an instrument or agent of the government,” making their conduct “state action” for purposes of Article I, section 9. It concluded that they had not. The trial court explained that the housekeeper “was not directed [by the DHS employee] to seize [T’s] underwear.” Rather, the employees themselves discussed and then executed a “plan of action.” The court noted that the DHS employee did not encourage or participate in the seizure of the underwear and that, while he “may have had an expectation that the housekeeper would likely obtain possession of the underwear,” he specifically told the housekeeper that he could not ask her to search for or seize it. The court stated that any “circumstantial encouragement” during his conversation with the housekeeper was “insufficient governmental involvement to warrant application of the exclusionary rule,” citing State v. Waterbury, 50 Or App 115, 622 P2d 330, rev den, 290 Or 651 (1981). Accordingly, the trial court ruled that the actions of defendant’s two employees “do not constitute state action.” The trial court also held that the police acquisition of the underwear from the housekeeper was not an unlawful seizure, because that action was supported by “an objectively reasonable belief that the child’s underwear contained evidence of a crime,” and that the testing of the underwear was not an unlawful search, because the information provided to police officers by the housekeeper, together with a visual examination of the underwear, supported the “objectively reasonable belief that * * * the underwear contained evidence of a crime and the testing would provide confirmation of that belief.”

At the trial that followed, the state introduced the test results for the confiscated garments, and a jury convicted defendant on four counts of first-degree sexual abuse involving T; it deadlocked or acquitted on the remaining counts.

Defendant appealed, arguing, in part, that any evidence derived from the seizure and testing of T’s underwear should have been suppressed as the fruit of several unlawful searches or seizures. Central to the issue now on review, defendant argued that, because the actions of those who took T’s underwear and gave it to the police constituted “state action” for purposes of the Oregon and United States constitutions, both the search and the resulting seizure had been unlawful because neither had been based on probable cause, a warrant or, alternatively, some exception to the warrant requirement. 3 In doing so, defendant acknowledged that Oregon courts had yet to clearly articulate a test to determine when a private citizen acts as an instrumentality of the government for search and seizure purposes. Defendant nevertheless asserted that, in his case, the Court of Appeals should adopt the two-part inquiry used by the Ninth Circuit Court of Appeals to analyze such actions under the Fourth Amendment: (1) Did the government know of and acquiesce in the conduct being examined, and (2) did the party performing the search intend to assist law enforcement rather than further the party’s own ends? See, e.g., United States v. Miller, 688 F2d 652, 657 (9th Cir 1982) (stating test). Affirmative answers to both questions, according to the Ninth Circuit, meant that the act, although performed by a non-state actor, nevertheless constituted state action.

In response, the state argued that the issue raised by defendant was controlled by Waterbury, the Court of Appeals decision relied upon by the trial court. In Waterbury, an informant related information concerning a possible marijuana grow to a sheriffs deputy, who pressed him for details concerning its location. The informant had been reluctant to provide that information and, instead, arranged to meet the deputy later without specifically apprising him of the reason for the meeting. When the informant arrived at the designated meeting place, however, he carried with him several plants taken from the grow site that the deputy immediately recognized as marijuana. The deputy and informant went before a judge, where the deputy obtained a search warrant for the property in question. The Court of Appeals subsequently held that the search and seizure performed by the informant had not involved state action:

“The trial court found that ‘there was no clear understanding about what the informant was going to do, but a reasonable expectation could have been that the informant was going to provide further evidence to enable the deputy to obtain a search warrant.’ While it could be said that the informant had some implicit encouragement from the police, absent any request or direct participation by the sheriff, we think such ‘circumstantial encouragement,’ if any, was insufficient official involvement to warrant applying the constraints of the exclusionary rule.”

Id. at 120. Citing that holding, the state argued on appeal that involvement by state actors in procuring T’s underwear had amounted, at most, to “circumstantial encouragement,” meaning that, like the evidence procured by the informant in Waterbury, the evidence procured by the housekeeper was the product of a private search and had been properly admitted at trial.

