321 Ga. 137 FINAL COPY
S24A1085. JONES v. THE STATE.
PINSON, Justice.
James Christopher Jones was charged with murder after the police identified her using the location history from her cell phone. The police got that location history through search warrants that authorized the police to obtain from Google an anonymized list of devices that reported their locations within 100 meters of the vic- tim’s home during the four hours when the murder happened — a process known as “geofencing” — and identifying information tied to the subset of devices that were relevant to the investigation. Before trial, Jones moved to suppress the evidence from the geofence war- rants, arguing that the warrants violated the Fourth Amendment to the United States Constitution because they were not supported by probable cause and failed to satisfy that Amendment’s particularity requirement. The trial court denied the motion, and we granted Jones’s application for an interlocutory appeal.
2
We now affirm. The warrant applications explained among other things that the suspect was caught on video using a cell phone near the victim’s home, that many cell phones generate Google loca- tion history data, and, later, that the movements of a specific cell phone “matched up” with what was known of the suspect’s move- ments. That information, together with the reasonable inferences and common sense that a reviewing magistrate may draw on in as- sessing probable cause, gave the magistrate here a substantial basis for concluding that accessing the location history and identifying in- formation sought from Google had a fair probability of helping the police identify the unknown murder suspect in the video. And the warrants satisfied the particularity requirement because they gave the police specific guidance as to what information they were au- thorized to access — a list of anonymized Google IDs and location history data from devices reporting their locations within 100 me- ters of the victim’s home during a given time frame, and then iden- tifying information tied to one of those Google IDs — and avoided the kind of unfettered discretion that would pose a particularity
3
problem.
1. Background (a) Jones was charged with malice murder and other crimes in connection with the stabbing death of Kay Conway Thomasson. Be- fore trial, Jones moved to suppress cell-phone-based evidence the police had obtained under two search warrants. The evidence intro- duced at the hearing on the motion to suppress showed the follow- ing.
On June 27, 2018, Detective J. T. Williams responded to a re-
port that a woman, Thomasson, had been found dead at a residential address in Sandy Springs. At the scene, Williams spoke to the vic- tim’s daughter and learned that the home was equipped with video surveillance, including one camera pointing down the driveway and another aimed at the back door. After getting a search warrant for the home, Williams reviewed the camera footage and saw a suspect arriving at the victim’s driveway at 1:10 a.m. on June 27, and the same suspect leaving in the victim’s car at 5:03 a.m.
Several months later, after the investigation had come to a
4
standstill, Williams applied for the first search warrant at issue in this appeal. The warrant was directed at Google, and it asked for two things: (1) “GPS, Wi-Fi, or Bluetooth sourced location history data generated from devices that reported” within 100 meters of Thomasson’s home between 1:00 a.m. and 5:00 a.m. on June 27, 2018; and (2) “identifying information for Google Accounts associ- ated with the responsive location history data.” The warrant de- scribed a three-step process for collecting that information. In step one, Google would provide an anonymized list of all devices that were detected within the original geographic and temporal parame- ters — 100 meters from the victim’s home, from 1:00 a.m. to 5:00 a.m. In step two, officers could request additional location infor- mation outside the initial 100-meter radius for any of the devices from step one that officers thought could be relevant to the investi- gation. And in step three, after further review, the police could ask for identifying information about the users associated with relevant devices.
5
Williams explained at the hearing that this location infor-
mation was routinely collected by Google from its users. According to Williams, Google tracks the physical location of everyone who uses a Google application like Gmail or Google Maps, or who uses a device that runs on Google’s Android operating system, through the device’s cellular, Wi-Fi, and GPS capabilities. Google also assigns a number, called a “Google ID,” to anyone who uses a Google applica- tion or an Android-running device. So in layman’s terms, the war- rant asked for the Google IDs, or account numbers, associated with any devices that came within 100 meters of the victim’s home during the four hours when the suspect was known to be there. The geo- graphic area targeted by this type of warrant — here, the 100-meter radius from the victim’s home — is called a “geofence,” and so this kind of search warrant is sometimes called a geofence warrant.
