Carpenter v. United States

585 U.S. 296, 138 S. Ct. 2206, 201 L. Ed. 2d 507, 2018 U.S. LEXIS 3844

• Court: Supreme Court of the United States
• Decided: June 22, 2018
• Docket: 16-402
• Status: Published

Opinion

(Slip Opinion) OCTOBER TERM, 2017 1

Syllabus

NOTE: Where it is feasible, a syllabus (headnote) will be released, as is being done in connection with this case, at the time the opinion is issued. The syllabus constitutes no part of the opinion of the Court but has been prepared by the Reporter of Decisions for the convenience of the reader. See United States v. Detroit Timber & Lumber Co., 200 U. S. 321, 337.

SUPREME COURT OF THE UNITED STATES

CARPENTER v. UNITED STATES

CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT

No. 16–402. Argued November 29, 2017—Decided June 22, 2018 Cell phones perform their wide and growing variety of functions by con- tinuously connecting to a set of radio antennas called “cell sites.” Each time a phone connects to a cell site, it generates a time-stamped record known as cell-site location information (CSLI). Wireless carri- ers collect and store this information for their own business purposes. Here, after the FBI identified the cell phone numbers of several rob- bery suspects, prosecutors were granted court orders to obtain the suspects’ cell phone records under the Stored Communications Act. Wireless carriers produced CSLI for petitioner Timothy Carpenter’s phone, and the Government was able to obtain 12,898 location points cataloging Carpenter’s movements over 127 days—an average of 101 data points per day. Carpenter moved to suppress the data, arguing that the Government’s seizure of the records without obtaining a warrant supported by probable cause violated the Fourth Amend- ment. The District Court denied the motion, and prosecutors used the records at trial to show that Carpenter’s phone was near four of the robbery locations at the time those robberies occurred. Carpen- ter was convicted. The Sixth Circuit affirmed, holding that Carpen- ter lacked a reasonable expectation of privacy in the location infor- mation collected by the FBI because he had shared that information with his wireless carriers. Held: 1. The Government’s acquisition of Carpenter’s cell-site records was a Fourth Amendment search. Pp. 4–18. (a) The Fourth Amendment protects not only property interests but certain expectations of privacy as well. Katz v. United States, 389 U. S. 347, 351. Thus, when an individual “seeks to preserve some- thing as private,” and his expectation of privacy is “one that society is 2 CARPENTER v. UNITED STATES

prepared to recognize as reasonable,” official intrusion into that sphere generally qualifies as a search and requires a warrant sup- ported by probable cause. Smith v. Maryland, 442 U. S. 735, 740 (in- ternal quotation marks and alterations omitted). The analysis re- garding which expectations of privacy are entitled to protection is informed by historical understandings “of what was deemed an un- reasonable search and seizure when [the Fourth Amendment] was adopted.” Carroll v. United States, 267 U. S. 132, 149. These Found- ing-era understandings continue to inform this Court when applying the Fourth Amendment to innovations in surveillance tools. See, e.g., Kyllo v. United States, 533 U. S. 27. Pp. 4–7. (b) The digital data at issue—personal location information maintained by a third party—does not fit neatly under existing prec- edents but lies at the intersection of two lines of cases. One set ad- dresses a person’s expectation of privacy in his physical location and movements. See, e.g., United States v. Jones, 565 U. S. 400 (five Jus- tices concluding that privacy concerns would be raised by GPS track- ing). The other addresses a person’s expectation of privacy in infor- mation voluntarily turned over to third parties. See United States v. Miller, 425 U. S. 435 (no expectation of privacy in financial records held by a bank), and Smith, 442 U. S. 735 (no expectation of privacy in records of dialed telephone numbers conveyed to telephone compa- ny). Pp. 7–10. (c) Tracking a person’s past movements through CSLI partakes of many of the qualities of GPS monitoring considered in Jones—it is detailed, encyclopedic, and effortlessly compiled. At the same time, however, the fact that the individual continuously reveals his loca- tion to his wireless carrier implicates the third-party principle of Smith and Miller. Given the unique nature of cell-site records, this Court declines to extend Smith and Miller to cover them. Pp. 10–18. (1) A majority of the Court has already recognized that indi- viduals have a reasonable expectation of privacy in the whole of their physical movements. Allowing government access to cell-site rec- ords—which “hold for many Americans the ‘privacies of life,’ ” Riley v. California, 573 U. S. ___, ___—contravenes that expectation. In fact, historical cell-site records present even greater privacy concerns than the GPS monitoring considered in Jones: They give the Government near perfect surveillance and allow it to travel back in time to retrace a person’s whereabouts, subject only to the five-year retention poli- cies of most wireless carriers. The Government contends that CSLI data is less precise than GPS information, but it thought the data ac- curate enough here to highlight it during closing argument in Car- penter’s trial. At any rate, the rule the Court adopts “must take ac- count of more sophisticated systems that are already in use or in Cite as: 585 U. S. ____ (2018) 3

development,” Kyllo, 533 U. S., at 36, and the accuracy of CSLI is rapidly approaching GPS-level precision. Pp. 12–15. (2) The Government contends that the third-party doctrine governs this case, because cell-site records, like the records in Smith and Miller, are “business records,” created and maintained by wire- less carriers. But there is a world of difference between the limited types of personal information addressed in Smith and Miller and the exhaustive chronicle of location information casually collected by wireless carriers. The third-party doctrine partly stems from the notion that an indi- vidual has a reduced expectation of privacy in information knowingly shared with another. Smith and Miller, however, did not rely solely on the act of sharing. They also considered “the nature of the partic- ular documents sought” and limitations on any “legitimate ‘expecta- tion of privacy’ concerning their contents.” Miller, 425 U. S., at 442. In mechanically applying the third-party doctrine to this case the Government fails to appreciate the lack of comparable limitations on the revealing nature of CSLI. Nor does the second rationale for the third-party doctrine— voluntary exposure—hold up when it comes to CSLI. Cell phone lo- cation information is not truly “shared” as the term is normally un- derstood. First, cell phones and the services they provide are “such a pervasive and insistent part of daily life” that carrying one is indis- pensable to participation in modern society. Riley, 573 U. S., at ___. Second, a cell phone logs a cell-site record by dint of its operation, without any affirmative act on the user’s part beyond powering up. Pp. 15–17. (d) This decision is narrow. It does not express a view on matters not before the Court; does not disturb the application of Smith and Miller or call into question conventional surveillance techniques and tools, such as security cameras; does not address other business rec- ords that might incidentally reveal location information; and does not consider other collection techniques involving foreign affairs or na- tional security. Pp. 17–18. 2. The Government did not obtain a warrant supported by proba- ble cause before acquiring Carpenter’s cell-site records. It acquired those records pursuant to a court order under the Stored Communi- cations Act, which required the Government to show “reasonable grounds” for believing that the records were “relevant and material to an ongoing investigation.” 18 U. S. C. §2703(d). That showing falls well short of the probable cause required for a warrant. Consequent- ly, an order issued under §2703(d) is not a permissible mechanism for accessing historical cell-site records. Not all orders compelling the production of documents will require a showing of probable cause. A 4 CARPENTER v. UNITED STATES

warrant is required only in the rare case where the suspect has a le- gitimate privacy interest in records held by a third party. And even though the Government will generally need a warrant to access CSLI, case-specific exceptions—e.g., exigent circumstances—may support a warrantless search. Pp. 18–22. 819 F. 3d 880, reversed and remanded.

ROBERTS, C. J., delivered the opinion of the Court, in which GINS- BURG, BREYER, SOTOMAYOR, and KAGAN, JJ., joined. KENNEDY, J., filed a dissenting opinion, in which THOMAS and ALITO, JJ., joined. THOMAS, J., filed a dissenting opinion. ALITO, J., filed a dissenting opinion, in which THOMAS, J., joined. GORSUCH, J., filed a dissenting opinion. Cite as: 585 U. S. ____ (2018) 1

Opinion of the Court

NOTICE: This opinion is subject to formal revision before publication in the preliminary print of the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Wash- ington, D. C. 20543, of any typographical or other formal errors, in order that corrections may be made before the preliminary print goes to press.

SUPREME COURT OF THE UNITED STATES _________________

No. 16–402 _________________

TIMOTHY IVORY CARPENTER, PETITIONER v.

UNITED STATES

ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF

APPEALS FOR THE SIXTH CIRCUIT

[June 22, 2018]

CHIEF JUSTICE ROBERTS delivered the opinion of the Court. This case presents the question whether the Govern- ment conducts a search under the Fourth Amendment when it accesses historical cell phone records that provide a comprehensive chronicle of the user’s past movements. I

A

There are 396 million cell phone service accounts in the United States—for a Nation of 326 million people. Cell phones perform their wide and growing variety of func- tions by connecting to a set of radio antennas called “cell sites.” Although cell sites are usually mounted on a tower, they can also be found on light posts, flagpoles, church steeples, or the sides of buildings. Cell sites typically have several directional antennas that divide the covered area into sectors. Cell phones continuously scan their environment look- ing for the best signal, which generally comes from the closest cell site. Most modern devices, such as smartphones, tap into the wireless network several times 2 CARPENTER v. UNITED STATES

a minute whenever their signal is on, even if the owner is not using one of the phone’s features. Each time the phone connects to a cell site, it generates a time-stamped record known as cell-site location information (CSLI). The precision of this information depends on the size of the geographic area covered by the cell site. The greater the concentration of cell sites, the smaller the coverage area. As data usage from cell phones has increased, wireless carriers have installed more cell sites to handle the traffic. That has led to increasingly compact coverage areas, especially in urban areas. Wireless carriers collect and store CSLI for their own business purposes, including finding weak spots in their network and applying “roaming” charges when another carrier routes data through their cell sites. In addition, wireless carriers often sell aggregated location records to data brokers, without individual identifying information of the sort at issue here. While carriers have long retained CSLI for the start and end of incoming calls, in recent years phone companies have also collected location infor- mation from the transmission of text messages and rou- tine data connections. Accordingly, modern cell phones generate increasingly vast amounts of increasingly precise CSLI. B In 2011, police officers arrested four men suspected of robbing a series of Radio Shack and (ironically enough) T- Mobile stores in Detroit. One of the men confessed that, over the previous four months, the group (along with a rotating cast of getaway drivers and lookouts) had robbed nine different stores in Michigan and Ohio. The suspect identified 15 accomplices who had participated in the heists and gave the FBI some of their cell phone numbers; the FBI then reviewed his call records to identify addi- tional numbers that he had called around the time of the Cite as: 585 U. S. ____ (2018) 3

robberies. Based on that information, the prosecutors applied for court orders under the Stored Communications Act to obtain cell phone records for petitioner Timothy Carpenter and several other suspects. That statute, as amended in 1994, permits the Government to compel the disclosure of certain telecommunications records when it “offers specific and articulable facts showing that there are reasonable grounds to believe” that the records sought “are relevant and material to an ongoing criminal investigation.” 18 U. S. C. §2703(d). Federal Magistrate Judges issued two orders directing Carpenter’s wireless carriers—MetroPCS and Sprint—to disclose “cell/site sector [information] for [Carpenter’s] telephone[ ] at call origination and at call termination for incoming and outgoing calls” during the four-month period when the string of robberies occurred. App. to Pet. for Cert. 60a, 72a. The first order sought 152 days of cell-site records from MetroPCS, which produced records spanning 127 days. The second order requested seven days of CSLI from Sprint, which produced two days of records covering the period when Carpenter’s phone was “roaming” in northeastern Ohio. Altogether the Govern- ment obtained 12,898 location points cataloging Carpen- ter’s movements—an average of 101 data points per day. Carpenter was charged with six counts of robbery and an additional six counts of carrying a firearm during a federal crime of violence. See 18 U. S. C. §§924(c), 1951(a). Prior to trial, Carpenter moved to suppress the cell-site data provided by the wireless carriers. He argued that the Government’s seizure of the records violated the Fourth Amendment because they had been obtained without a warrant supported by probable cause. The District Court denied the motion. App. to Pet. for Cert. 38a–39a. At trial, seven of Carpenter’s confederates pegged him as the leader of the operation. In addition, FBI agent Christopher Hess offered expert testimony about the cell- 4 CARPENTER v. UNITED STATES

site data. Hess explained that each time a cell phone taps into the wireless network, the carrier logs a time-stamped record of the cell site and particular sector that were used. With this information, Hess produced maps that placed Carpenter’s phone near four of the charged robberies. In the Government’s view, the location records clinched the case: They confirmed that Carpenter was “right where the . . . robbery was at the exact time of the robbery.” App. 131 (closing argument). Carpenter was convicted on all but one of the firearm counts and sentenced to more than 100 years in prison. The Court of Appeals for the Sixth Circuit affirmed. 819 F. 3d 880 (2016). The court held that Carpenter lacked a reasonable expectation of privacy in the location infor- mation collected by the FBI because he had shared that information with his wireless carriers. Given that cell phone users voluntarily convey cell-site data to their carriers as “a means of establishing communication,” the court concluded that the resulting business records are not entitled to Fourth Amendment protection. Id., at 888 (quoting Smith v. Maryland, 442 U. S. 735, 741 (1979)). We granted certiorari. 582 U. S. ___ (2017). II