The Court of Appeals noted that this court had held in State v. Tucker, 330 Or 85, 90, 997 P2d 182 (2000), that Article I, section 9, applies where a private party conducts a search “because of and within the scope of’ a request by a state officer, but it also observed that “[n] either we nor the Supreme Court has explained with precision how much or what kind of state involvement is sufficient to trigger the protection of Article I, section 9.” Sines, 263 Or App at 349. Ultimately, the Court of Appeals agreed with defendant, concluding that the employees’ seizure of the underwear and its delivery to state officials constituted state action. That was so, the court said, for three reasons. First, according to the Court of Appeals, the DHS employee “knew what [the housekeeper] planned to do and that she was likely to do it.” Second, the DHS employee had “communicated with [the housekeeper] about her plans and offered law enforcement support if she conducted the seizure * * * ” Third, the DHS employee had “delayed the safety check to allow [the housekeeper] to accomplish the planned seizure.” Id. at 356. Because no warrant had authorized the resulting search and seizure and the state had not argued that any other exception to the warrant requirement applied, the court held that the trial court erred in denying the motion to suppress and reversed.

DISCUSSION

We begin with a brief overview of first principles. Article I, section 9, of the Oregon Constitution, provides:

“No law shall violate the right of the people to be secure in their persons, houses, papers, and effects, against unreasonable search, or seizure; and no warrant shall issue but upon probable cause, supported by oath, or affirmation, and particularly describing the place to be searched, and the person or thing to be seized.”

That provision protects individuals “against unreasonable search, or seizure,” as well as both possessory and privacy rights in effects. State v. Owens, 302 Or 196, 206, 729 P2d 524 (1986). It is axiomatic, however, that Article I, section 9, applies only to government-conducted or directed searches and seizures, not those of private citizens. Tucker, 330 Or at 89; see State v. Tanner, 304 Or 312, 321, 745 P2d 757 (1987) (privacy interest protected by Article I, section 9, “is an interest against the state,” and “is not an interest against private parties”). That is true even if citizens act unlawfully in obtaining the evidence that later makes its way into the state’s possession. State v. Luman, 347 Or 487, 492, 223 P3d 1041 (2009).

That said, situations can and do arise in which a private citizen’s conduct in pursuing his or her own search and seizure may become so intertwined with the conduct of a state actor that the private citizen’s actions are essentially those of the state and should be subject to constitutional restrictions on state searches and seizures. State v. Tucker is one example. There, a state trooper investigating an accident called the private tow truck operator who had towed the vehicle and asked him to search the vehicle. This court had little difficulty concluding that Article I, section 9, is implicated “if a state officer requests a private person to search a particular place or thing, and if the private person acts because of and within the scope of the state’s request.” 330 Or at 90. See also Wayne R. LaFave, 1 Search and Seizure: A Treatise on the Fourth Amendment § 1.8(b), at 370 (5th ed 2012) (“Quite clearly, a search is not private in nature if it has been ordered or requested by a government official.”). Similarly, although no Article I, section 9, cases are directly on point, Fourth Amendment cases generally hold that when law enforcement officers participate with private individuals in a search or seizure, the action is not a “private search.” Id. at 372 (“A search will also be deemed subject to Fourth Amendment protections if it is a ‘joint endeavor’ involving both a private person and a government official, as where a detective and a victim of a theft together enter a suspect’s apartment to retrieve the stolen goods.” (Footnotes omitted.)).