Google responded to the geofence warrant and provided the re-
quested information. In its raw form, Google’s data were a list of every time a Google ID was detected within the geofence. For each hit, the data showed the Google ID, the time stamp, the latitude and
6
longitude coordinates, and the means of detection (GPS, Wi-Fi, or cellular).
When police reviewed the data, two Google IDs “sparked [their]
interest.” One was the victim’s phone. The other was device 1258821290, which police found to be “very suspicious.” Device 290 was shown to be in or near the victim’s home during the time of the killing, moving around the side of the home where the suspect was seen in the surveillance video.
Police then went to step two of the search warrant and asked for additional location data from Google about the victim’s phone and device 290. They received all the location information for those two phones, including locations outside the geofence, for an hour be- fore and an hour after the original window of 1:00 a.m. to 5:00 a.m. The data from the extended timeframe showed that the two devices — the victim’s phone and device 290 — both left the area of the vic- tim’s home after the killing. The phones traveled together to a nearby business, where the victim’s phone remained until it went
7
dead 12 hours later. (The victim’s phone was never recovered.) De- vice 290, however, continued on to an apartment complex in Cham- blee.
Williams then explained how the police used the location data they received from Google together with other evidence they col- lected. Because the victim’s security cameras showed that the sus- pect left the scene in the victim’s car, the police were able to track the route the victim’s car took using license-plate-reading cameras in the area. Comparing the car’s route with the location data for the victim’s phone and device 290, they determined that the car and the devices were traveling together. And police later found the victim’s car at the same apartment complex where device 290 ended up. From this information, police surmised that device 290 belonged to the suspect.
That left identification of the suspect. Although step three of the search warrant authorized the police to ask for identifying infor- mation about relevant Google IDs, Detective Williams did not ask for that identifying information under the original warrant. Instead,
8
he completed a second, updated warrant application, and obtained a second search warrant that authorized police to obtain identifying information for the Google user associated with device 290.
Google provided police with that identifying information: a Gmail address and a cell phone number. Police found that the phone number was on record with the Department of Public Safety and was associated with the recently renewed driver’s license of Jones. Jones was thus identified as the suspect, and she was arrested and charged with the victim’s death.
(b) Jones moved to suppress the evidence obtained under the search warrants, arguing that the warrants violated the Fourth Amendment to the United States Constitution because they were not supported by probable cause and did not satisfy the particularity requirement. After a hearing, the trial court denied the motion, con- cluding that the search warrants were supported by probable cause and were sufficiently particular. The court explained that the affi- davit supporting the first warrant established that the suspect used a cell phone around the victim’s home within the timeframe when
9
the victim was killed, and that the affidavit for the second warrant, which was drafted after police obtained the anonymized location his- tory data from the first warrant, established that device 290 was at the victim’s home at the time of the alleged murder and left the house when the suspect did. The trial court concluded that those af- fidavits together authorized a practical, commonsense determina- tion that the location history data, and the later identifying infor- mation about device 290, had a fair probability of providing evidence of a crime. As to particularity, the trial court noted that the first affidavit referred to the specific crime, date, time, and location at issue, and the court found that the 100-meter radius was reasonable to help officers determine whether the suspect captured on video was moving around the area, entering other homes, or meeting other individuals. The court further observed that the second warrant identified the specific Google ID for which it sought identifying in- formation.
The court granted a certificate of immediate review, and we granted Jones’s interlocutory application to appeal.
10
2. Analysis The Fourth Amendment to the United States Constitution was “the founding generation’s response to the reviled ‘general warrants’ and ‘writs of assistance’ of the colonial era, which allowed British officers to rummage through homes in an unrestrained search for evidence of criminal activity.” Riley v. California, 573 U.S. 373, 403 (IV) (134 SCt 2473, 189 LE2d 430) (2014). The Amendment guards against such practices (and their modern equivalents) in a couple of ways. It guarantees a general “right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures.” U.S. Const. Amend. IV. And it protects that right with specific requirements for warrants, which authorize the government to carry out searches and seizures. See id. A warrant to carry out a search under the Fourth Amendment is properly issued if it (1) is supported by “probable cause” and (2) “particularly de- scrib[es] the place to be searched, and the persons or things to be seized.” Id.