The Fourth Amendment protects “[t]he right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures.” The “basic purpose of this Amendment,” our cases have recog- nized, “is to safeguard the privacy and security of individ- uals against arbitrary invasions by governmental offi- cials.” Camara v. Municipal Court of City and County of San Francisco, 387 U. S. 523, 528 (1967). The Founding generation crafted the Fourth Amendment as a “response to the reviled ‘general warrants’ and ‘writs of assistance’ of the colonial era, which allowed British officers to rum- Cite as: 585 U. S. ____ (2018) 5

mage through homes in an unrestrained search for evi- dence of criminal activity.” Riley v. California, 573 U. S. ___, ___ (2014) (slip op., at 27). In fact, as John Adams recalled, the patriot James Otis’s 1761 speech condemning writs of assistance was “the first act of opposition to the arbitrary claims of Great Britain” and helped spark the Revolution itself. Id., at ___–___ (slip op., at 27–28) (quot- ing 10 Works of John Adams 248 (C. Adams ed. 1856)). For much of our history, Fourth Amendment search doctrine was “tied to common-law trespass” and focused on whether the Government “obtains information by physi- cally intruding on a constitutionally protected area.” United States v. Jones, 565 U. S. 400, 405, 406, n. 3 (2012). More recently, the Court has recognized that “property rights are not the sole measure of Fourth Amendment violations.” Soldal v. Cook County, 506 U. S. 56, 64 (1992). In Katz v. United States, 389 U. S. 347, 351 (1967), we established that “the Fourth Amendment protects people, not places,” and expanded our conception of the Amendment to protect certain expectations of privacy as well. When an individual “seeks to preserve something as private,” and his expectation of privacy is “one that society is prepared to recognize as reasonable,” we have held that official intrusion into that private sphere generally quali- fies as a search and requires a warrant supported by probable cause. Smith, 442 U. S., at 740 (internal quota- tion marks and alterations omitted). Although no single rubric definitively resolves which expectations of privacy are entitled to protection,1 the —————— 1 JUSTICE KENNEDY believes that there is such a rubric—the “proper-

ty-based concepts” that Katz purported to move beyond. Post, at 3 (dissenting opinion). But while property rights are often informative, our cases by no means suggest that such an interest is “fundamental” or “dispositive” in determining which expectations of privacy are legitimate. Post, at 8–9. JUSTICE THOMAS (and to a large extent JUSTICE GORSUCH) would have us abandon Katz and return to an 6 CARPENTER v. UNITED STATES

analysis is informed by historical understandings “of what was deemed an unreasonable search and seizure when [the Fourth Amendment] was adopted.” Carroll v. United States, 267 U. S. 132, 149 (1925). On this score, our cases have recognized some basic guideposts. First, that the Amendment seeks to secure “the privacies of life” against “arbitrary power.” Boyd v. United States, 116 U. S. 616, 630 (1886). Second, and relatedly, that a central aim of the Framers was “to place obstacles in the way of a too permeating police surveillance.” United States v. Di Re, 332 U. S. 581, 595 (1948). We have kept this attention to Founding-era under- standings in mind when applying the Fourth Amendment to innovations in surveillance tools. As technology has enhanced the Government’s capacity to encroach upon areas normally guarded from inquisitive eyes, this Court has sought to “assure[ ] preservation of that degree of privacy against government that existed when the Fourth Amendment was adopted.” Kyllo v. United States, 533 U. S. 27, 34 (2001). For that reason, we rejected in Kyllo a “mechanical interpretation” of the Fourth Amendment and held that use of a thermal imager to detect heat radiating from the side of the defendant’s home was a search. Id., at 35. Because any other conclusion would leave homeown- ers “at the mercy of advancing technology,” we determined that the Government—absent a warrant—could not capi- talize on such new sense-enhancing technology to explore —————— exclusively property-based approach. Post, at 1–2, 17–21 (THOMAS J., dissenting); post, at 6–9 (GORSUCH, J., dissenting). Katz of course “discredited” the “premise that property interests control,” 389 U. S., at 353, and we have repeatedly emphasized that privacy interests do not rise or fall with property rights, see, e.g., United States v. Jones, 565 U. S. 400, 411 (2012) (refusing to “make trespass the exclusive test”); Kyllo v. United States, 533 U. S. 27, 32 (2001) (“We have since decou- pled violation of a person’s Fourth Amendment rights from trespassory violation of his property.”). Neither party has asked the Court to reconsider Katz in this case. Cite as: 585 U. S. ____ (2018) 7

what was happening within the home. Ibid. Likewise in Riley, the Court recognized the “immense storage capacity” of modern cell phones in holding that police officers must generally obtain a warrant before searching the contents of a phone. 573 U. S., at ___ (slip op., at 17). We explained that while the general rule allowing warrantless searches incident to arrest “strikes the appropriate balance in the context of physical objects, neither of its rationales has much force with respect to” the vast store of sensitive information on a cell phone. Id., at ___ (slip op., at 9). B The case before us involves the Government’s acquisi- tion of wireless carrier cell-site records revealing the location of Carpenter’s cell phone whenever it made or received calls. This sort of digital data—personal location information maintained by a third party—does not fit neatly under existing precedents. Instead, requests for cell-site records lie at the intersection of two lines of cases, both of which inform our understanding of the privacy interests at stake. The first set of cases addresses a person’s expectation of privacy in his physical location and movements. In United States v. Knotts, 460 U. S. 276 (1983), we considered the Government’s use of a “beeper” to aid in tracking a vehicle through traffic. Police officers in that case planted a beeper in a container of chloroform before it was pur- chased by one of Knotts’s co-conspirators. The officers (with intermittent aerial assistance) then followed the automobile carrying the container from Minneapolis to Knotts’s cabin in Wisconsin, relying on the beeper’s signal to help keep the vehicle in view. The Court concluded that the “augment[ed]” visual surveillance did not constitute a search because “[a] person traveling in an automobile on public thoroughfares has no reasonable expectation of 8 CARPENTER v. UNITED STATES

privacy in his movements from one place to another.” Id., at 281, 282. Since the movements of the vehicle and its final destination had been “voluntarily conveyed to anyone who wanted to look,” Knotts could not assert a privacy interest in the information obtained. Id., at 281. This Court in Knotts, however, was careful to distin- guish between the rudimentary tracking facilitated by the beeper and more sweeping modes of surveillance. The Court emphasized the “limited use which the government made of the signals from this particular beeper” during a discrete “automotive journey.” Id., at 284, 285. Signifi- cantly, the Court reserved the question whether “different constitutional principles may be applicable” if “twenty-four hour surveillance of any citizen of this country [were] possible.” Id., at 283–284. Three decades later, the Court considered more sophis- ticated surveillance of the sort envisioned in Knotts and found that different principles did indeed apply. In United States v. Jones, FBI agents installed a GPS tracking de- vice on Jones’s vehicle and remotely monitored the vehi- cle’s movements for 28 days. The Court decided the case based on the Government’s physical trespass of the vehi- cle. 565 U. S., at 404–405. At the same time, five Justices agreed that related privacy concerns would be raised by, for example, “surreptitiously activating a stolen vehicle detection system” in Jones’s car to track Jones himself, or conducting GPS tracking of his cell phone. Id., at 426, 428 (ALITO, J., concurring in judgment); id., at 415 (SOTOMAYOR, J., concurring). Since GPS monitoring of a vehicle tracks “every movement” a person makes in that vehicle, the concurring Justices concluded that “longer term GPS monitoring in investigations of most offenses impinges on expectations of privacy”—regardless whether those movements were disclosed to the public at large. Id., at 430 (opinion of ALITO, J.); id., at 415 (opinion of Cite as: 585 U. S. ____ (2018) 9

SOTOMAYOR, J.).2 In a second set of decisions, the Court has drawn a line between what a person keeps to himself and what he shares with others. We have previously held that “a per- son has no legitimate expectation of privacy in information he voluntarily turns over to third parties.” Smith, 442 U. S., at 743–744. That remains true “even if the infor- mation is revealed on the assumption that it will be used only for a limited purpose.” United States v. Miller, 425 U. S. 435, 443 (1976). As a result, the Government is typically free to obtain such information from the recipient without triggering Fourth Amendment protections. This third-party doctrine largely traces its roots to Miller. While investigating Miller for tax evasion, the Government subpoenaed his banks, seeking several months of canceled checks, deposit slips, and monthly statements. The Court rejected a Fourth Amendment challenge to the records collection. For one, Miller could “assert neither ownership nor possession” of the docu- ments; they were “business records of the banks.” Id., at 440. For another, the nature of those records confirmed Miller’s limited expectation of privacy, because the checks were “not confidential communications but negotiable instruments to be used in commercial transactions,” and the bank statements contained information “exposed to —————— 2 JUSTICE KENNEDY argues that this case is in a different category

from Jones and the dragnet-type practices posited in Knotts because the disclosure of the cell-site records was subject to “judicial authorization.” Post, at 14–16. That line of argument conflates the threshold question whether a “search” has occurred with the separate matter of whether the search was reasonable. The subpoena process set forth in the Stored Communications Act does not determine a target’s expectation of privacy. And in any event, neither Jones nor Knotts purported to resolve the question of what authorization may be required to conduct such electronic surveillance techniques. But see Jones, 565 U. S., at 430 (ALITO, J., concurring in judgment) (indicating that longer term GPS tracking may require a warrant). 10 CARPENTER v. UNITED STATES

[bank] employees in the ordinary course of business.” Id., at 442. The Court thus concluded that Miller had “take[n] the risk, in revealing his affairs to another, that the in- formation [would] be conveyed by that person to the Gov- ernment.” Id., at 443. Three years later, Smith applied the same principles in the context of information conveyed to a telephone com- pany. The Court ruled that the Government’s use of a pen register—a device that recorded the outgoing phone num- bers dialed on a landline telephone—was not a search. Noting the pen register’s “limited capabilities,” the Court “doubt[ed] that people in general entertain any actual expectation of privacy in the numbers they dial.” 442 U. S., at 742. Telephone subscribers know, after all, that the numbers are used by the telephone company “for a variety of legitimate business purposes,” including routing calls. Id., at 743. And at any rate, the Court explained, such an expectation “is not one that society is prepared to recognize as reasonable.” Ibid. (internal quotation marks omitted). When Smith placed a call, he “voluntarily con- veyed” the dialed numbers to the phone company by “ex- pos[ing] that information to its equipment in the ordinary course of business.” Id., at 744 (internal quotation marks omitted). Once again, we held that the defendant “as- sumed the risk” that the company’s records “would be divulged to police.” Id., at 745. III The question we confront today is how to apply the Fourth Amendment to a new phenomenon: the ability to chronicle a person’s past movements through the record of his cell phone signals. Such tracking partakes of many of the qualities of the GPS monitoring we considered in Jones. Much like GPS tracking of a vehicle, cell phone location information is detailed, encyclopedic, and effort- lessly compiled. Cite as: 585 U. S. ____ (2018) 11

At the same time, the fact that the individual continu- ously reveals his location to his wireless carrier implicates the third-party principle of Smith and Miller. But while the third-party doctrine applies to telephone numbers and bank records, it is not clear whether its logic extends to the qualitatively different category of cell-site records. After all, when Smith was decided in 1979, few could have imagined a society in which a phone goes wherever its owner goes, conveying to the wireless carrier not just dialed digits, but a detailed and comprehensive record of the person’s movements. We decline to extend Smith and Miller to cover these novel circumstances. Given the unique nature of cell phone location records, the fact that the information is held by a third party does not by itself overcome the user’s claim to Fourth Amendment protection. Whether the Government employs its own surveillance technology as in Jones or leverages the technology of a wireless carrier, we hold that an individual maintains a legitimate expectation of privacy in the record of his physical movements as captured through CSLI. The location information ob- tained from Carpenter’s wireless carriers was the product of a search.3

—————— 3 The parties suggest as an alternative to their primary submissions

that the acquisition of CSLI becomes a search only if it extends beyond a limited period. See Reply Brief 12 (proposing a 24-hour cutoff); Brief for United States 55–56 (suggesting a seven-day cutoff). As part of its argument, the Government treats the seven days of CSLI requested from Sprint as the pertinent period, even though Sprint produced only two days of records. Brief for United States 56. Contrary to JUSTICE KENNEDY’s assertion, post, at 19, we need not decide whether there is a limited period for which the Government may obtain an individual’s historical CSLI free from Fourth Amendment scrutiny, and if so, how long that period might be. It is sufficient for our purposes today to hold that accessing seven days of CSLI constitutes a Fourth Amendment search. 12 CARPENTER v. UNITED STATES