A more difficult question arises in cases like this one, where a state officer does not instigate or participate directly in a search or seizure, but nevertheless has some communication or involvement related to the search or seizure with the private person before that person engages in the conduct at issue. On review, the parties present us with two somewhat different approaches for determining when a search and seizure conducted by a citizen should be construed as state action and therefore subject to the constitutional protections provided by Article I, section 9, of the Oregon Constitution. 4

The state suggests that common-law agency principles are useful in determining when a private citizen is acting on behalf of or under the authority of the state and therefore subject to constitutional search and seizure limitations. Under the common law, the state notes, an agency relationship “results from the manifestation of consent by one person to another that the other shall act on behalf and subject to his control, and consent by the other so to act.” Vaughn v. First Transit, Inc., 346 Or 128, 135, 206 P3d 181 (2009) (quoting Hampton Tree Farms v. Jewett, 320 Or 599, 617, 892 P2d 883 (1995) (emphasis and internal quotations omitted)). In such a relationship, the principal is vicariously liable for the acts of its agent “only if the principal ‘intended’ or ‘authorized the result []or the manner of performance’ of that act.” Vaughn, 346 Or at 137 (brackets in original) (quoting Restatement (Second) of Agency § 250). In other words, the state continues, for a principal to be held responsible for the acts of its agent, the principal must have conveyed to the agent that he or she is, in fact, authorized to act on the principal’s behalf. Relying on that analysis, the state proposes that we adopt the following rule and apply it to this case: A seizure of property by a private citizen becomes state action for purposes of Article I, section 9, only if the citizen was acting “on the state’s behalf and at the state’s behest,” i.e., that the state “must have directed or controlled the seizure, or must have conveyed to the citizen that he or she was authorized to act on the state’s behalf by asking or actively encouraging the person to do so.”

Defendant urges us instead to adopt the two-part federal test, mentioned previously, viz.: (1) Did the government know of and acquiesce in the conduct being examined, and (2) did the party performing the search intend to assist law enforcement rather than further the party’s own ends? 5 Defendant argues that the Court of Appeals essentially applied that test and correctly held that it was met here.

Defendant’s analysis, like that of the Court of Appeals, begins with what the DHS employee “knew” and “believed” about the housekeeper’s likely actions. He then argues, citing dictionary definitions of the word “acquiesce,” that the state essentially made her conduct its own by not objecting to the potential taking of the underwear or cautioning the housekeeper that doing so would be a crime. The state further supported the housekeeper’s possible actions, defendant asserts, by offering to arrange testing of the underwear if she took it, and by delaying the safety check. Those actions, he argues, so encouraged the housekeeper and defendant’s other employee to take T’s underwear that the state must be viewed as having essentially “caused” that search and seizure.

As discussed above, our cases make clear that Article I, section 9, is a restriction on government searches and seizures, not private ones. Government generally acts, of course, through government employees, but it may also act through nonemployee agents, and searches or seizures by those agents are subject to constitutional protections. See State v. Nagel, 320 Or 24, 29, 880 P2d 451 (1994) (“Under Article I, section 9, a search is an intrusion by a government officer, agent, or employee into the protected privacy interest of an individual.” (Internal quotation marks omitted.)). Constitutional protections against unreasonable searches and seizures would be easily circumvented if the government was not held responsible—and the exclusionary rule not applied—to the actions of private individuals taken on behalf of government. But if a private person cannot be said to be acting on behalf of government in some sense—that is, subject to the government’s control as its agent—it is difficult to see how a search or seizure by that person implicates the rights that Article I, section 9, protects.

We confronted a similar issue in State v. Smith, 310 Or 1, 791 P2d 836 (1990), where a defendant’s cellmate asked him questions that led to incriminating statements, and the defendant sought to suppress those statements at trial on the theory that the questioning had not been preceded by Miranda warnings and therefore violated Article I, section 11. This court rejected that argument, concluding that the cellmate was not acting as a state agent and that Article I, section 11, therefore was not implicated. We noted that the cellmate had initiated the contact with law enforcement officers about the defendant’s statement, and they had told him that if he heard something he wanted to pass on, he could. 310 Or at 14. The officers, however, told the cellmate that he was not required to pass any information along and they made no deals with him. This court concluded that the cellmate was not an agent acting on behalf of the state, because the officers involved were not “directly or indirectly involved to a sufficient extent in initiating, planning, controlling, or supporting the informant’s activities” such that the cellmate could be described as having acted “at the behest” of the state. Id. at 15.