Jones claims that neither of these requirements is met by the
11
warrants here. To summarize once more the content of those war- rants: The first warrant authorized the police to access “location his- tory data” generated from devices that reported a location within a radius of 100 meters around the victim’s home between 1:00 a.m. and 5:00 a.m. on the night she was killed. The warrant further al- lowed the police to access location history data outside that geofence if needed to determine whether a device was “relevant” to the inves- tigation. And it allowed the police to obtain identifying information for the Google Account associated with any devices determined to be relevant to the investigation. But as to that third step of obtaining identifying information, the police did not rely on the first warrant, but instead sought and secured the second warrant. The second war- rant authorized the police to obtain identifying information tied to two cell phones that reported location history data near the victim’s home, one of which matched the movements of the suspect that the police had seen on video surveillance. In short, these search war- rants authorized the police to access the anonymized location history reported from devices within a specified area and timeframe, and
12
then to obtain the identifying information tied to specified devices within that “geofence.”
We address Jones’s claims that these warrants violated the probable cause and particularity requirements in turn.1 (a) Probable Cause (i) The standard for assessing whether a warrant is supported
1 In granting Jones’s application for an interlocutory appeal, we also asked the parties to address the threshold question whether the government conducts a search under the Fourth Amendment when it accesses a person’s cell-phone-GPS-based location history. Cf. Carpenter v. United States, 585 U.S. 296, 304 (II) (A) (138 SCt 2206, 201 LE2d 507) (2018). Compare United States v. Chatrie, 107 F4th 319, 332 (II) (B) (4th Cir. 2024) (concluding that obtaining geofenced location history was not a search in large part because the defendant “knowingly and voluntarily chose to allow Google to collect and store his location information”) with United States v. Smith, 110 F4th 817, 836 (III) (A) (5th Cir. 2024) (concluding that obtaining geofenced location history was a search because, “[g]iven the intrusiveness and ubiquity of Location History data,” the defendants had “a ‘reasonable expectation of privacy’ in their respective data”). Having reviewed the record, however, it is now apparent that both the trial court and the parties, including the State, took as a given that a search under the Fourth Amendment was carried out here. Because that issue was neither raised nor ruled on below, there is not a sufficient record on which we could properly address and answer this threshold question in this posture. And in light of our disposition of Jones’s claims in this decision (see below), we need not do so. Thus, like the parties and the trial court below, we take as a given (but do not decide) that the police carried out a search under the Fourth Amendment when they accessed Jones’s location history here.
We also asked the parties to address these questions under the Georgia Constitution, but as the State rightly points out, Jones did not raise a separate claim under the Georgia Constitution below, and the trial court thus did not rule on one, so no such claim is properly before us now. We therefore assess Jones’s claims under the Fourth Amendment only.
13
by probable cause balances core Fourth Amendment concerns — “safeguard[ing] citizens from rash and unreasonable interferences with privacy and from unfounded charges of crime” — with the need to “give fair leeway for enforcing the law in the community’s protec- tion.” Brinegar v. United States, 338 U.S. 160, 176 (II) (69 SCt 1302, 93 LE 1879) (1949). That balance requires a “practical, nontechnical conception” of the rule. Id. So probable cause does not require proof “beyond a reasonable doubt” or even by a “preponderance of the evi- dence” that a search will turn up evidence of a crime, but only the “‘fair probability’ on which ‘reasonable and prudent (people,) not le- gal technicians, act.’” Florida v. Harris, 568 U.S. 237, 243-244 (II) (133 SCt 1050, 185 LE2d 61) (2013) (quoting Illinois v. Gates, 462 U.S. 213, 231, 235, 238 (III) (103 SCt 2317, 76 LE2d 527) (1983)). See also Pugh v. State, 318 Ga. 706, 713 (2) (b) (899 SE2d 653) (2024). In keeping with this commonsense standard, a magistrate making a probable-cause determination not only takes into account all of the evidence, see Pugh, 318 Ga. at 716 (2) (b), but also may
14
draw reasonable inferences from the information in the warrant af- fidavit, see id., consider the “factual and practical considerations of everyday life,” Copeland v. State, 314 Ga. 44, 49 (3) (875 SE2d 636) (2022) (citation and punctuation omitted), and depend on “common- sense conclusions about human behavior,” Pugh, 318 Ga. at 714 (2) (b) (quoting Gates, 462 U.S. at 231 (III)).