A A person does not surrender all Fourth Amendment protection by venturing into the public sphere. To the contrary, “what [one] seeks to preserve as private, even in an area accessible to the public, may be constitutionally protected.” Katz, 389 U. S., at 351–352. A majority of this Court has already recognized that individuals have a reasonable expectation of privacy in the whole of their physical movements. Jones, 565 U. S., at 430 (ALITO, J., concurring in judgment); id., at 415 (SOTOMAYOR, J., concurring). Prior to the digital age, law enforcement might have pursued a suspect for a brief stretch, but doing so “for any extended period of time was difficult and costly and therefore rarely undertaken.” Id., at 429 (opinion of ALITO, J.). For that reason, “society’s expectation has been that law enforcement agents and others would not— and indeed, in the main, simply could not—secretly moni- tor and catalogue every single movement of an individual’s car for a very long period.” Id., at 430. Allowing government access to cell-site records contra- venes that expectation. Although such records are gener- ated for commercial purposes, that distinction does not negate Carpenter’s anticipation of privacy in his physical location. Mapping a cell phone’s location over the course of 127 days provides an all-encompassing record of the holder’s whereabouts. As with GPS information, the time- stamped data provides an intimate window into a person’s life, revealing not only his particular movements, but through them his “familial, political, professional, reli- gious, and sexual associations.” Id., at 415 (opinion of SOTOMAYOR, J.). These location records “hold for many Americans the ‘privacies of life.’ ” Riley, 573 U. S., at ___ (slip op., at 28) (quoting Boyd, 116 U. S., at 630). And like GPS monitoring, cell phone tracking is remarkably easy, cheap, and efficient compared to traditional investigative tools. With just the click of a button, the Government can Cite as: 585 U. S. ____ (2018) 13

access each carrier’s deep repository of historical location information at practically no expense. In fact, historical cell-site records present even greater privacy concerns than the GPS monitoring of a vehicle we considered in Jones. Unlike the bugged container in Knotts or the car in Jones, a cell phone—almost a “feature of human anatomy,” Riley, 573 U. S., at ___ (slip op., at 9)—tracks nearly exactly the movements of its owner. While individuals regularly leave their vehicles, they compulsively carry cell phones with them all the time. A cell phone faithfully follows its owner beyond public thor- oughfares and into private residences, doctor’s offices, political headquarters, and other potentially revealing locales. See id., at ___ (slip op., at 19) (noting that “nearly three-quarters of smart phone users report being within five feet of their phones most of the time, with 12% admit- ting that they even use their phones in the shower”); contrast Cardwell v. Lewis, 417 U. S. 583, 590 (1974) (plurality opinion) (“A car has little capacity for escaping public scrutiny.”). Accordingly, when the Government tracks the location of a cell phone it achieves near perfect surveillance, as if it had attached an ankle monitor to the phone’s user. Moreover, the retrospective quality of the data here gives police access to a category of information otherwise unknowable. In the past, attempts to reconstruct a per- son’s movements were limited by a dearth of records and the frailties of recollection. With access to CSLI, the Government can now travel back in time to retrace a person’s whereabouts, subject only to the retention polices of the wireless carriers, which currently maintain records for up to five years. Critically, because location infor- mation is continually logged for all of the 400 million devices in the United States—not just those belonging to persons who might happen to come under investigation— this newfound tracking capacity runs against everyone. 14 CARPENTER v. UNITED STATES

Unlike with the GPS device in Jones, police need not even know in advance whether they want to follow a particular individual, or when. Whoever the suspect turns out to be, he has effectively been tailed every moment of every day for five years, and the police may—in the Government’s view—call upon the results of that surveillance without regard to the con- straints of the Fourth Amendment. Only the few with- out cell phones could escape this tireless and absolute surveillance. The Government and JUSTICE KENNEDY contend, how- ever, that the collection of CSLI should be permitted because the data is less precise than GPS information. Not to worry, they maintain, because the location records did “not on their own suffice to place [Carpenter] at the crime scene”; they placed him within a wedge-shaped sector ranging from one-eighth to four square miles. Brief for United States 24; see post, at 18–19. Yet the Court has already rejected the proposition that “inference insulates a search.” Kyllo, 533 U. S., at 36. From the 127 days of location data it received, the Government could, in combi- nation with other information, deduce a detailed log of Carpenter’s movements, including when he was at the site of the robberies. And the Government thought the CSLI accurate enough to highlight it during the closing argu- ment of his trial. App. 131. At any rate, the rule the Court adopts “must take ac- count of more sophisticated systems that are already in use or in development.” Kyllo, 533 U. S., at 36. While the records in this case reflect the state of technology at the start of the decade, the accuracy of CSLI is rapidly ap- proaching GPS-level precision. As the number of cell sites has proliferated, the geographic area covered by each cell sector has shrunk, particularly in urban areas. In addi- tion, with new technology measuring the time and angle of signals hitting their towers, wireless carriers already have Cite as: 585 U. S. ____ (2018) 15

the capability to pinpoint a phone’s location within 50 meters. Brief for Electronic Frontier Foundation et al. as Amici Curiae 12 (describing triangulation methods that estimate a device’s location inside a given cell sector). Accordingly, when the Government accessed CSLI from the wireless carriers, it invaded Carpenter’s reason- able expectation of privacy in the whole of his physical movements. B The Government’s primary contention to the contrary is that the third-party doctrine governs this case. In its view, cell-site records are fair game because they are “business records” created and maintained by the wireless carriers. The Government (along with JUSTICE KENNEDY) recognizes that this case features new technology, but asserts that the legal question nonetheless turns on a garden-variety request for information from a third-party witness. Brief for United States 32–34; post, at 12–14. The Government’s position fails to contend with the seismic shifts in digital technology that made possible the tracking of not only Carpenter’s location but also everyone else’s, not for a short period but for years and years. Sprint Corporation and its competitors are not your typi- cal witnesses. Unlike the nosy neighbor who keeps an eye on comings and goings, they are ever alert, and their memory is nearly infallible. There is a world of difference between the limited types of personal information ad- dressed in Smith and Miller and the exhaustive chronicle of location information casually collected by wireless carriers today. The Government thus is not asking for a straightforward application of the third-party doctrine, but instead a significant extension of it to a distinct cate- gory of information. The third-party doctrine partly stems from the notion that an individual has a reduced expectation of privacy in 16 CARPENTER v. UNITED STATES

information knowingly shared with another. But the fact of “diminished privacy interests does not mean that the Fourth Amendment falls out of the picture entirely.” Riley, 573 U. S., at ___ (slip op., at 16). Smith and Miller, after all, did not rely solely on the act of sharing. Instead, they considered “the nature of the particular documents sought” to determine whether “there is a legitimate ‘expec- tation of privacy’ concerning their contents.” Miller, 425 U. S., at 442. Smith pointed out the limited capabilities of a pen register; as explained in Riley, telephone call logs reveal little in the way of “identifying information.” Smith, 442 U. S., at 742; Riley, 573 U. S., at ___ (slip op., at 24). Miller likewise noted that checks were “not confi- dential communications but negotiable instruments to be used in commercial transactions.” 425 U. S., at 442. In mechanically applying the third-party doctrine to this case, the Government fails to appreciate that there are no comparable limitations on the revealing nature of CSLI. The Court has in fact already shown special solicitude for location information in the third-party context. In Knotts, the Court relied on Smith to hold that an individ- ual has no reasonable expectation of privacy in public movements that he “voluntarily conveyed to anyone who wanted to look.” Knotts, 460 U. S., at 281; see id., at 283 (discussing Smith). But when confronted with more per- vasive tracking, five Justices agreed that longer term GPS monitoring of even a vehicle traveling on public streets constitutes a search. Jones, 565 U. S., at 430 (ALITO, J., concurring in judgment); id., at 415 (SOTOMAYOR, J., concurring). JUSTICE GORSUCH wonders why “someone’s location when using a phone” is sensitive, post, at 3, and JUSTICE KENNEDY assumes that a person’s discrete movements “are not particularly private,” post, at 17. Yet this case is not about “using a phone” or a person’s move- ment at a particular time. It is about a detailed chronicle of a person’s physical presence compiled every day, every Cite as: 585 U. S. ____ (2018) 17

moment, over several years. Such a chronicle implicates privacy concerns far beyond those considered in Smith and Miller. Neither does the second rationale underlying the third- party doctrine—voluntary exposure—hold up when it comes to CSLI. Cell phone location information is not truly “shared” as one normally understands the term. In the first place, cell phones and the services they provide are “such a pervasive and insistent part of daily life” that carrying one is indispensable to participation in modern society. Riley, 573 U. S., at ___ (slip op., at 9). Second, a cell phone logs a cell-site record by dint of its operation, without any affirmative act on the part of the user beyond powering up. Virtually any activity on the phone gener- ates CSLI, including incoming calls, texts, or e-mails and countless other data connections that a phone automati- cally makes when checking for news, weather, or social media updates. Apart from disconnecting the phone from the network, there is no way to avoid leaving behind a trail of location data. As a result, in no meaningful sense does the user voluntarily “assume[ ] the risk” of turning over a comprehensive dossier of his physical movements. Smith, 442 U. S., at 745. We therefore decline to extend Smith and Miller to the collection of CSLI. Given the unique nature of cell phone location information, the fact that the Government ob- tained the information from a third party does not over- come Carpenter’s claim to Fourth Amendment protection. The Government’s acquisition of the cell-site records was a search within the meaning of the Fourth Amendment. * * * Our decision today is a narrow one. We do not express a view on matters not before us: real-time CSLI or “tower dumps” (a download of information on all the devices that connected to a particular cell site during a particular 18 CARPENTER v. UNITED STATES

interval). We do not disturb the application of Smith and Miller or call into question conventional surveillance techniques and tools, such as security cameras. Nor do we address other business records that might incidentally reveal location information. Further, our opinion does not consider other collection techniques involving foreign affairs or national security. As Justice Frankfurter noted when considering new innovations in airplanes and radios, the Court must tread carefully in such cases, to ensure that we do not “embarrass the future.” Northwest Air- lines, Inc. v. Minnesota, 322 U. S. 292, 300 (1944).4 IV Having found that the acquisition of Carpenter’s CSLI was a search, we also conclude that the Government must generally obtain a warrant supported by probable cause before acquiring such records. Although the “ultimate measure of the constitutionality of a governmental search is ‘reasonableness,’ ” our cases establish that warrantless searches are typically unreasonable where “a search is undertaken by law enforcement officials to discover evi- dence of criminal wrongdoing.” Vernonia School Dist. 47J v. Acton, 515 U. S. 646, 652–653 (1995). Thus, “[i]n the absence of a warrant, a search is reasonable only if it falls within a specific exception to the warrant requirement.” Riley, 573 U. S., at ___ (slip op., at 5). The Government acquired the cell-site records pursuant to a court order issued under the Stored Communications Act, which required the Government to show “reasonable grounds” for believing that the records were “relevant and —————— 4 JUSTICE GORSUCH faults us for not promulgating a complete code

addressing the manifold situations that may be presented by this new technology—under a constitutional provision turning on what is “rea- sonable,” no less. Post, at 10–12. Like JUSTICE GORSUCH, we “do not begin to claim all the answers today,” post, at 13, and therefore decide no more than the case before us. Cite as: 585 U. S. ____ (2018) 19

material to an ongoing investigation.” 18 U. S. C. §2703(d). That showing falls well short of the probable cause required for a warrant. The Court usually requires “some quantum of individualized suspicion” before a search or seizure may take place. United States v. Martinez-Fuerte, 428 U. S. 543, 560–561 (1976). Under the standard in the Stored Communications Act, however, law enforcement need only show that the cell-site evidence might be pertinent to an ongoing investigation—a “gigan- tic” departure from the probable cause rule, as the Gov- ernment explained below. App. 34. Consequently, an order issued under Section 2703(d) of the Act is not a permissible mechanism for accessing historical cell-site records. Before compelling a wireless carrier to turn over a subscriber’s CSLI, the Government’s obligation is a familiar one—get a warrant. JUSTICE ALITO contends that the warrant requirement simply does not apply when the Government acquires records using compulsory process. Unlike an actual search, he says, subpoenas for documents do not involve the direct taking of evidence; they are at most a “construc- tive search” conducted by the target of the subpoena. Post, at 12. Given this lesser intrusion on personal privacy, JUSTICE ALITO argues that the compulsory production of records is not held to the same probable cause standard. In his view, this Court’s precedents set forth a categorical rule—separate and distinct from the third-party doc- trine—subjecting subpoenas to lenient scrutiny without regard to the suspect’s expectation of privacy in the rec- ords. Post, at 8–19. But this Court has never held that the Government may subpoena third parties for records in which the suspect has a reasonable expectation of privacy. Almost all of the examples JUSTICE ALITO cites, see post, at 14–15, contem- plated requests for evidence implicating diminished pri- 20 CARPENTER v. UNITED STATES

vacy interests or for a corporation’s own books.5 The lone exception, of course, is Miller, where the Court’s analysis of the third-party subpoena merged with the application of the third-party doctrine. 425 U. S., at 444 (concluding that Miller lacked the necessary privacy interest to contest the issuance of a subpoena to his bank). JUSTICE ALITO overlooks the critical issue. At some point, the dissent should recognize that CSLI is an entirely different species of business record—something that implicates basic Fourth Amendment concerns about arbi- trary government power much more directly than corpo- rate tax or payroll ledgers. When confronting new con- cerns wrought by digital technology, this Court has been careful not to uncritically extend existing precedents. See Riley, 573 U. S., at ___ (slip op., at 10) (“A search of the information on a cell phone bears little resemblance to the type of brief physical search considered [in prior precedents].”). If the choice to proceed by subpoena provided a categori- cal limitation on Fourth Amendment protection, no type of record would ever be protected by the warrant require- ment. Under JUSTICE ALITO’s view, private letters, digital contents of a cell phone—any personal information re- duced to document form, in fact—may be collected by —————— 5 See United States v. Dionisio, 410 U. S. 1, 14 (1973) (“No person can

have a reasonable expectation that others will not know the sound of his voice”); Donovan v. Lone Steer, Inc., 464 U. S. 408, 411, 415 (1984) (payroll and sales records); California Bankers Assn. v. Shultz, 416 U. S. 21, 67 (1974) (Bank Secrecy Act reporting requirements); See v. Seattle, 387 U. S. 541, 544 (1967) (financial books and records); United States v. Powell, 379 U. S. 48, 49, 57 (1964) (corporate tax records); McPhaul v. United States, 364 U. S. 372, 374, 382 (1960) (books and records of an organization); United States v. Morton Salt Co., 338 U. S. 632, 634, 651–653 (1950) (Federal Trade Commission reporting re- quirement); Oklahoma Press Publishing Co. v. Walling, 327 U. S. 186, 189, 204–208 (1946) (payroll records); Hale v. Henkel, 201 U. S. 43, 45, 75 (1906) (corporate books and papers). Cite as: 585 U. S. ____ (2018) 21