Other courts have used a similar agency analysis in the search and seizure context. In United States v. Jarrett, 338 F3d 339 (4th Cir 2003), for example, the court examined whether a computer hacker who turned over evidence of child pornography to federal authorities had acted on behalf of the government. The court did so by engaging in “a fact-intensive inquiry that is guided by common law agency principles.” Id. at 344. See also Skinner v. Railway Labor Exec. Assn., 489 US 602, 614, 109 S Ct 1402, 103 L Ed 2d 639 (1989) (“Although the Fourth Amendment does not apply to a search or seizure * * * effected by a private party on its own initiative, the Amendment protects against such intrusions if the private party acted as an instrument or agent of the Government.” (Emphasis added.)); State v. Wall, 154 NH 237, 240, 910 A2d 1253 (2006) (applying test of whether “an agency relationship existed between the government and a private individual”). In United States v. Koenig, 856 F2d 843, 847 (7th Cir 1988), the court used common-law agency principles to explain its rejection of the defendant’s proposed “brightline” rule that “knowledge plus acquiescence equals agency.” Although the court noted that the constitutional issue of when a private search may be deemed state action is not necessarily governed by the common-law definition of agency, id. at 847 n 1, it nevertheless quoted and followed the Restatement (Second) of Agency in holding that a FedEx employee had not acted as the government’s agent when he searched a package addressed to the defendant, despite earlier and contemporaneous communications between the employee and governmental officers. Id. at 850.

In our view, too, common-law agency principles can provide substantial assistance in determining when a private citizen’s search or seizure should be considered state action for purposes of Article I, section 9. The state’s formulation of its proposed test—whether a private party acts “on behalf’ and “at the behest” of state officials—does have a conclusory ring, but the factual considerations that lead to those conclusions are helpful, because they look to the objective statements and conduct of the parties to assess whether the conduct of a private individual should be attributed to the government. Common-law agency exists where a principal “manifests assent to another person”—the agent—that the agent “shall act on the principal’s behalf and subject to the principal’s control, and the agent manifests assent or otherwise consents so to act.” Restatement (Third) of Agency § 1.01 (2006). The considerations relevant to the existence of an agent’s actual authority to act on behalf of the principal focus on the “principal’s manifestation to an agent that, as reasonably understood by the agent, expresses the principal’s assent that the agent take action on the principal’s behalf.” Id. § 3.01. Whether the principal “manifests” assent for the agent to act, and whether the agent manifests assent or otherwise agrees so to act, are determined by “written or spoken words or other conduct.” Id. § 1.03. 6

One advantage of the common-law agency analysis is that, in determining whether agency exists, the emphasis is on “manifestations” that can be assessed objectively, in contrast to the Court of Appeals’ test (and defendant’s proposed test), which tend to focus on the subjective motives of the principal and agent, or on what the principal “knew” or “thought” that the agent might do. See Sines, 263 Or App at 353-56 (emphasizing what state officer “knew” and “understood”). Indeed “manifestation,” as used in the Restatement, means “conduct by a person, observable by others, that expresses meaning,” and includes but is not limited to, written or spoken words. Id. § 1.03 comment b. 7

In the criminal search context, those agency concepts examine the conduct of government officials that would communicate to the putative agent that the agent was acting on behalf of the government. One treatise sums up the inquiry as follows:

“[A] defendant must show that the government affirmatively encouraged, initiated, or otherwise participated in the private action. Whether there is sufficient government involvement in a search to transform it into state action is a question of fact that is determined by looking at the totality of the circumstances. Generally speaking, however, courts are likely to find sufficient government involvement where a government official orders, requests, or directs a search. Similarly, even in the absence of an order to search, the use of coercion or affirmative suggestion is usually sufficient to transform an otherwise private search into state action. By contrast, the fact that an officer did not discourage the private party from undertaking the search generally has been found insufficient to bring the search within the scope of the Fourth Amendment.”