Once made, the magistrate’s decision about probable cause is entitled to “great deference” from reviewing courts — both the trial court ruling on a motion to suppress and any appellate court review- ing that ruling on appeal. Gates, 462 U.S. at 236 (III) (citation and punctuation omitted). So when we review such a decision, our job is simply to ensure that the magistrate had a “substantial basis” for concluding that probable cause supported the warrant, id., and if we have doubts, we are to resolve them in favor of upholding the war- rant, see Pugh, 318 Ga. at 714 (2) (b).
Applying this deferential standard of review, the magistrate here had a substantial basis for concluding that the warrants here were supported by probable cause.
15
Start with the first search warrant. Here is what the magis-
trate had in front of him: The affidavit submitted with the applica- tion for the first warrant explained that the victim was found dead in her home. The affidavit then described in some detail how, on the night of the killing, a suspect was seen on video surveillance pulling into the victim’s driveway at 1:10 a.m., walking up the driveway, circling the house, entering the home with a flashlight and baseball bat, leaving the home at 4:43 a.m. and finally driving away at 5:03 a.m. And important here, the affidavit told the magistrate that the video showed the suspect talking on a cell phone, and it further explained that “[n]early every cellular phone using the Android op- erating system has an associated Google account,” and that Google collects and retains location data from Android devices when the user has enabled Google location services.
This information, together with reasonable inferences and common sense, gave the magistrate a substantial basis for finding probable cause to issue the first warrant. Surveillance video showed that the person suspected of murdering the victim at her home that
16
night was carrying a cell phone that was likely powered on around the time of the murder. And given the ubiquity of cell phones capable of reporting location information to Google — supported by the affi- davit and common knowledge, see, e.g., Carpenter v. United States, 585 U.S. 296, 300 (I) (A) (138 SCt 2206, 201 LE2d 507) (2018) (noting in 2018 that “[t]here are 396 million cell phone service accounts in the United States—for a Nation of 326 million people”) — the mag- istrate reasonably could have inferred that the suspect’s cell phone likely was reporting its location to Google at the time. All of this adds up to a fair probability that the requested search would turn up evidence that would help the police identify the killer: Allowing the police to access a list of devices reporting their location to Google close in time and place to where the murder suspect was shown on video surveillance could link a device to the suspect’s movements shown on that video. Allowing the police to access further anony- mized location history data outside that geofence could help the po- lice narrow down which device was likely to be the one held by the suspect in the video. And obtaining the identifying information tied
17
to such a device would likely identify the primary suspect in the murder. In short, the magistrate had a substantial basis for his de- termination that the search authorized by the warrant had a fair probability of turning up evidence that would help them solve a mur- der.
The second warrant was also supported by probable cause. The application for that warrant, which the police sought after they had identified device 290 as likely belonging to the suspect, explained that the location history of that device “match[ed] up” with the movements of the suspect as seen on security and traffic cameras. Device 290 arrived at the victim’s home when the suspect did, left when the suspect did, and reported its location at two area busi- nesses when the suspect was seen at those businesses. The route of device 290 also matched that of the victim’s phone. Based on that “match[ing] up,” the affidavit attested that device 290 was the sus- pect’s device, and that any identifying information about the owner of the device “will greatly assist in identifying who the murder sus- pect is.” That affidavit gave the magistrate a substantial basis for
18
concluding that learning the identifying information tied to device 290 would have a “fair probability” of yielding further evidence likely to identify the person who committed the murder. See Pugh, 318 Ga. at 714 (2) (b).