subpoena for no reason other than “official curiosity.” United States v. Morton Salt Co., 338 U. S. 632, 652 (1950). JUSTICE KENNEDY declines to adopt the radical implications of this theory, leaving open the question whether the warrant requirement applies “when the Gov- ernment obtains the modern-day equivalents of an indi- vidual’s own ‘papers’ or ‘effects,’ even when those papers or effects are held by a third party. ” Post, at 13 (citing United States v. Warshak, 631 F. 3d 266, 283–288 (CA6 2010)). That would be a sensible exception, because it would prevent the subpoena doctrine from overcoming any reasonable expectation of privacy. If the third-party doc- trine does not apply to the “modern-day equivalents of an individual’s own ‘papers’ or ‘effects,’ ” then the clear impli- cation is that the documents should receive full Fourth Amendment protection. We simply think that such pro- tection should extend as well to a detailed log of a person’s movements over several years. This is certainly not to say that all orders compelling the production of documents will require a showing of proba- ble cause. The Government will be able to use subpoenas to acquire records in the overwhelming majority of inves- tigations. We hold only that a warrant is required in the rare case where the suspect has a legitimate privacy in- terest in records held by a third party. Further, even though the Government will generally need a warrant to access CSLI, case-specific exceptions may support a warrantless search of an individual’s cell- site records under certain circumstances. “One well- recognized exception applies when ‘ “the exigencies of the situation” make the needs of law enforcement so compel- ling that [a] warrantless search is objectively reasonable under the Fourth Amendment.’ ” Kentucky v. King, 563 U. S. 452, 460 (2011) (quoting Mincey v. Arizona, 437 U. S. 385, 394 (1978)). Such exigencies include the need to pursue a fleeing suspect, protect individuals who are 22 CARPENTER v. UNITED STATES

threatened with imminent harm, or prevent the imminent destruction of evidence. 563 U. S., at 460, and n. 3. As a result, if law enforcement is confronted with an urgent situation, such fact-specific threats will likely justify the warrantless collection of CSLI. Lower courts, for instance, have approved warrantless searches related to bomb threats, active shootings, and child abductions. Our decision today does not call into doubt warrantless access to CSLI in such circumstances. While police must get a warrant when collecting CSLI to assist in the mine- run criminal investigation, the rule we set forth does not limit their ability to respond to an ongoing emergency. * * * As Justice Brandeis explained in his famous dissent, the Court is obligated—as “[s]ubtler and more far-reaching means of invading privacy have become available to the Government”—to ensure that the “progress of science” does not erode Fourth Amendment protections. Olmstead v. United States, 277 U. S. 438, 473–474 (1928). Here the progress of science has afforded law enforcement a power- ful new tool to carry out its important responsibilities. At the same time, this tool risks Government encroachment of the sort the Framers, “after consulting the lessons of history,” drafted the Fourth Amendment to prevent. Di Re, 332 U. S., at 595. We decline to grant the state unrestricted access to a wireless carrier’s database of physical location infor- mation. In light of the deeply revealing nature of CSLI, its depth, breadth, and comprehensive reach, and the inescapable and automatic nature of its collection, the fact that such information is gathered by a third party does not make it any less deserving of Fourth Amendment protec- tion. The Government’s acquisition of the cell-site records here was a search under that Amendment. The judgment of the Court of Appeals is reversed, and Cite as: 585 U. S. ____ (2018) 23

the case is remanded for further proceedings consistent with this opinion. It is so ordered. Cite as: 585 U. S. ____ (2018) 1

KENNEDY, J., dissenting

JUSTICE KENNEDY, with whom JUSTICE THOMAS and JUSTICE ALITO join, dissenting. This case involves new technology, but the Court’s stark departure from relevant Fourth Amendment precedents and principles is, in my submission, unnecessary and incorrect, requiring this respectful dissent. The new rule the Court seems to formulate puts needed, reasonable, accepted, lawful, and congressionally author- ized criminal investigations at serious risk in serious cases, often when law enforcement seeks to prevent the threat of violent crimes. And it places undue restrictions on the lawful and necessary enforcement powers exercised not only by the Federal Government, but also by law enforcement in every State and locality throughout the Nation. Adherence to this Court’s longstanding prece- dents and analytic framework would have been the proper and prudent way to resolve this case. The Court has twice held that individuals have no Fourth Amendment interests in business records which are possessed, owned, and controlled by a third party. United States v. Miller, 425 U. S. 435 (1976); Smith v. Maryland, 442 U. S. 735 (1979). This is true even when the records contain personal and sensitive information. So when the Government uses a subpoena to obtain, for example, bank records, telephone records, and credit card 2 CARPENTER v. UNITED STATES

statements from the businesses that create and keep these records, the Government does not engage in a search of the business’s customers within the meaning of the Fourth Amendment. In this case petitioner challenges the Government’s right to use compulsory process to obtain a now-common kind of business record: cell-site records held by cell phone service providers. The Government acquired the records through an investigative process enacted by Congress. Upon approval by a neutral magistrate, and based on the Government’s duty to show reasonable necessity, it au- thorizes the disclosure of records and information that are under the control and ownership of the cell phone service provider, not its customer. Petitioner acknowledges that the Government may obtain a wide variety of business records using compulsory process, and he does not ask the Court to revisit its precedents. Yet he argues that, under those same precedents, the Government searched his records when it used court-approved compulsory process to obtain the cell-site information at issue here. Cell-site records, however, are no different from the many other kinds of business records the Government has a lawful right to obtain by compulsory process. Customers like petitioner do not own, possess, control, or use the records, and for that reason have no reasonable expecta- tion that they cannot be disclosed pursuant to lawful compulsory process. The Court today disagrees. It holds for the first time that by using compulsory process to obtain records of a business entity, the Government has not just engaged in an impermissible action, but has conducted a search of the business’s customer. The Court further concludes that the search in this case was unreasonable and the Government needed to get a warrant to obtain more than six days of cell-site records. In concluding that the Government engaged in a search, Cite as: 585 U. S. ____ (2018) 3

the Court unhinges Fourth Amendment doctrine from the property-based concepts that have long grounded the analytic framework that pertains in these cases. In doing so it draws an unprincipled and unworkable line between cell-site records on the one hand and financial and tele- phonic records on the other. According to today’s majority opinion, the Government can acquire a record of every credit card purchase and phone call a person makes over months or years without upsetting a legitimate expecta- tion of privacy. But, in the Court’s view, the Government crosses a constitutional line when it obtains a court’s approval to issue a subpoena for more than six days of cell-site records in order to determine whether a person was within several hundred city blocks of a crime scene. That distinction is illogical and will frustrate principled application of the Fourth Amendment in many routine yet vital law enforcement operations. It is true that the Cyber Age has vast potential both to expand and restrict individual freedoms in dimensions not contemplated in earlier times. See Packingham v. North Carolina, 582 U. S. ___, ___–___ (2017) (slip op., at 46). For the reasons that follow, however, there is simply no basis here for concluding that the Government interfered with information that the cell phone customer, either from a legal or commonsense standpoint, should have thought the law would deem owned or controlled by him. I Before evaluating the question presented it is helpful to understand the nature of cell-site records, how they are commonly used by cell phone service providers, and their proper use by law enforcement. When a cell phone user makes a call, sends a text mes- sage or e-mail, or gains access to the Internet, the cell phone establishes a radio connection to an antenna at a nearby cell site. The typical cell site covers a more-or-less 4 CARPENTER v. UNITED STATES

circular geographic area around the site. It has three (or sometimes six) separate antennas pointing in different directions. Each provides cell service for a different 120- degree (or 60-degree) sector of the cell site’s circular cover- age area. So a cell phone activated on the north side of a cell site will connect to a different antenna than a cell phone on the south side. Cell phone service providers create records each time a cell phone connects to an antenna at a cell site. For a phone call, for example, the provider records the date, time, and duration of the call; the phone numbers making and receiving the call; and, most relevant here, the cell site used to make the call, as well as the specific antenna that made the connection. The cell-site and antenna data points, together with the date and time of connection, are known as cell-site location information, or cell-site records. By linking an individual’s cell phone to a particular 120- or 60-degree sector of a cell site’s coverage area at a par- ticular time, cell-site records reveal the general location of the cell phone user. The location information revealed by cell-site records is imprecise, because an individual cell-site sector usually covers a large geographic area. The FBI agent who offered expert testimony about the cell-site records at issue here testified that a cell site in a city reaches between a half mile and two miles in all directions. That means a 60- degree sector covers between approximately one-eighth and two square miles (and a 120-degree sector twice that area). To put that in perspective, in urban areas cell-site records often would reveal the location of a cell phone user within an area covering between around a dozen and several hundred city blocks. In rural areas cell-site rec- ords can be up to 40 times more imprecise. By contrast, a Global Positioning System (GPS) can reveal an individ- ual’s location within around 15 feet. Major cell phone service providers keep cell-site records Cite as: 585 U. S. ____ (2018) 5

for long periods of time. There is no law requiring them to do so. Instead, providers contract with their customers to collect and keep these records because they are valuable to the providers. Among other things, providers aggregate the records and sell them to third parties along with other information gleaned from cell phone usage. This data can be used, for example, to help a department store deter- mine which of various prospective store locations is likely to get more foot traffic from middle-aged women who live in affluent zip codes. The market for cell phone data is now estimated to be in the billions of dollars. See Brief for Technology Experts as Amici Curiae 23. Cell-site records also can serve an important investiga- tive function, as the facts of this case demonstrate. Peti- tioner, Timothy Carpenter, along with a rotating group of accomplices, robbed at least six RadioShack and T-Mobile stores at gunpoint over a 2-year period. Five of those robberies occurred in the Detroit area, each crime at least four miles from the last. The sixth took place in Warren, Ohio, over 200 miles from Detroit. The Government, of course, did not know all of these details in 2011 when it began investigating Carpenter. In April of that year police arrested four of Carpenter’s co- conspirators. One of them confessed to committing nine robberies in Michigan and Ohio between December 2010 and March 2011. He identified 15 accomplices who had participated in at least one of those robberies; named Carpenter as one of the accomplices; and provided Carpen- ter’s cell phone number to the authorities. The suspect also warned that the other members of the conspiracy planned to commit more armed robberies in the immediate future. The Government at this point faced a daunting task. Even if it could identify and apprehend the suspects, still it had to link each suspect in this changing criminal gang to specific robberies in order to bring charges and convict. 6 CARPENTER v. UNITED STATES

And, of course, it was urgent that the Government take all necessary steps to stop the ongoing and dangerous crime spree. Cell-site records were uniquely suited to this task. The geographic dispersion of the robberies meant that, if Car- penter’s cell phone were within even a dozen to several hundred city blocks of one or more of the stores when the different robberies occurred, there would be powerful circumstantial evidence of his participation; and this would be especially so if his cell phone usually was not located in the sectors near the stores except during the robbery times. To obtain these records, the Government applied to federal magistrate judges for disclosure orders pursuant to §2703(d) of the Stored Communications Act. That Act authorizes a magistrate judge to issue an order requiring disclosure of cell-site records if the Government demon- strates “specific and articulable facts showing that there are reasonable grounds to believe” the records “are rele- vant and material to an ongoing criminal investigation.” 18 U. S. C. §§2703(d), 2711(3). The full statutory provi- sion is set out in the Appendix, infra. From Carpenter’s primary service provider, MetroPCS, the Government obtained records from between December 2010 and April 2011, based on its understanding that nine robberies had occurred in that timeframe. The Govern- ment also requested seven days of cell-site records from Sprint, spanning the time around the robbery in Warren, Ohio. It obtained two days of records. These records confirmed that Carpenter’s cell phone was in the general vicinity of four of the nine robberies, includ- ing the one in Ohio, at the times those robberies occurred. II The first Clause of the Fourth Amendment provides that “the right of the people to be secure in their persons, houses, Cite as: 585 U. S. ____ (2018) 7

papers, and effects, against unreasonable searches and seizures, shall not be violated.” The customary beginning point in any Fourth Amendment search case is whether the Government’s actions constitute a “search” of the defendant’s person, house, papers, or effects, within the meaning of the constitutional provision. If so, the next question is whether that search was reasonable. Here the only question necessary to decide is whether the Government searched anything of Carpenter’s when it used compulsory process to obtain cell-site records from Carpenter’s cell phone service providers. This Court’s decisions in Miller and Smith dictate that the answer is no, as every Court of Appeals to have considered the ques- tion has recognized. See United States v. Thompson, 866 F. 3d 1149 (CA10 2017); United States v. Graham, 824 F. 3d 421 (CA4 2016) (en banc); Carpenter v. United States, 819 F. 3d 880 (CA6 2016); United States v. Davis, 785 F. 3d 498 (CA11 2015) (en banc); In re Application of U. S. for Historical Cell Site Data, 724 F. 3d 600 (CA5 2013). A Miller and Smith hold that individuals lack any protected Fourth Amendment interests in records that are pos- sessed, owned, and controlled only by a third party. In Miller federal law enforcement officers obtained four months of the defendant’s banking records. 425 U. S., at 437438. And in Smith state police obtained records of the phone numbers dialed from the defendant’s home phone. 442 U. S., at 737. The Court held in both cases that the officers did not search anything belonging to the defendants within the meaning of the Fourth Amendment. The defendants could “assert neither ownership nor pos- session” of the records because the records were created, owned, and controlled by the companies. Miller, supra, at 440; see Smith, supra, at 741. And the defendants had no 8 CARPENTER v. UNITED STATES