Barbara Bergman and Theresa Duncan, 4 Wharton’s Criminal Procedure § 24:20, 24-77 to 78 (14th ed 2009) (footnotes omitted). 8

Defendant urges us to adopt the two-part test described above. Applying the first part of that test, he focuses on the DHS employee and the deputy sheriffs “knowledge of’ and “acquiescence in” the conduct of defendant’s employees, arguing that those facts support his claim that the employees acted as agents of the the government. The federal cases, however, use those terms in a way that does not aid defendant. In United States v. Smythe, 84 F3d 1240, 1242-43 (10th Cir 1996), for example, the court stated, “Knowledge and acquiescence * * * encompass the requirement that the government must also affirmatively encourage, initiate or instigate the private action” (Emphasis added.) Similarly, the court in Jarrett set out the test as quoted, but in applying it observed that “we have required more than mere knowledge and passive acquiescence by the Government before finding an agency relationship.” 338 F3d at 346; see also Koenig, 856 F2d at 847 (rejecting “simple, brightline rule” that “knowledge plus acquiescence equals agency”); United States v. Walther, 652 F2d 788, 792 (9th Cir 1981) (“Mere governmental authorization of a particular type of private search in the absence of more active participation or encouragement is similarly insufficient to require the application of Fourth Amendment standards.”).

Thus, although the first part of defendant’s proposed test is phrased in terms of “knowledge” and “acquiescence,” those terms are not used in their ordinary sense, and courts applying the test have also frequently required affirmative “intiation,” “instigation,” “participation,” or “encouragement.” Indeed, in application, the first part of defendant’s proposed test—although using different words—does not appear to differ substantively from the agency principles we have discussed.

Neither party disputes that the second part of the Miller test—that the private individuals acted with the intent to deter crime and assist law enforcement rather than to “further their own ends”—was met in this case; defendant’s housekeeper so testified. However, that part of the test is problematic in some circumstances, like those here, where defendant’s employees could have intended to protect defendant’s children from sex abuse as well as wanting to assist law enforcement. Such mixed motivations have been noted by several courts. See United States v. Cameron, 699 F3d 621, 638 (1st Cir 2012) (denying motion to suppress and noting that although the government has an interest in combating child pornography, “this does not mean that Yahoo! cannot voluntarily choose to have the same interest”); United States v. Day, 591 F3d 679, 688 (4th Cir 2010) (denying motion to suppress and noting, “Of course, the objective of ‘deterring crime’ is entirely consistent with [the private security guards’] responsibility to protect the tenants and property of the Regency Lake apartment complex, irrespective of any simultaneous goal of assisting law enforcement.”).

We decline to adopt defendant’s proposed test. The first part of that test purports to rely on whether the government “knew of and acquiesced in” the private conduct. See Miller, 688 F2d at 657. But those considerations tell us little about particular government actions that would communicate to a private person any authority or permission to act as an agent or instrument of the government. They also ask the factfinder to consider subjective mental states, rather than statements and conduct that can be assessed objectively. As discussed above, even the federal courts that use defendant’s proposed test require “active participation or encouragement,” see, e.g., Walther, 652 F2d at 792 (so stating), and conclude that “knowledge and acquiescence” without more is insufficient to establish state action. Koenig, 856 F2d at 847. In our view, a test that, in application, uses words in ways that are at odds with the ordinary meaning of those words is of limited utility. It is, in any event, more difficult to apply a test that relies on an assessment of what persons “knew” and what they “acquisesced” in, than it is to apply a test that examines statements and affirmative conduct for manifestations of an intent to confer authority. The state’s proposed use of common-law agency principles to determine whether, in particular circumstances, a private actor should be considered a state agent for purposes of Article I, section 9, is, in contrast, clearer and more easily applied. 9