(ii) Jones does not seriously contest that the access to location history data these warrants authorized had a fair probability of turning up evidence helpful to the murder investigation, including the identity of the suspect who had been caught on video. Instead, she contends that the warrants were “overbroad” — that the scope of the search they authorized exceeded whatever search might have been supported by probable cause. See Pugh, 318 Ga. at 718 (2) (c). See also Wayne R. LaFave, 2 Search & Seizure § 4.5(a) (6th ed.) (“While a description of the place to be searched must be particular enough to guide the executing officer as to where the warrant is to be executed, there is in addition the requirement that the descrip- tion be sufficiently narrow in the sense of not outrunning the prob- able cause showing.”). Jones makes two arguments that fit under the “overbreadth” label: one about the nature of the search of
19
Google’s database itself, and another specific to the scope of the geofence.
Jones’s database-search argument is that these warrants were overbroad because Google’s initial search for the requested infor- mation would “touch” the data of all Google users. That is, when Google searches its database in response to a geofence warrant, Jones asserts that Google effectively “looks at” every Google account in the world to see whether the account ID fits the temporal and geographic parameters of the geofence request. In Jones’s view, that means the warrants authorized a “search” of every Google user’s lo- cation history data.
Color us skeptical. Database searches are a routine part of criminal investigations. They can be used to look up a physical ad- dress, see Jones v. State, 314 Ga. 605, 607 (1) (878 SE2d 505) (2022), a phone number, see Young v. State, 309 Ga. 529, 530 (1) (847 SE2d 347) (2020), a license plate number, see Felton v. State, 304 Ga. 565, 566 (819 SE2d 461) (2018), or a DNA profile, see Bharadia v. State, 297 Ga. 567, 568-569 (1) (774 SE2d 90) (2015), among other things.
20
Speaking generally, when an investigator runs one of those searches, he types a set of search terms or criteria into a search box, and then he sees the results, if any, that the query returns. He does not view every license plate or phone number in the system. (Nor, we presume, would he want to: he entered the search criteria to find the information he was looking for, not anything else.) As a result, the privacy of the information that is not surfaced by the search query is preserved from the investigator. And if that’s so, it is hard to see how one could call what the investigator did a Fourth Amend- ment “search” of the information in the database that was not re- turned as a search result. See Camara v. Mun. Ct. of City and County of San Francisco, 387 U.S. 523, 528 (I) (87 SCt 1727, 18 LE2d 930) (1967) (“The basic purpose of [the Fourth] Amendment, as rec- ognized in countless decisions of this Court, is to safeguard the pri- vacy and security of individuals against arbitrary invasions by gov- ernmental officials.”); United States v. Sparks, 806 F3d 1323, 1336 (III) (B) (2) (11th Cir. 2015) (holding that a detective exceeded the scope of a private search to the extent that he viewed a second video
21
stored within the same digital album that contained the video viewed by the private citizen, because the private search had not “expose[d] every part of the information contained in the cell phone”), overruled on other grounds by United States v. Ross, 963 F3d 1056 (11th Cir. 2020). See also Orin S. Kerr, The Digital Fourth Amendment 78-80 (2025) (noting that treating the scope of a com- puter search as the information exposed by the search, not the whole device, tracks the rule for home searches, where “each new exposure of something in the home is treated as a distinct search”).
But we need not resolve this argument definitively because Jones lacks the Fourth Amendment standing to make it. Fourth Amendment standing is a “useful shorthand for capturing the idea that a person must have a cognizable Fourth Amendment interest in the place searched before seeking relief for an unconstitutional search.” Byrd v. United States, 584 U.S. 395, 410 (IV) (138 SCt 1518, 200 LE2d 805) (2018). At most, Jones has a Fourth Amendment in- terest in her own location history data, not the data of any of Google’s other users she claims were searched. As a result, she lacks
22
Fourth Amendment standing to challenge the warrant on the basis that it authorizes a search of other users’ data. See United States v. Davis, 109 F4th 1320, 1329-1330 (III) (A) (2) (11th Cir. 2024) (de- fendant lacked Fourth Amendment standing to challenge collection of Google location history data associated with someone else’s phone and Google account); Presley v. United States, 895 F3d 1284, 1289 (III) (A) (11th Cir. 2018) (explaining that “[p]laintiffs [who] contest only others’ privacy rights . . . ordinarily lack Fourth Amendment standing”).