reasonable expectation of privacy in information they “voluntarily conveyed to the [companies] and exposed to their employees in the ordinary course of business.” Mil- ler, supra, at 442; see Smith, 442 U. S., at 744. Rather, the defendants “assumed the risk that the information would be divulged to police.” Id., at 745. Miller and Smith have been criticized as being based on too narrow a view of reasonable expectations of privacy. See, e.g., Ashdown, The Fourth Amendment and the “Le- gitimate Expectation of Privacy,” 34 Vand. L. Rev. 1289, 13131316 (1981). Those criticisms, however, are unwar- ranted. The principle established in Miller and Smith is correct for two reasons, the first relating to a defendant’s attenuated interest in property owned by another, and the second relating to the safeguards inherent in the use of compulsory process. First, Miller and Smith placed necessary limits on the ability of individuals to assert Fourth Amendment inter- ests in property to which they lack a “requisite connec- tion.” Minnesota v. Carter, 525 U. S. 83, 99 (1998) (KENNEDY, J., concurring). Fourth Amendment rights, after all, are personal. The Amendment protects “[t]he right of the people to be secure in their . . . persons, houses, papers, and effects”—not the persons, houses, papers, and effects of others. (Emphasis added.) The concept of reasonable expectations of privacy, first announced in Katz v. United States, 389 U. S. 347 (1967), sought to look beyond the “arcane distinctions developed in property and tort law” in evaluating whether a person has a sufficient connection to the thing or place searched to assert Fourth Amendment interests in it. Rakas v. Illinois, 439 U. S. 128, 143 (1978). Yet “property concepts” are, nonetheless, fundamental “in determining the pres- ence or absence of the privacy interests protected by that Amendment.” Id., at 143144, n. 12. This is so for at least two reasons. First, as a matter of settled expectations Cite as: 585 U. S. ____ (2018) 9

from the law of property, individuals often have greater expectations of privacy in things and places that belong to them, not to others. And second, the Fourth Amendment’s protections must remain tethered to the text of that Amendment, which, again, protects only a person’s own “persons, houses, papers, and effects.” Katz did not abandon reliance on property-based con- cepts. The Court in Katz analogized the phone booth used in that case to a friend’s apartment, a taxicab, and a hotel room. 389 U. S., at 352, 359. So when the defendant “shu[t] the door behind him” and “pa[id] the toll,” id., at 352, he had a temporary interest in the space and a legit- imate expectation that others would not intrude, much like the interest a hotel guest has in a hotel room, Stoner v. California, 376 U. S. 483 (1964), or an overnight guest has in a host’s home, Minnesota v. Olson, 495 U. S. 91 (1990). The Government intruded on that space when it attached a listening device to the phone booth. Katz, 389 U. S., at 348. (And even so, the Court made it clear that the Government’s search could have been reasonable had there been judicial approval on a case-specific basis, which, of course, did occur here. Id., at 357359.) Miller and Smith set forth an important and necessary limitation on the Katz framework. They rest upon the commonsense principle that the absence of property law analogues can be dispositive of privacy expectations. The defendants in those cases could expect that the third-party businesses could use the records the companies collected, stored, and classified as their own for any number of business and commercial purposes. The businesses were not bailees or custodians of the records, with a duty to hold the records for the defendants’ use. The defendants could make no argument that the records were their own papers or effects. See Miller, supra, at 440 (“the docu- ments subpoenaed here are not respondent’s ‘private papers’ ”); Smith, supra, at 741 (“petitioner obviously 10 CARPENTER v. UNITED STATES

cannot claim that his ‘property’ was invaded”). The rec- ords were the business entities’ records, plain and simple. The defendants had no reason to believe the records were owned or controlled by them and so could not assert a reasonable expectation of privacy in the records. The second principle supporting Miller and Smith is the longstanding rule that the Government may use compul- sory process to compel persons to disclose documents and other evidence within their possession and control. See United States v. Nixon, 418 U. S. 683, 709 (1974) (it is an “ancient proposition of law” that “the public has a right to every man’s evidence” (internal quotation marks and alterations omitted)). A subpoena is different from a warrant in its force and intrusive power. While a warrant allows the Government to enter and seize and make the examination itself, a subpoena simply requires the person to whom it is directed to make the disclosure. A subpoena, moreover, provides the recipient the “opportunity to pre- sent objections” before complying, which further mitigates the intrusion. Oklahoma Press Publishing Co. v. Walling, 327 U. S. 186, 195 (1946). For those reasons this Court has held that a subpoena for records, although a “constructive” search subject to Fourth Amendment constraints, need not comply with the procedures applicable to warrants—even when challenged by the person to whom the records belong. Id., at 202, 208. Rather, a subpoena complies with the Fourth Amendment’s reasonableness requirement so long as it is “ ‘sufficiently limited in scope, relevant in purpose, and specific in directive so that compliance will not be unrea- sonably burdensome.’ ” Donovan v. Lone Steer, Inc., 464 U. S. 408, 415 (1984). Persons with no meaningful inter- ests in the records sought by a subpoena, like the defend- ants in Miller and Smith, have no rights to object to the records’ disclosure—much less to assert that the Govern- ment must obtain a warrant to compel disclosure of the Cite as: 585 U. S. ____ (2018) 11

records. See Miller, 425 U. S., at 444446; SEC v. Jerry T. O’Brien, Inc., 467 U. S. 735, 742743 (1984). Based on Miller and Smith and the principles underly- ing those cases, it is well established that subpoenas may be used to obtain a wide variety of records held by busi- nesses, even when the records contain private information. See 2 W. LaFave, Search and Seizure §4.13 (5th ed. 2012). Credit cards are a prime example. State and federal law enforcement, for instance, often subpoena credit card statements to develop probable cause to prosecute crimes ranging from drug trafficking and distribution to healthcare fraud to tax evasion. See United States v. Phibbs, 999 F. 2d 1053 (CA6 1993) (drug distribution); McCune v. DOJ, 592 Fed. Appx. 287 (CA5 2014) (healthcare fraud); United States v. Green, 305 F. 3d 422 (CA6 2002) (drug trafficking and tax evasion); see also 12 U. S. C. §§3402(4), 3407 (allowing the Government to subpoena financial records if “there is reason to believe that the records sought are relevant to a legitimate law enforcement inquiry”). Subpoenas also may be used to obtain vehicle registration records, hotel records, employ- ment records, and records of utility usage, to name just a few other examples. See 1 LaFave, supra, §2.7(c). And law enforcement officers are not alone in their reliance on subpoenas to obtain business records for legit- imate investigations. Subpoenas also are used for investi- gatory purposes by state and federal grand juries, see United States v. Dionisio, 410 U. S. 1 (1973), state and federal administrative agencies, see Oklahoma Press, supra, and state and federal legislative bodies, see McPhaul v. United States, 364 U. S. 372 (1960). B Carpenter does not question these traditional investiga- tive practices. And he does not ask the Court to reconsider Miller and Smith. Carpenter argues only that, under 12 CARPENTER v. UNITED STATES

Miller and Smith, the Government may not use compulsory process to acquire cell-site records from cell phone service providers. There is no merit in this argument. Cell-site records, like all the examples just discussed, are created, kept, classified, owned, and controlled by cell phone service providers, which aggregate and sell this information to third parties. As in Miller, Carpenter can “assert neither ownership nor possession” of the records and has no con- trol over them. 425 U. S., at 440. Carpenter argues that he has Fourth Amendment inter- ests in the cell-site records because they are in essence his personal papers by operation of 47 U. S. C. §222. That statute imposes certain restrictions on how providers may use “customer proprietary network information”—a term that encompasses cell-site records. §§222(c), (h)(1)(A). The statute in general prohibits providers from disclosing personally identifiable cell-site records to private third parties. §222(c)(1). And it allows customers to request cell-site records from the provider. §222(c)(2). Carpenter’s argument is unpersuasive, however, for §222 does not grant cell phone customers any meaningful interest in cell-site records. The statute’s confidentiality protections may be overridden by the interests of the providers or the Government. The providers may disclose the records “to protect the[ir] rights or property” or to “initiate, render, bill, and collect for telecommunications services.” §§222(d)(1), (2). They also may disclose the records “as required by law”—which, of course, is how they were disclosed in this case. §222(c)(1). Nor does the stat- ute provide customers any practical control over the rec- ords. Customers do not create the records; they have no say in whether or for how long the records are stored; and they cannot require the records to be modified or de- stroyed. Even their right to request access to the records is limited, for the statute “does not preclude a carrier from Cite as: 585 U. S. ____ (2018) 13

being reimbursed by the customers . . . for the costs asso- ciated with making such disclosures.” H. R. Rep. No. 104– 204, pt. 1, p. 90 (1995). So in every legal and practical sense the “network information” regulated by §222 is, under that statute, “proprietary” to the service providers, not Carpenter. The Court does not argue otherwise. Because Carpenter lacks a requisite connection to the cell-site records, he also may not claim a reasonable expec- tation of privacy in them. He could expect that a third party—the cell phone service provider—could use the information it collected, stored, and classified as its own for a variety of business and commercial purposes. All this is not to say that Miller and Smith are without limits. Miller and Smith may not apply when the Gov- ernment obtains the modern-day equivalents of an indi- vidual’s own “papers” or “effects,” even when those papers or effects are held by a third party. See Ex parte Jackson, 96 U. S. 727, 733 (1878) (letters held by mail carrier); United States v. Warshak, 631 F. 3d 266, 283288 (CA6 2010) (e-mails held by Internet service provider). As already discussed, however, this case does not involve property or a bailment of that sort. Here the Govern- ment’s acquisition of cell-site records falls within the heartland of Miller and Smith. In fact, Carpenter’s Fourth Amendment objection is even weaker than those of the defendants in Miller and Smith. Here the Government did not use a mere sub- poena to obtain the cell-site records. It acquired the records only after it proved to a Magistrate Judge reasonable grounds to believe that the records were relevant and material to an ongoing criminal investigation. See 18 U. S. C. §2703(d). So even if §222 gave Carpenter some attenuated interest in the records, the Government’s conduct here would be reasonable under the standards governing subpoenas. See Donovan, 464 U. S., at 415. Under Miller and Smith, then, a search of the sort that 14 CARPENTER v. UNITED STATES

requires a warrant simply did not occur when the Gov- ernment used court-approved compulsory process, based on a finding of reasonable necessity, to compel a cell phone service provider, as owner, to disclose cell-site records. III The Court rejects a straightforward application of Miller and Smith. It concludes instead that applying those cases to cell-site records would work a “significant extension” of the principles underlying them, ante, at 15, and holds that the acquisition of more than six days of cell-site records constitutes a search, ante, at 11, n. 3. In my respectful view the majority opinion misreads this Court’s precedents, old and recent, and transforms Miller and Smith into an unprincipled and unworkable doctrine. The Court’s newly conceived constitutional standard will cause confusion; will undermine traditional and important law enforcement practices; and will allow the cell phone to become a protected medium that dangerous persons will use to commit serious crimes. A The Court errs at the outset by attempting to sidestep Miller and Smith. The Court frames this case as following instead from United States v. Knotts, 460 U. S. 276 (1983), and United States v. Jones, 565 U. S. 400 (2012). Those cases, the Court suggests, establish that “individuals have a reasonable expectation of privacy in the whole of their physical movements.” Ante, at 79, 12. Knotts held just the opposite: “A person traveling in an automobile on public thoroughfares has no reasonable expectation of privacy in his movements from one place to another.” 460 U. S., at 281. True, the Court in Knotts also suggested that “different constitutional principles may be applicable” to “dragnet-type law enforcement practices.” Id., at 284. But by dragnet practices the Court was refer- Cite as: 585 U. S. ____ (2018) 15

ring to “ ‘twenty-four hour surveillance of any citizen of this country . . . without judicial knowledge or supervi- sion.’ ” Id., at 283. Those “different constitutional principles” mentioned in Knotts, whatever they may be, do not apply in this case. Here the Stored Communications Act requires a neutral judicial officer to confirm in each case that the Govern- ment has “reasonable grounds to believe” the cell-site records “are relevant and material to an ongoing criminal investigation.” 18 U. S. C. §2703(d). This judicial check mitigates the Court’s concerns about “ ‘a too permeating police surveillance.’ ” Ante, at 6 (quoting United States v. Di Re, 332 U. S. 581, 595 (1948)). Here, even more so than in Knotts, “reality hardly suggests abuse.” 460 U. S., at 284. The Court’s reliance on Jones fares no better. In Jones the Government installed a GPS tracking device on the defendant’s automobile. The Court held the Government searched the automobile because it “physically occupied private property [of the defendant] for the purpose of obtaining information.” 565 U. S., at 404. So in Jones it was “not necessary to inquire about the target’s expecta- tion of privacy in his vehicle’s movements.” Grady v. North Carolina, 575 U. S. ___, ___ (2015) (per curiam) (slip op., at 3). Despite that clear delineation of the Court’s holding in Jones, the Court today declares that Jones applied the “ ‘different constitutional principles’ ” alluded to in Knotts to establish that an individual has an expectation of pri- vacy in the sum of his whereabouts. Ante, at 8, 12. For that proposition the majority relies on the two concurring opinions in Jones, one of which stated that “longer term GPS monitoring in investigations of most offenses impinges on expectations of privacy.” 565 U. S., at 430 (ALITO, J., concurring). But Jones involved direct governmental surveillance of a defendant’s automobile without judicial 16 CARPENTER v. UNITED STATES