With that background, we return to the essentially undisputed facts here, focusing on the statements and objective conduct of the individuals involved. Defendant’s housekeeper anonymously and on her own initiative called DHS to report the suspected abuse. The DHS employee with whom she spoke did not direct the housekeeper to search the house or to seize evidence, saying, instead, that the decision was “up to her.” The housekeeper also raised the issue of obtaining evidence and the possibility of underwear as evidence; she testified that getting the underwear “was my idea.” In response to her question about what they could determine from underwear, the DHS employee said there was a lab locally and they could “probably tell a lot.” The DHS employee gave the housekeeper his direct telephone number. He also contacted the sheriffs office to discuss the call, and he and his contact at the sheriffs office agreed to conduct the routine follow-up safety check within five days, rather than within the usual 24-hour period from the initial report. Neither the 24-hour safety check protocol or the modification that DHS and the sheriffs office agreed to was communicated to the housekeeper. The housekeeper subsequently called defendant’s other employee, and that employee took the underwear from the laundry room in defendant’s house the following day, while T and her brother were at school, and then gave it to the housekeeper. The housekeeper turned the underwear over to state officials the next day.

The question is whether those facts, and in particular the conduct and statements of the state officials, demonstrate that those officials communicated to the housekeeper (and defendant’s other employee) that they were authorized to act as agents of the state. The DHS employee did not direct or request the housekeeper to take the underwear. The idea of taking evidence from the house, and of taking underwear in particular, came from the housekeeper. Although defendant and the Court of Appeals focus on what DHS employees “knew” or “thought” or “understood” the housekeeper might do, the common-law agency analysis that we outlined above looks first to objective manifestations by the principal to the agent that the agent should or may act on behalf of the principal. That is consistent with the federal courts’ emphasis on affirmative government conduct vis á vis the private person. See Koening, 856 F2d at 850 (“It is only by the exercise of some form of control that the actions of one may be attributed to another. Restatement (Second) of Agency § 14 (1958). Mere knowledge of another’s independent action does not produce vicarious responsibility absent some manifestation of consent and the ability to control.”); Smythe, 84 F3d at 1242-43 (to make private conduct state action, government agent must “affirmatively encourage, initiate, or instigate the private action”). There was little, if any, such affirmative encouragement, initiation, or instigation here.

Defendant nevertheless argues that the state encouraged and “supported” the private search in several ways. First, he argues that DHS delayed its usual 24-hour safety check to allow the housekeeper sufficient time to conduct the search. That unilateral action by the state, however, was never communicated to the housekeeper, and could not have affected her or her decision to act. Although the delay suggests that state officials hoped that the housekeeper’s actions would assist them in investigating the alleged abuse, it is irrelevant to whether the state consented to have her act on the state’s behalf. Moreover, on the facts here, there is no showing that the delay had an effect on the search and seizure in any event, because the search and seizure occurred within 24 hours of the housekeeper’s initial call.

Defendant next asserts that the DHS employee’s communications with the housekeeper, including discussing testing the underwear and giving her his direct phone number, demonstrate a level of indirect support of the housekeeper’s conduct sufficient to make her an agent of the state. We disagree. The fact that the DHS employee truthfully answered the anonymous caller’s unsolicited question about what they could determine from particular evidence and provided his direct phone number do not rise to the level of state instigation or direction to make the caller’s subsequent search state action. 10

Finally, defendant makes two other, related arguments for suppressing the results of the search and seizure here. First, he asserts that the DHS employee indirectly encouraged the housekeeper by failing to warn her that taking property from defendant’s house would constitute theft. Second, he contends that the evidence should be suppressed because defendants’ employees stole it from his house.

This court addressed the latter argument in State v. Luman, 347 Or 487, where, after reiterating that Article I, section 9, does not apply to private searches, the court stated that that “principle applies even if the private parties acted unlawfully in conducting the search and seizure that ultimately led to police possession of the evidence.” The court distinguished the issue of the criminality of the private conduct from the issue of whether the actions of the private parties could be attributed to the state. Even if the private party had stolen the evidence in question and given it to the sheriffs office, “that fact would not somehow turn that conduct into state action or render the sheriffs office’s later possession of the videotape unlawful.” Id. at 493.