As for her geofence-based argument, Jones points out that the first warrant authorized the police to access a list of devices report- ing their locations within a 100-meter radius of the victim’s home, which reached beyond the victim’s home and yard into other homes, yards, and roads. In her view, that is like Ybarra v. Illinois, 444 U.S. 85 (100 SCt 338, 62 LE2d 238) (1979), where the Supreme Court held that police could not search every patron of a bar based on prob- able cause to believe that the bartender — and only the bartender — was selling illegal drugs. Jones emphasizes the point from Ybarra
23
that the probable cause requirement “cannot be undercut or avoided by simply pointing to the fact that coincidentally there exists proba- ble cause to search or seize another or to search the premises where the person may happen to be.” Id. at 91 (II).
This argument reflects a misunderstanding of the search here.
The first warrant did not authorize the police to search a person, like Ybarra, who was known not to be the suspect and was thus out- side the scope of the probable cause for a search. Indeed, it did not authorize a search of any person, or even any place. Instead, the warrant authorized the police to access information: an anonymized list of Google IDs and location history data associated with devices reporting their locations near the victim’s home around the time of the murder (and later, identifying information tied to relevant Google IDs).2 Such access was permissibly authorized by this war- rant because reviewing location history data in that window of time
2 To the extent Jones’s argument is that the police violated the Fourth
Amendment rights of the other users whose anonymized location history data was accessed (i.e., those other than Jones), she lacks Fourth Amendment standing to assert the rights of those other users. See, e.g., Davis, 109 F4th at 1329-1330 (III) (A) (2).
24
and space was likely enough to help police pinpoint the cell phone already known to have been used by the suspect in the same area at the same time and, in turn, to identify the suspect.3 The possibility that the access the warrant authorized could also allow the police to view anonymized data not associated with the suspect does not af- fect the probable-cause assessment, which turns on the likelihood that the access granted by the warrant could lead to the suspect’s identity. Nor does that possibility on its own make this warrant overbroad — just as a search warrant for a person’s papers is not overbroad because not all of the documents examined will be evi- dence of a crime. See Andresen v. Maryland, 427 U.S. 463, 482 (III) n.11 (96 SCt 2737, 49 LE2d 627) (1976) (recognizing that when a search warrant authorizes the search and seizure of a person’s pa- pers, “it is certain that some innocuous documents will be examined, at least cursorily, in order to determine whether they are, in fact,
3 We do not decide here whether a geofence search warrant could be sup-
ported by probable cause even without a showing that a cell phone or other device was used within the temporal and spatial parameters of the geofence.
25
among those papers authorized to be seized”); United States v. Ul- bricht, 858 F3d 71, 100 (I) (B) (2d Cir. 2017) (“[T]raditional searches for paper records, like searches for electronic records, have always entailed the exposure of records that are not the objects of the search to at least superficial examination in order to identify and seize those records that are.”), abrogated on other grounds by Carpenter, 585 U.S. at 309-310 (III).4 It is possible that overbreadth could be a problem for a similar warrant that authorized a more expansive geofence request, if it gave police access to enough location history data not reasonably calculated to lead them to the device held by the unknown suspect or that suspect’s identity. Cf. United States v.
4 The access authorized by the warrant here is more like the search up-
held in Zurcher v. Stanford Daily, 436 U.S. 547 (98 SCt 1970, 56 LE2d 525) (1978). There, the Supreme Court upheld the constitutionality of a warrant to search a newspaper’s records for photos of an assault, when police had reason to believe that a photographer from the newspaper had photographed the crime and that the photos would help police identify the suspects. It did not matter that police did not know who the suspects were, or that the photos would inevitably show people who had nothing to do with the assault, or that the newspaper itself was not suspected of any crime. The search warrant was constitutional because “there [was] probable cause to believe that fruits, instrumentalities , or evidence of a crime [would] be found.” Id. at 554 (II). That is, there was a fair probability that the newspaper’s photos would show the people who were at the scene of the crime. Just so here.
26
McCall, 84 F4th 1317, 1328 (III) (B) (11th Cir. 2023) (assuming a search warrant that authorized a search of the entirety of the de- fendant’s iCloud account was overly broad). But that does not de- scribe the geofence request here, which was appropriately tailored to the evidence in this case.