authorization—specifically, GPS surveillance accurate within 50 to 100 feet. Id., at 402403. Even assuming that the different constitutional principles mentioned in Knotts would apply in a case like Jones—a proposition the Court was careful not to announce in Jones, supra, at 412413—those principles are inapplicable here. Cases like this one, where the Government uses court-approved compulsory process to obtain records owned and controlled by a third party, are governed by the two majority opin- ions in Miller and Smith. B The Court continues its analysis by misinterpreting Miller and Smith, and then it reaches the wrong outcome on these facts even under its flawed standard. The Court appears, in my respectful view, to read Miller and Smith to establish a balancing test. For each “quali- tatively different category” of information, the Court suggests, the privacy interests at stake must be weighed against the fact that the information has been disclosed to a third party. See ante, at 11, 1517. When the privacy interests are weighty enough to “overcome” the third-party disclosure, the Fourth Amendment’s protections apply. See ante, at 17. That is an untenable reading of Miller and Smith. As already discussed, the fact that information was relin- quished to a third party was the entire basis for conclud- ing that the defendants in those cases lacked a reasonable expectation of privacy. Miller and Smith do not establish the kind of category-by-category balancing the Court today prescribes. But suppose the Court were correct to say that Miller and Smith rest on so imprecise a foundation. Still the Court errs, in my submission, when it concludes that cell- site records implicate greater privacy interests—and thus deserve greater Fourth Amendment protection—than Cite as: 585 U. S. ____ (2018) 17

financial records and telephone records. Indeed, the opposite is true. A person’s movements are not particularly private. As the Court recognized in Knotts, when the defendant there “traveled over the public streets he voluntarily conveyed to anyone who wanted to look the fact that he was traveling over particular roads in a particular direction, the fact of whatever stops he made, and the fact of his final destination.” 460 U. S., at 281282. Today expectations of privacy in one’s location are, if anything, even less reasonable than when the Court decided Knotts over 30 years ago. Millions of Americans choose to share their location on a daily basis, whether by using a variety of location-based services on their phones, or by sharing their location with friends and the public at large via social media. And cell-site records, as already discussed, disclose a person’s location only in a general area. The records at issue here, for example, revealed Carpenter’s location within an area covering between around a dozen and several hundred city blocks. “Areas of this scale might encompass bridal stores and Bass Pro Shops, gay bars and straight ones, a Methodist church and the local mosque.” 819 F. 3d 880, 889 (CA6 2016). These records could not reveal where Carpenter lives and works, much less his “ ‘familial, political, professional, religious, and sexual associations.’ ” Ante, at 12 (quoting Jones, supra, at 415 (SOTOMAYOR, J., concurring)). By contrast, financial records and telephone records do “ ‘revea[l] . . . personal affairs, opinions, habits and associ- ations.’ ” Miller, 425 U. S., at 451 (Brennan, J., dissent- ing); see Smith, 442 U. S., at 751 (Marshall, J., dissent- ing). What persons purchase and to whom they talk might disclose how much money they make; the political and religious organizations to which they donate; whether they have visited a psychiatrist, plastic surgeon, abortion clinic, or AIDS treatment center; whether they go to gay bars or 18 CARPENTER v. UNITED STATES

straight ones; and who are their closest friends and family members. The troves of intimate information the Gov- ernment can and does obtain using financial records and telephone records dwarfs what can be gathered from cell- site records. Still, the Court maintains, cell-site records are “unique” because they are “comprehensive” in their reach; allow for retrospective collection; are “easy, cheap, and efficient compared to traditional investigative tools”; and are not exposed to cell phone service providers in a meaningfully voluntary manner. Ante, at 1113, 17, 22. But many other kinds of business records can be so described. Fi- nancial records are of vast scope. Banks and credit card companies keep a comprehensive account of almost every transaction an individual makes on a daily basis. “With just the click of a button, the Government can access each [company’s] deep repository of historical [financial] infor- mation at practically no expense.” Ante, at 1213. And the decision whether to transact with banks and credit card companies is no more or less voluntary than the decision whether to use a cell phone. Today, just as when Miller was decided, “ ‘it is impossible to participate in the economic life of contemporary society without maintaining a bank account.’ ” 425 U. S., at 451 (Brennan, J., dissent- ing). But this Court, nevertheless, has held that individ- uals do not have a reasonable expectation of privacy in financial records. Perhaps recognizing the difficulty of drawing the consti- tutional line between cell-site records and financial and telephonic records, the Court posits that the accuracy of cell-site records “is rapidly approaching GPS-level preci- sion.” Ante, at 14. That is certainly plausible in the era of cyber technology, yet the privacy interests associated with location information, which is often disclosed to the public at large, still would not outweigh the privacy interests implicated by financial and telephonic records. Cite as: 585 U. S. ____ (2018) 19

Perhaps more important, those future developments are no basis upon which to resolve this case. In general, the Court “risks error by elaborating too fully on the Fourth Amendment implications of emerging technology before its role in society has become clear.” Ontario v. Quon, 560 U. S. 746, 759 (2010). That judicial caution, prudent in most cases, is imperative in this one. Technological changes involving cell phones have com- plex effects on crime and law enforcement. Cell phones make crimes easier to coordinate and conceal, while also providing the Government with new investigative tools that may have the potential to upset traditional privacy expectations. See Kerr, An Equilibrium-Adjustment Theory of the Fourth Amendment, 125 Harv. L. Rev 476, 512517 (2011). How those competing effects balance against each other, and how property norms and expecta- tions of privacy form around new technology, often will be difficult to determine during periods of rapid technological change. In those instances, and where the governing legal standard is one of reasonableness, it is wise to defer to legislative judgments like the one embodied in §2703(d) of the Stored Communications Act. See Jones, 565 U. S., at 430 (ALITO, J., concurring). In §2703(d) Congress weighed the privacy interests at stake and imposed a judicial check to prevent executive overreach. The Court should be wary of upsetting that legislative balance and erecting constitu- tional barriers that foreclose further legislative instruc- tions. See Quon, supra, at 759. The last thing the Court should do is incorporate an arbitrary and outside limit—in this case six days’ worth of cell-site records—and use it as the foundation for a new constitutional framework. The Court’s decision runs roughshod over the mechanism Congress put in place to govern the acquisition of cell-site records and closes off further legislative debate on these issues. 20 CARPENTER v. UNITED STATES

C The Court says its decision is a “narrow one.” Ante, at 17. But its reinterpretation of Miller and Smith will have dramatic consequences for law enforcement, courts, and society as a whole. Most immediately, the Court’s holding that the Gov- ernment must get a warrant to obtain more than six days of cell-site records limits the effectiveness of an important investigative tool for solving serious crimes. As this case demonstrates, cell-site records are uniquely suited to help the Government develop probable cause to apprehend some of the Nation’s most dangerous criminals: serial killers, rapists, arsonists, robbers, and so forth. See also, e.g., Davis, 785 F. 3d, at 500501 (armed robbers); Brief for Alabama et al. as Amici Curiae 2122 (serial killer). These records often are indispensable at the initial stages of investigations when the Government lacks the evidence necessary to obtain a warrant. See United States v. Pem- brook, 876 F. 3d 812, 816819 (CA6 2017). And the long- term nature of many serious crimes, including serial crimes and terrorism offenses, can necessitate the use of significantly more than six days of cell-site records. The Court’s arbitrary 6-day cutoff has the perverse effect of nullifying Congress’ reasonable framework for obtain- ing cell-site records in some of the most serious criminal investigations. The Court’s decision also will have ramifications that extend beyond cell-site records to other kinds of infor- mation held by third parties, yet the Court fails “to pro- vide clear guidance to law enforcement” and courts on key issues raised by its reinterpretation of Miller and Smith. Riley v. California, 573 U. S. ___, ___ (2014) (slip op., at 22). First, the Court’s holding is premised on cell-site records being a “distinct category of information” from other busi- Cite as: 585 U. S. ____ (2018) 21

ness records. Ante, at 15. But the Court does not explain what makes something a distinct category of information. Whether credit card records are distinct from bank rec- ords; whether payment records from digital wallet applica- tions are distinct from either; whether the electronic bank records available today are distinct from the paper and microfilm records at issue in Miller; or whether cell-phone call records are distinct from the home-phone call records at issue in Smith, are just a few of the difficult questions that require answers under the Court’s novel conception of Miller and Smith. Second, the majority opinion gives courts and law en- forcement officers no indication how to determine whether any particular category of information falls on the finan- cial-records side or the cell-site-records side of its newly conceived constitutional line. The Court’s multifactor analysis—considering intimacy, comprehensiveness, ex- pense, retrospectivity, and voluntariness—puts the law on a new and unstable foundation. Third, even if a distinct category of information is deemed to be more like cell-site records than financial records, courts and law enforcement officers will have to guess how much of that information can be requested before a warrant is required. The Court suggests that less than seven days of location information may not require a warrant. See ante, at 11, n. 3; see also ante, at 1718 (expressing no opinion on “real-time CSLI,” tower dumps, and security-camera footage). But the Court does not explain why that is so, and nothing in its opinion even alludes to the considerations that should determine whether greater or lesser thresholds should apply to in- formation like IP addresses or website browsing history. Fourth, by invalidating the Government’s use of court- approved compulsory process in this case, the Court calls into question the subpoena practices of federal and state grand juries, legislatures, and other investigative bodies, 22 CARPENTER v. UNITED STATES

as JUSTICE ALITO’s opinion explains. See post, at 219 (dissenting opinion). Yet the Court fails even to mention the serious consequences this will have for the proper administration of justice. In short, the Court’s new and uncharted course will inhibit law enforcement and “keep defendants and judges guessing for years to come.” Riley, 573 U. S., at ___ (slip op., at 25) (internal quotation marks omitted). * * * This case should be resolved by interpreting accepted property principles as the baseline for reasonable expecta- tions of privacy. Here the Government did not search anything over which Carpenter could assert ownership or control. Instead, it issued a court-authorized subpoena to a third party to disclose information it alone owned and controlled. That should suffice to resolve this case. Having concluded, however, that the Government searched Carpenter when it obtained cell-site records from his cell phone service providers, the proper resolution of this case should have been to remand for the Court of Appeals to determine in the first instance whether the search was reasonable. Most courts of appeals, believing themselves bound by Miller and Smith, have not grappled with this question. And the Court’s reflexive imposition of the warrant requirement obscures important and difficult issues, such as the scope of Congress’ power to authorize the Government to collect new forms of information using processes that deviate from traditional warrant proce- dures, and how the Fourth Amendment’s reasonableness requirement should apply when the Government uses compulsory process instead of engaging in an actual, physical search. These reasons all lead to this respectful dissent. Cite as: 585 U. S. ____ (2018) 23

KENNEDY Appendix , J., dissenting to opinion of KENNEDY, J.

APPENDIX

Ҥ2703. Required disclosure of customer communi- cations or records

“(d) REQUIREMENTS FOR COURT ORDER.—A court order for disclosure under subsection (b) or (c) may be issued by any court that is a court of competent jurisdiction and shall issue only if the governmental entity offers specific and articulable facts showing that there are reasonable grounds to believe that the contents of a wire or electronic communication, or the records or other information sought, are relevant and material to an ongoing criminal investigation. In the case of a State governmental author- ity, such a court order shall not issue if prohibited by the law of such State. A court issuing an order pursuant to this section, on a motion made promptly by the service provider, may quash or modify such order, if the infor- mation or records requested are unusually voluminous in nature or compliance with such order otherwise would cause an undue burden on such provider.” Cite as: 585 U. S. ____ (2018) 1

THOMAS, J., dissenting

JUSTICE THOMAS, dissenting. This case should not turn on “whether” a search oc­ curred. Ante, at 1. It should turn, instead, on whose property was searched. The Fourth Amendment guaran­ tees individuals the right to be secure from unreasonable searches of “their persons, houses, papers, and effects.” (Emphasis added.) In other words, “each person has the right to be secure against unreasonable searches . . . in his own person, house, papers, and effects.” Minnesota v. Carter, 525 U. S. 83, 92 (1998) (Scalia, J., concurring). By obtaining the cell-site records of MetroPCS and Sprint, the Government did not search Carpenter’s property. He did not create the records, he does not maintain them, he cannot control them, and he cannot destroy them. Neither the terms of his contracts nor any provision of law makes the records his. The records belong to MetroPCS and Sprint. The Court concludes that, although the records are not Carpenter’s, the Government must get a warrant because Carpenter had a reasonable “expectation of privacy” in the location information that they reveal. Ante, at 11. I agree with JUSTICE KENNEDY, JUSTICE ALITO, JUSTICE GORSUCH, and every Court of Appeals to consider the question that this is not the best reading of our precedents. 2 CARPENTER v. UNITED STATES