The former argument—that the DHS employee’s failure to warn the housekeeper that taking the underwear was a crime or otherwise to dissuade her from stealing from defendant was sufficient government support to make the private conduct state action—also is not well taken. The ultimate issue is whether the housekeeper acted on behalf of the state, which we determine by considering whether the state’s conduct would have conveyed to her that she was so authorized. Failing to warn or advise the housekeeper against engaging in a potentially criminal act is not such conduct. As we previously emphasized, “the fact that an officer did not discourage the private party from undertaking the search generally has been found insufficient to bring the search within the scope of the Fourth Amendment.” Bergman and Duncan, 4 Wharton’s Criminal Procedure § 24:20 at 24-78; see also Jarrett, 338 F3d at 347 (“that the government did not actively discourage Unknownuser from engaging in illict hacking does not transform Unknownuser into a Government agent”; government had no special obligation to discourage illegal hacking by private party); United States v. Souza, 223 F3d 1197, 1202 (10th Cir 2000) (“The police are under no duty to discourage private citizens from conducting searches of their own volition.”).

We conclude, based on the facts explicitly and implicitly found by the trial court, that the actions of defendant’s employees in searching for and seizing the underwear constituted private conduct and therefore did not violate Article I, section 9. We acknowledge that this is a close case. Contacts between private individuals and state officers before a private search always require careful examination to determine whether, given all the circumstances, the state officers provided such affirmative encouragement and authorization to the private individuals so as to render them agents of the state. 11 In this case, for the reasons described above, we hold that they did not. Accordingly, we reverse the Court of Appeals decision and remand to that court for consideration of other assignments of error that it did not address.

The decision of the Court of Appeals is reversed, and the case is remanded to the Court of Apeáis for further proceedings.

1 The Court of Appeals considered the validity of the search and seizure here only under Article I, section 9, and did not discuss or resolve any claim by defendant under the Fourth Amendment. Similarly, the state’s petition for review was based solely on Article I, section 9. The briefs of both parties refer to Fourth Amendment cases only in the context of their competing arguments on the proper interpretation of Article I, section 9, and neither party develops any independent argument as to the validity or invalidity of the search here under the Fourth Amendment. Accordingly, we express no opinion on that issue, nor do we express any opinion as to whether that issue was properly raised, preserved, or developed below, and leave those questions, in the first instance, to the Court of Appeals on remand. Additionally, because the Court of Appeals reversed and remanded on the legality of the initial taking of the underwear, it did not address defendant’s other assignments of error. Depending on the Court of Appeals’ rulings on remand, it may be appropriate for that court to consider defendant’s other assignments of error at that time.

2 On cross-examination, during the hearing on the motion to suppress, counsel for defendant and the DHS worker engaged in the following exchange:

“[Defense counsel:] And so you offered to her the services of the Oregon State Police Crime Lab?

“[DHS:] I told her that I could hook her up with—I told her that I could hook her up with people who could make that happen.

“[Defense counsel:] So if she stole underwear, you could hook her up with people who could examine it at the Oregon State Crime Lab. You told her that in the first call?

“[DHS:] I told her that those services were available in—right here within the community.

“ [Defense counsel:] And that you would hook her up with them?

“[DHS:] I could—I would hook her up with a worker and with law enforcement who could make that happen.”

3 Defendant also contended, as he had at trial, that (1) acceptance of the underwear by the deputy sheriff—which he knew had been taken without permission from defendant’s house—constituted a second unlawful seizure, and (2) testing the underwear constituted a second unlawful search. Because it disposed of this case on defendant’s first argument, the Court of Appeals did not reach those issues.

4 In discussing the tests suggested by the parties, it is worth noting that each party also argues that it would prevail under the other party’s test, as well as the test that it proposes. Although we find the state’s proposed test more useful for the reasons discussed below, it appears that most of the cases cited by the parties would have reached the same result under either test.