(b) Particularity In addition to being supported by probable cause, a search war-
rant must “particularly describ[e] the place to be searched, and the persons or things to be seized.” U.S. Const. Amend. IV. The particu- larity requirement safeguards against general searches — the “ ‘gen- eral, exploratory rummaging in a person’s belongings’ by the gov- ernment that has been rejected since the founding.” State v. Wilson, 315 Ga. 613, 614 (884 SE2d 298) (2023) (quoting Coolidge v. New Hampshire, 403 U.S. 443, 467 (II) (C) (91 SCt 2022, 29 LE2d 564) (1971)). That safeguard is put in place when the warrant’s descrip- tion gives officers “enough guidance to locate and seize only those items the warrant authorizes.” Id. A description will generally sat-
27
isfy the particularity requirement if it is “as specific as the circum- stances and nature of activity under investigation permit.” Id. at 615 (citation and punctuation omitted).
Given the “circumstances and nature of activity” in this case, the warrants here were particularized. The first warrant — which is the main focus of Jones’s challenge — gave a clear description of what information police could access. In step one, it authorized po- lice to view the results of a Google database query, which was de- signed to turn up the location history data of devices reporting a lo- cation within 100 meters of the victim’s home between 1:00 a.m. and 5:00 a.m. on the night of the murder. As discussed above, the “place to be searched” in the warrant was not, in any real sense, Google’s entire database, but the part of the database that the police wanted to see. That is consistent with how the particularity requirement has been understood when the government is interested in a distinct piece of a larger whole. See, e.g., United States v. Karo, 468 U.S. 705, 718 (III) (104 SCt 3296, 82 LE2d 530) (1984) (a warrant to place a location-tracking beeper in an object in the suspect’s possession to
28
determine the suspect’s location did not have to describe the “place” to be searched as the entire world; the warrant could be particular- ized by describing “the object into which the beeper is to be placed, the circumstances that led agents to wish to install the beeper, and the length of time for which beeper surveillance is requested”). And the search parameters authorized by the warrant were as specific as the circumstances permitted at that initial stage: the time range matched the approximate time period when the suspect was seen at and around the victim’s home, and the geographic range was rea- sonably targeted to capture the suspect’s movements, especially given that Google location history is not precise. See Wilson, 315 Ga. at 615.
As for step two of the first warrant, it is true that police were allowed some discretion to determine whether they needed addi- tional location history for a subset of devices returned from the ini- tial list. But that discretion did not create a particularity problem. First, the officers’ discretion was not generalized. It was, instead, limited to the devices identified in step one, and dependent on the
29
facts of the investigation. That is, the officers could ask for addi- tional location history, but only for those devices that were not ex- cluded as irrelevant to the investigation because they were “outside the Target Location,” “not within the Target Location for a long enough period of time,” “moving through the Target Location in a manner inconsistent with the facts of the underlying case,” or for any other reason. And second, viewed together, step one and step two of the first warrant narrowed, rather than broadened, the access authorized by the warrant. Rather than authorizing a significantly broader initial geofence or time period, the first warrant authorized one that was narrow and particularized, and then allowed an expan- sion of the location history search that targeted only a subset of the devices as needed to determine their relevance to the investigation. In short, the description of the search in steps one and two of the first warrant was not the kind that raises the specter of a general rummaging, and it thus satisfies the particularity requirement. See Wilson, 315 Ga. at 614.
Jones takes issue with step three of the first warrant, which
30
authorized police to get identifying information tied to Google IDs “identified as relevant.” But even assuming this step gave the police too much discretion to satisfy the particularity requirement, it was not a basis for suppressing the evidence gained from these searches, because the police did not rely on that step to obtain Jones’s identi- fying information — instead, they got a new search warrant. See Pugh, 318 Ga. at 719 (2) (c) (explaining that a warrant with provi- sions that lack probable cause or are too general can be “cured” by striking those provisions and preserving the provisions that satisfy the Fourth Amendment, and holding that no evidence needed to be suppressed where the defendant had not shown that the seizure of any evidence admitted against him was authorized only by the al- legedly invalid portions of the warrant).