The more fundamental problem with the Court’s opin­ ion, however, is its use of the “reasonable expectation of privacy” test, which was first articulated by Justice Har­ lan in Katz v. United States, 389 U. S. 347, 360–361 (1967) (concurring opinion). The Katz test has no basis in the text or history of the Fourth Amendment. And, it invites courts to make judgments about policy, not law. Until we confront the problems with this test, Katz will continue to distort Fourth Amendment jurisprudence. I respectfully dissent. I Katz was the culmination of a series of decisions apply­ ing the Fourth Amendment to electronic eavesdropping. The first such decision was Olmstead v. United States, 277 U. S. 438 (1928), where federal officers had intercepted the defendants’ conversations by tapping telephone lines near their homes. Id., at 456–457. In an opinion by Chief Justice Taft, the Court concluded that this wiretap did not violate the Fourth Amendment. No “search” occurred, according to the Court, because the officers did not physi­ cally enter the defendants’ homes. Id., at 464–466. And neither the telephone lines nor the defendants’ intangible conversations qualified as “persons, houses, papers, [or] effects” within the meaning of the Fourth Amendment. Ibid.1 In the ensuing decades, this Court adhered to

—————— 1 Justice Brandeis authored the principal dissent in Olmstead. He

consulted the “underlying purpose,” rather than “the words of the [Fourth] Amendment,” to conclude that the wiretap was a search. 277 U. S., at 476. In Justice Brandeis’ view, the Framers “recognized the significance of man’s spiritual nature, of his feelings and of his intel­ lect” and “sought to protect Americans in their beliefs, their thoughts, their emotions and their sensations.” Id., at 478. Thus, “every unjusti­ fiable intrusion by the Government upon the privacy of the individual, whatever the means employed,” should constitute an unreasonable search under the Fourth Amendment. Ibid. Cite as: 585 U. S. ____ (2018) 3

Olmstead and rejected Fourth Amendment challenges to various methods of electronic surveillance. See On Lee v. United States, 343 U. S. 747, 749–753 (1952) (use of mi­ crophone to overhear conversations with confidential informant); Goldman v. United States, 316 U. S. 129, 131– 132, 135–136 (1942) (use of detectaphone to hear conver­ sations in office next door). In the 1960’s, however, the Court began to retreat from Olmstead. In Silverman v. United States, 365 U. S. 505 (1961), for example, federal officers had eavesdropped on the defendants by driving a “spike mike” several inches into the house they were occupying. Id., at 506–507. This was a “search,” the Court held, because the “unauthorized physical penetration into the premises” was an “actual intrusion into a constitutionally protected area.” Id., at 509, 512. The Court did not mention Olmstead’s other holding that intangible conversations are not “persons, houses, papers, [or] effects.” That omission was signifi­ cant. The Court confirmed two years later that “[i]t fol­ lows from [Silverman] that the Fourth Amendment may protect against the overhearing of verbal statements as well as against the more traditional seizure of ‘papers and effects.’ ” Wong Sun v. United States, 371 U. S. 471, 485 (1963); accord, Berger v. New York, 388 U. S. 41, 51 (1967). In Katz, the Court rejected Olmstead’s remaining hold- ing—that eavesdropping is not a search absent a physical intrusion into a constitutionally protected area. The federal officers in Katz had intercepted the defendant’s conversations by attaching an electronic device to the outside of a public telephone booth. 389 U. S., at 348. The Court concluded that this was a “search” because the officers “violated the privacy upon which [the defendant] justifiably relied while using the telephone booth.” Id., at 353. Although the device did not physically penetrate the booth, the Court overruled Olmstead and held that “the reach of [the Fourth] Amendment cannot turn upon the 4 CARPENTER v. UNITED STATES

presence or absence of a physical intrusion.” 389 U. S., at 353. The Court did not explain what should replace Olmstead’s physical-intrusion requirement. It simply asserted that “the Fourth Amendment protects people, not places” and “what [a person] seeks to preserve as private . . . may be constitutionally protected.” 389 U. S., at 351. Justice Harlan’s concurrence in Katz attempted to artic­ ulate the standard that was missing from the majority opinion. While Justice Harlan agreed that “ ‘the Fourth Amendment protects people, not places,’ ” he stressed that “[t]he question . . . is what protection it affords to those people,” and “the answer . . . requires reference to a ‘place.’ ” Id., at 361. Justice Harlan identified a “twofold requirement” to determine when the protections of the Fourth Amendment apply: “first that a person have exhib­ ited an actual (subjective) expectation of privacy and, second, that the expectation be one that society is pre­ pared to recognize as ‘reasonable.’ ” Ibid. Justice Harlan did not cite anything for this “expecta­ tion of privacy” test, and the parties did not discuss it in their briefs. The test appears to have been presented for the first time at oral argument by one of the defendant’s lawyers. See Winn, Katz and the Origins of the “Reason- able Expectation of Privacy” Test, 40 McGeorge L. Rev. 1, 9–10 (2009). The lawyer, a recent law-school graduate, apparently had an “[e]piphany” while preparing for oral argument. Schneider, Katz v. United States: The Untold Story, 40 McGeorge L. Rev. 13, 18 (2009). He conjectured that, like the “reasonable person” test from his Torts class, the Fourth Amendment should turn on “whether a rea­ sonable person . . . could have expected his communication to be private.” Id., at 19. The lawyer presented his new theory to the Court at oral argument. See, e.g., Tr. of Oral Arg. in Katz v. United States, O. T. 1967, No. 35, p. 5 (proposing a test of “whether or not, objectively speaking, the communication was intended to be private”); id., at 11 Cite as: 585 U. S. ____ (2018) 5

(“We propose a test using a way that’s not too dissimilar from the tort ‘reasonable man’ test”). After some question­ ing from the Justices, the lawyer conceded that his test should also require individuals to subjectively expect privacy. See id., at 12. With that modification, Justice Harlan seemed to accept the lawyer’s test almost verbatim in his concurrence. Although the majority opinion in Katz had little practi­ cal significance after Congress enacted the Omnibus Crime Control and Safe Streets Act of 1968, Justice Har­ lan’s concurrence profoundly changed our Fourth Amend­ ment jurisprudence. It took only one year for the full Court to adopt his two-pronged test. See Terry v. Ohio, 392 U. S. 1, 10 (1968). And by 1979, the Court was de­ scribing Justice Harlan’s test as the “lodestar” for deter­ mining whether a “search” had occurred. Smith v. Mary- land, 442 U. S. 735, 739 (1979). Over time, the Court minimized the subjective prong of Justice Harlan’s test. See Kerr, Katz Has Only One Step: The Irrelevance of Subjective Expectations, 82 U. Chi. L. Rev. 113 (2015). That left the objective prong—the “reasonable expectation of privacy” test that the Court still applies today. See ante, at 5; United States v. Jones, 565 U. S. 400, 406 (2012). II Under the Katz test, a “search” occurs whenever “gov­ ernment officers violate a person’s ‘reasonable expectation of privacy.’ ” Jones, supra, at 406. The most glaring prob­ lem with this test is that it has “no plausible foundation in the text of the Fourth Amendment.” Carter, 525 U. S., at 97 (opinion of Scalia, J.). The Fourth Amendment, as relevant here, protects “[t]he right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches.” By defining “search” to mean “any violation of a reasonable expectation of pri- 6 CARPENTER v. UNITED STATES

vacy,” the Katz test misconstrues virtually every one of these words. A The Katz test distorts the original meaning of “searc[h]”—the word in the Fourth Amendment that it purports to define, see ante, at 5; Smith, supra. Under the Katz test, the government conducts a search anytime it violates someone’s “reasonable expectation of privacy.” That is not a normal definition of the word “search.” At the founding, “search” did not mean a violation of someone’s reasonable expectation of privacy. The word was probably not a term of art, as it does not appear in legal dictionaries from the era. And its ordinary meaning was the same as it is today: “ ‘[t]o look over or through for the purpose of finding something; to explore; to examine by inspection; as, to search the house for a book; to search the wood for a thief.’ ” Kyllo v. United States, 533 U. S. 27, 32, n. 1 (2001) (quoting N. Webster, An American Diction­ ary of the English Language 66 (1828) (reprint 6th ed. 1989)); accord, 2 S. Johnson, A Dictionary of the English Language (5th ed. 1773) (“Inquiry by looking into every suspected place”); N. Bailey, An Universal Etymological English Dictionary (22d ed. 1770) (“a seeking after, a looking for, &c.”); 2 J. Ash, The New and Complete Dic­ tionary of the English Language (2d ed. 1795) (“An en­ quiry, an examination, the act of seeking, an enquiry by looking into every suspected place; a quest; a pursuit”); T. Sheridan, A Complete Dictionary of the English Language (6th ed. 1796) (similar). The word “search” was not asso­ ciated with “reasonable expectation of privacy” until Jus­ tice Harlan coined that phrase in 1967. The phrase “ex­ pectation(s) of privacy” does not appear in the pre-Katz federal or state case reporters, the papers of prominent Cite as: 585 U. S. ____ (2018) 7

Founders,2 early congressional documents and debates,3 collections of early American English texts,4 or early American newspapers.5 B The Katz test strays even further from the text by focus­ ing on the concept of “privacy.” The word “privacy” does not appear in the Fourth Amendment (or anywhere else in the Constitution for that matter). Instead, the Fourth Amendment references “[t]he right of the people to be secure.” It then qualifies that right by limiting it to “per­ sons” and three specific types of property: “houses, papers, and effects.” By connecting the right to be secure to these four specific objects, “[t]he text of the Fourth Amendment reflects its close connection to property.” Jones, supra, at 405. “[P]rivacy,” by contrast, “was not part of the political vocabulary of the [founding]. Instead, liberty and privacy rights were understood largely in terms of property rights.” Cloud, Property Is Privacy: Locke and Brandeis in the Twenty-First Century, 55 Am. Crim. L. Rev. 37, 42 (2018). Those who ratified the Fourth Amendment were quite familiar with the notion of security in property. Security in property was a prominent concept in English law. See, e.g., 3 W. Blackstone, Commentaries on the Laws of Eng-

—————— 2 National Archives, Library of Congress, Founders Online, https:// founders.archives.gov (all Internet materials as last visited June 18, 2018). 3 A Century of Lawmaking For A New Nation, U. S. Congressional

Documents and Debates, 1774–1875 (May 1, 2003), https://memory.loc .gov/ammem/amlaw/lawhome.html. 4 Corpus of Historical American English, https://corpus.byu.edu/coha;

Google Books (American), https://googlebooks.byu.edu/x.asp; Corpus of Founding Era American English, https://lawncl.byu.edu/cofea. 5 Readex, America’s Historical Newspapers (2018), https:// www.readex.com/content/americas-historical-newspapers. 8 CARPENTER v. UNITED STATES

land 288 (1768) (“[E]very man’s house is looked upon by the law to be his castle”); 3 E. Coke, Institutes of Laws of England 162 (6th ed. 1680) (“[F]or a man[’]s house is his Castle, & domus sua cuique est tutissimum refugium [each man’s home is his safest refuge]”). The political philosophy of John Locke, moreover, “permeated the 18th­ century political scene in America.” Obergefell v. Hodges, 576 U. S. ___, ___ (2015) (THOMAS, J., dissenting) (slip op., at 8). For Locke, every individual had a property right “in his own person” and in anything he “removed from the common state [of] Nature” and “mixed his labour with.” Second Treatise of Civil Government §27 (1690). Because property is “very unsecure” in the state of nature, §123, individuals form governments to obtain “a secure enjoy­ ment of their properties.” §95. Once a government is formed, however, it cannot be given “a power to destroy that which every one designs to secure”; it cannot legiti­ mately “endeavour to take away, and destroy the property of the people,” or exercise “an absolute power over [their] lives, liberties, and estates.” §222. The concept of security in property recognized by Locke and the English legal tradition appeared throughout the materials that inspired the Fourth Amendment. In Entick v. Carrington, 19 How. St. Tr. 1029 (C. P. 1765)—a her­ alded decision that the founding generation considered “the true and ultimate expression of constitutional law,” Boyd v. United States, 116 U. S. 616, 626 (1886)—Lord Camden explained that “[t]he great end, for which men entered into society, was to secure their property.” 19 How. St. Tr., at 1066. The American colonists echoed this reasoning in their “widespread hostility” to the Crown’s writs of assistance6—a practice that inspired the Revolu­

—————— 6 Writs of assistance were “general warrants” that gave “customs

officials blanket authority to search where they pleased for goods Cite as: 585 U. S. ____ (2018) 9

tion and became “[t]he driving force behind the adoption of the [Fourth] Amendment.” United States v. Verdugo- Urquidez, 494 U. S. 259, 266 (1990). Prominent colonists decried the writs as destroying “ ‘domestic security’ ” by permitting broad searches of homes. M. Smith, The Writs of Assistance Case 475 (1978) (quoting a 1772 Boston town meeting); see also id., at 562 (complaining that “ ‘every householder in this province, will necessarily become less secure than he was before this writ’ ” (quoting a 1762 article in the Boston Gazette)); id., at 493 (complaining that the writs were “ ‘expressly contrary to the common law, which ever regarded a man’s house as his castle, or a place of perfect security’ ” (quoting a 1768 letter from John Dickinson)). John Otis, who argued the famous Writs of Assistance case, contended that the writs violated “ ‘the fundamental Principl[e] of Law’ ” that “ ‘[a] Man who is quiet, is as secure in his House, as a Prince in his Castle.’ ” Id., at 339 (quoting John Adam’s notes). John Adams attended Otis’ argument and later drafted Article XIV of the Massachusetts Constitution,7 which served as a model for the Fourth Amendment. See Clancy, The Framers’ Intent: John Adams, His Era, and the Fourth Amendment, 86 Ind. L. J. 979, 982 (2011); Donahue, The Original Fourth Amendment, 83 U. Chi. L. Rev. 1181, 1269 (2016)