5 Although both parties suggest that this two-part test is generally used by the federal courts, the landscape is actually more diverse. For example, the First Circuit considers three factors: (1) the extent of the government’s role in instigating or participating in the search; (2) the government intent and the degree of control it exercises over the search and the private party; and (3) the extent to which the private party aims primarily to help the government or to serve its own interests. United States v. Cameron, 699 F3d 621, 637 (1st Cir 2012), cert den, 133 S Ct 1845 (2013). The Fourth Circuit considers whether the private citizen was an instrument or agent of the government, given the government’s participation or affirmative encouragement. United States v. Richardson, 607 F3d 357, 364 (4th Cir 2010). See also United States v. Smythe, 84 F3d 1240, 1243 (10th Cir 1996) (for private search to be state action, government must “affirmatively encourage, initiate or instigate the private action”). Moreover, as we discuss later in the text, some of the words in defendant’s proposed test are used in a different sense than their ordinary dictionary definitions.

6 Determining whether a principal has assented for another to act as its agent will often depend on the context in which the conduct of the parties occurs and other aspects of the relationship between the principal and agent. See Restatement (Third) of Agency § 1.03 comment e (discussing context in which assent is manifested) and comment c (discussing assent by organization for person to act as its agent).

7 The reference to conduct “observable by others” simply highlights one kind of communication by a principal that can confer authority to act on an agent. It need not be “observable” by third persons dealing with the agent, and in many circumstances involving police—such as the use of an informant—the police and the agent will seek to prevent the person dealing with the agent from knowing the agent’s authority. In this context, the communication that establishes the agency relationship is that between the principal and the agent, and the comment quoted in the text is a reminder that the conduct must be such that a reasonable observer—such as the agent or a later factfinder—would understand the conduct to be intended by the principal to assent to the creation of an agency relationship. It is important to remember that, in the kind of agency at issue here, the agent has “actual authority” to act “on behalf of the principal, consistent with the principal’s manifestations to the agent that the principal wishes the agent so to act ’’Restatement (Third) of Agency § 2.01 (2006) (describing actual authority), rather than “apparent authority,” which can exist when “a third party reasonably believes the [agent] has authority to act on behalf of the principal.” Id. § 2.03 (describing apparent authority).

8 We quote this passage for its summary of the case law. We do not necessarily agree with the comment in the second sentence that whether a particular search by a private citizen is state action is a “question of fact.” The determination will depend upon the facts, but the conclusion ultimately is a legal one.

9 We also find the second part of defendant’s proposed test to be unhelpful because of the “mixed motivations” issue identified by several federal courts and discussed above. Of course, the circumstances of the private actor’s conduct, including job responsibilities and any relationship with the defendant, may well be relevant facts in deciding whether that conduct constituted state action.

10 The circumstances here are substantially different from State v. Lowry, 37 Or App 641, 588 P2d 623 (1978), where the court concluded that incriminating statements made by the defendant to an informant without Miranda warnings should have been suppressed because the informant was acting as an agent of the state. The informant—a person well known to authorities as a “dedicated and accomplished ‘stool pigeon’” who had “survived for years by (and in spite of) providing information to various police and penal authorities,” id. at 643 — obtained admissions from the defendant, his cellmate, and then negotiated a deal with prison officials pursuant to which he recounted the admissions to the officials, who then paid him $50 and, at his request, transferred him to a federal corrections facility the next day. Although the Court of Appeals here relied heavily on Lowry, see Sines, 263 Or App at 349-53, neither party cited that case in their briefs before the Court of Appeals or this court, and we do not find it helpful.

11 Because we conclude that the state officials did not authorize the private individuals involved here to act as agents of the state, we do not consider the second step in the common-law agency analysis, viz., whether the “agent manifest[ed] or otherwise consent[ed] so to act.” Restatement (Third) of Agency § 1.01.