And that second search warrant had no particularity problem.
That warrant, which police secured after reviewing the results from the first, authorized police to obtain identifying information associ- ated with a specific device: device 290. As discussed above, the sec-
31
ond warrant was supported by an updated probable cause state- ment, and there is no question that its description of what the police could access — subscriber information sufficient to identify the in- dividual associated with device 290 — was particularized. As the warrant and its application explained, the user of device 290 was very likely the suspect, so knowing the identity of that user was rea- sonably likely to lead to evidence of a crime. The second warrant authorized police to obtain just that information.5 That was easily enough to satisfy the particularity requirement. See Pugh, 318 Ga. at 713 (2) (b); id. at 725 (Pinson, J., concurring).6
5 On its face, the second search warrant also authorized police to access
a wider range of information from device 290, including its GPS history, Internet browser history, Google search history, and application history, for two months before and after the night of the murder. But the application for the second warrant specified that police sought all that information only insofar “as it will greatly assist in identifying who the murder suspect is,” and Detective Williams testified that he actually received from Google only the Gmail address and cell phone number associated with device 290 — enough to identify the user, just as the warrant application said. So even if the warrant’s broader description of items to be seized might raise concerns about particularity , it would not invalidate the warrant here, because the police neither obtained nor used any evidence beyond what was needed to identify Jones. See Pugh, 318 Ga. at 719 (2) (c).
6 When we granted Jones’s application, we also asked the parties to ad-
dress whether Gary v. State, 262 Ga. 573 (422 SE2d 426) (1992), which de-
32
Judgment affirmed. All the Justices concur, except Peterson, PJ, and LaGrua, J., who concur specially.
PETERSON, Presiding Justice, concurring specially.
For the same reasons that I first articulated in my special concurrence in State v. Ledbetter, 318 Ga. 457, 481 (899 SE2d 222) (2024) (Peterson, PJ, concurring specially), I would overrule the last remnants of our misbegotten decision in Gary v. State, 262 Ga. 573 (422 SE2d 426) (1992). Taking that approach would render unnecessary the majority decision’s thoughtful analysis of whether the warrants were supported by probable cause, and so I do not join that analysis.
clined to adopt the federal “good-faith exception” to excluding evidence obtained in reliance on a warrant not supported by probable cause, should be overruled. Many of us still doubt that Gary can remain good law given our later decision in Mobley v. State, 307 Ga. 59 (834 SE2d 785) (2019), which rejected the reasoning of Gary and adopted a different exception to the exclusionary rule. See State v. Ledbetter, 318 Ga. 457, 480 (899 SE2d 222) (2024) (Bethel, J., concurring) (recognizing Gary’s “imperiled and untenable position in our law”); id. at 481-483 (Peterson, PJ, concurring specially) (arguing that Gary should be overruled). But the question of what to do with the “mess” of our precedent in this area, see id. at 480 (Bethel, J., concurring), 484 (Peterson, PJ, concurring specially), is not ultimately presented in this case, because the search warrants the police relied on for the evidence Jones sought to suppress were valid.
33
In so doing, I remind the Court again of what I said in Ledbetter:
(1) there is no support in the text of OCGA § 17-5-30 for the idea that the statute permits some exceptions to the exclusionary rule but excludes others; (2) the current state of the law is a mess of our own making, and we should clean it up; and (3) until we do, trial courts will be compelled to exclude evidence that federal law would admit, and the Court of Appeals will be compelled to affirm that exclusion.
318 Ga. at 484. We should not miss yet another opportunity to clean up our own mess.
I am authorized to state that Justice LaGrua joins in this concurrence.
34
Decided March 4, 2025.
Suppression of evidence. Fulton Superior Court. Before Judge Carnesale.
Stephen R. Scarborough, Jennifer Lubinsky, for appellant.
Fani T. Willis, District Attorney, Kevin C. Armstrong, Jeffrey L.
Lee, Assistant District Attorneys; Christopher M. Carr, Attorney Gen- eral, Beth A. Burton, Deputy Attorney General, Meghan H. Hill, Sen- ior Assistant Attorney General, for appellee.
Cory H. Isaacson, amicus curiae.