——————

imported in violation of the British tax laws.” Stanford v. Texas, 379

U. S. 476, 481 (1965). 7 “Every subject has a right to be secure from all unreasonable

searches and seizures of his person, his house, his papers, and all his possessions. All warrants, therefore, are contrary to right, if the cause or foundation of them be not previously supported by oath or affirma­ tion, and if the order in the warrant to a civil officer, to make search in suspected places, or to arrest one or more suspected persons, or to seize their property, be not accompanied with a special designation of the person or objects of search, arrest, or seizure; and no warrant ought to be issued but in cases, and with the formalities prescribed by the laws.” Mass. Const., pt. I, Art. XIV (1780). 10 CARPENTER v. UNITED STATES

(Donahue). Adams agreed that “[p]roperty must be se­ cured, or liberty cannot exist.” Discourse on Davila, in 6 The Works of John Adams 280 (C. Adams ed. 1851). Of course, the founding generation understood that, by securing their property, the Fourth Amendment would often protect their privacy as well. See, e.g., Boyd, supra, at 630 (explaining that searches of houses invade “the privacies of life”); Wilkes v. Wood, 19 How. St. Tr. 1153, 1154 (C. P. 1763) (argument of counsel contending that seizures of papers implicate “our most private concerns”). But the Fourth Amendment’s attendant protection of privacy does not justify Katz’s elevation of privacy as the sine qua non of the Amendment. See T. Clancy, The Fourth Amendment: Its History and Interpretation §3.4.4, p. 78 (2008) (“[The Katz test] confuse[s] the reasons for exercising the protected right with the right itself. A purpose of exercising one’s Fourth Amendment rights might be the desire for privacy, but the individual’s motiva­ tion is not the right protected”); cf. United States v. Gonzalez- Lopez, 548 U. S. 140, 145 (2006) (rejecting “a line of reasoning that ‘abstracts from the right to its purposes, and then eliminates the right’ ”). As the majority opinion in Katz recognized, the Fourth Amendment “cannot be translated into a general constitutional ‘right to privacy,’ ” as its protections “often have nothing to do with privacy at all.” 389 U. S., at 350. Justice Harlan’s focus on privacy in his concurrence—an opinion that was issued between Griswold v. Connecticut, 381 U. S. 479 (1965), and Roe v. Wade, 410 U. S. 113 (1973)—reflects privacy’s status as the organizing constitutional idea of the 1960’s and 1970’s. The organizing constitutional idea of the founding era, by contrast, was property. C In shifting the focus of the Fourth Amendment from property to privacy, the Katz test also reads the words Cite as: 585 U. S. ____ (2018) 11

“persons, houses, papers, and effects” out of the text. At its broadest formulation, the Katz test would find a search “wherever an individual may harbor a reasonable ‘expecta­ tion of privacy.’ ” Terry, 392 U. S., at 9 (emphasis added). The Court today, for example, does not ask whether cell- site location records are “persons, houses, papers, [or] effects” within the meaning of the Fourth Amendment.8 Yet “persons, houses, papers, and effects” cannot mean “anywhere” or “anything.” Katz’s catchphrase that “the Fourth Amendment protects people, not places,” is not a serious attempt to reconcile the constitutional text. See Carter, 525 U. S., at 98, n. 3 (opinion of Scalia, J.). The Fourth Amendment obviously protects people; “[t]he ques­ tion . . . is what protection it affords to those people.” Katz, 389 U. S., at 361 (Harlan, J., concurring). The Founders decided to protect the people from unreasonable searches and seizures of four specific things—persons, houses, papers, and effects. They identified those four categories as “the objects of privacy protection to which the Constitution would extend, leaving further expansion to the good judgment . . . of the people through their rep­ resentatives in the legislature.” Carter, supra, at 97–98 (opinion of Scalia, J.). This limiting language was important to the founders. Madison’s first draft of the Fourth Amendment used a different phrase: “their persons, their houses, their papers, and their other property.” 1 Annals of Cong. 452 (1789)

—————— 8 The answer to that question is not obvious. Cell-site location rec­ ords are business records that mechanically collect the interactions between a person’s cell phone and the company’s towers; they are not private papers and do not reveal the contents of any communications. Cf. Schnapper, Unreasonable Searches and Seizures of Papers, 71 Va. L. Rev. 869, 923–924 (1985) (explaining that business records that do not reveal “personal or speech-related confidences” might not satisfy the original meaning of “papers”). 12 CARPENTER v. UNITED STATES

(emphasis added). In one of the few changes made to Madison’s draft, the House Committee of Eleven changed “other property” to “effects.” See House Committee of Eleven Report (July 28, 1789), in N. Cogan, The Complete Bill of Rights 334 (2d ed. 2015). This change might have narrowed the Fourth Amendment by clarifying that it does not protect real property (other than houses). See Oliver v. United States, 466 U. S. 170, 177, and n. 7 (1984); Da­ vies, Recovering the Original Fourth Amendment, 98 Mich. L. Rev. 547, 709–714 (1999) (Davies). Or the change might have broadened the Fourth Amendment by clarify­ ing that it protects commercial goods, not just personal possessions. See Donahue 1301. Or it might have done both. Whatever its ultimate effect, the change reveals that the Founders understood the phrase “persons, houses, papers, and effects” to be an important measure of the Fourth Amendment’s overall scope. See Davies 710. The Katz test, however, displaces and renders that phrase entirely “superfluous.” Jones, 565 U. S., at 405. D “[P]ersons, houses, papers, and effects” are not the only words that the Katz test reads out of the Fourth Amend­ ment. The Fourth Amendment specifies that the people have a right to be secure from unreasonable searches of “their” persons, houses, papers, and effects. Although phrased in the plural, “[t]he obvious meaning of [‘their’] is that each person has the right to be secure against unrea­ sonable searches and seizures in his own person, house, papers, and effects.” Carter, supra, at 92 (opinion of Sca- lia, J.); see also District of Columbia v. Heller, 554 U. S. 570, 579 (2008) (explaining that the Constitution uses the plural phrase “the people” to “refer to individual rights, not ‘collective’ rights”). Stated differently, the word “their” means, at the very least, that individuals do not have Fourth Amendment rights in someone else’s property. See Cite as: 585 U. S. ____ (2018) 13

Carter, supra, at 92–94 (opinion of Scalia, J.). Yet, under the Katz test, individuals can have a reasonable expecta­ tion of privacy in another person’s property. See, e.g., Carter, 525 U. S., at 89 (majority opinion) (“[A] person may have a legitimate expectation of privacy in the house of someone else”). Until today, our precedents have not acknowledged that individuals can claim a reasonable expectation of privacy in someone else’s business records. See ante, at 2 (KENNEDY, J., dissenting). But the Court erases that line in this case, at least for cell-site location records. In doing so, it confirms that the Katz test does not necessarily require an individual to prove that the government searched his person, house, paper, or effect. Carpenter attempts to argue that the cell-site records are, in fact, his “papers,” see Brief for Petitioner 32–35; Reply Brief 14–15, but his arguments are unpersuasive, see ante, at 12–13 (opinion of KENNEDY, J.); post, at 20–23 (ALITO, J., dissenting). Carpenter stipulated below that the cell-site records are the business records of Sprint and MetroPCS. See App. 51. He cites no property law in his briefs to this Court, and he does not explain how he has a property right in the companies’ records under the law of any jurisdiction at any point in American history. If someone stole these records from Sprint or MetroPCS, Carpenter does not argue that he could recover in a tradi­ tional tort action. Nor do his contracts with Sprint and MetroPCS make the records his, even though such provi­ sions could exist in the marketplace. Cf., e.g., Google Terms of Service, https://policies.google.com/terms (“Some of our Services allow you to upload, submit, store, send or receive content. You retain ownership of any intellectual property rights that you hold in that content. In short, what belongs to you stays yours”). Instead of property, tort, or contract law, Carpenter relies on the federal Telecommunications Act of 1996 to demonstrate that the cell site records are his papers. The 14 CARPENTER v. UNITED STATES

Telecommunications Act generally bars cell-phone compa­ nies from disclosing customers’ cell site location infor­ mation to the public. See 47 U. S. C. §222(c). This is sufficient to make the records his, Carpenter argues, because the Fourth Amendment merely requires him to identify a source of “positive law” that “protects against access by the public without consent.” Brief for Petitioner 32–33 (citing Baude & Stern, The Positive Law Model of the Fourth Amendment, 129 Harv. L. Rev. 1821, 1825– 1826 (2016); emphasis deleted). Carpenter is mistaken. To come within the text of the Fourth Amendment, Carpenter must prove that the cell- site records are his; positive law is potentially relevant only insofar as it answers that question. The text of the Fourth Amendment cannot plausibly be read to mean “any violation of positive law” any more than it can plausibly be read to mean “any violation of a reasonable expectation of privacy.” Thus, the Telecommunications Act is insufficient be­ cause it does not give Carpenter a property right in the cell-site records. Section 222, titled “Privacy of customer information,” protects customers’ privacy by preventing cell-phone companies from disclosing sensitive information about them. The statute creates a “duty to protect the confidentiality” of information relating to customers, §222(a), and creates “[p]rivacy requirements” that limit the disclosure of that information, §222(c)(1). Nothing in the text pre-empts state property law or gives customers a property interest in the companies’ business records (as­ suming Congress even has that authority).9 Although —————— 9 Carpenter relies on an order from the Federal Communications

Commission (FCC), which weakly states that “ ‘[t]o the extent [a cus­ tomer’s location information] is property, . . . it is better understood as belonging to the customer, not the carrier.’ ” Brief for Petitioner 34, and n. 23 (quoting 13 FCC Rcd. 8061, 8093 ¶43 (1998); emphasis added). Cite as: 585 U. S. ____ (2018) 15

§222 “protects the interests of individuals against wrong­ ful uses or disclosures of personal data, the rationale for these legal protections has not historically been grounded on a perception that people have property rights in per­ sonal data as such.” Samuelson, Privacy as Intellectual Property? 52 Stan. L. Rev. 1125, 1130–1131 (2000) (foot­ note omitted). Any property rights remain with the companies. E The Katz test comes closer to the text of the Fourth Amendment when it asks whether an expectation of pri- vacy is “reasonable,” but it ultimately distorts that term as well. The Fourth Amendment forbids “unreasonable searches.” In other words, reasonableness determines the legality of a search, not “whether a search . . . within the meaning of the Constitution has occurred.” Carter, 525 U. S., at 97 (opinion of Scalia, J.) (internal quotation marks omitted). Moreover, the Katz test invokes the concept of reason- ableness in a way that would be foreign to the ratifiers of the Fourth Amendment. Originally, the word “unreason- able” in the Fourth Amendment likely meant “against reason”—as in “against the reason of the common law.” See Donahue 1270–1275; Davies 686–693; California v. Acevedo, 500 U. S. 565, 583 (1991) (Scalia, J., concurring in judgment). At the founding, searches and seizures were

—————— But this order was vacated by the Court of Appeals for the Tenth Circuit. U. S. West, Inc. v. FCC, 182 F. 3d 1224, 1240 (1999). Notably, the carrier in that case argued that the FCC’s regulation of customer information was a taking of its property. See id., at 1230. Although the panel majority had no occasion to address this argument, see id., at 1239, n. 14, the dissent concluded that the carrier had failed to prove the information was “property” at all, see id., at 1247–1248 (opinion of Briscoe, J.). 16 CARPENTER v. UNITED STATES

regulated by a robust body of common-law rules. See generally W. Cuddihy, The Fourth Amendment: Origins and Original Meaning 602–1791 (2009); e.g., Wilson v. Arkansas, 514 U. S. 927, 931–936 (1995) (discussing the common-law knock-and-announce rule). The search-and­ seizure practices that the Founders feared most—such as general warrants—were already illegal under the common law, and jurists such as Lord Coke described violations of the common law as “against reason.” See Donahue 1270– 1271, and n. 513. Locke, Blackstone, Adams, and other influential figures shortened the phrase “against reason” to “unreasonable.” See id., at 1270–1275. Thus, by pro­ hibiting “unreasonable” searches and seizures in the Fourth Amendment, the Founders ensured that the newly created Congress could not use legislation to abolish the established common-law rules of search and seizure. See T. Cooley, Constitutional Limitations *303 (2d ed. 1871); 3 J. Story, Commentaries on the Constitution of the United States §1895, p. 748 (1833). Although the Court today maintains that its decision is based on “Founding-era understandings,” ante, at 6, the Founders would be puzzled by the Court’s conclusion as well as its reasoning. The Court holds that the Govern­ ment unreasonably searched Carpenter by subpoenaing the cell-site records of Sprint and MetroPCS without a warrant. But the Founders would not recognize the Court’s “warrant requirement.” Ante, at 21. The common law required warrants for some types of searches and seizures, but not for many others. The relevant rule de­ pended on context. See Acevedo, supra, at 583–584 (opin­ ion of Scalia, J.); Amar, Fourth Amendment First Princi­ ples, 107 Harv. L. Rev. 757, 763–770 (1994); Davies 738– 739. In cases like this one, a subpoena for third-party documents was not a “search” to begin with, and the com­ mon law did not limit the government’s authority to sub­ poena third parties. See post, at 2–12 (ALITO, J., dissent­ Cite as: 585 U. S. ____ (2018) 17

ing). Suffice it to say, the Founders would be confused by this Court’s transformation of their common-law protec­ tion of property into a “warrant requirement” and a vague inquiry into “reasonable expectations of privacy.” III That the Katz test departs so far from the text of the Fourth Amendment is reason enough to reject it. But the Katz test also has proved unworkable in practice. Jurists and commentators tasked with deciphering our jurispru­ dence have described the Katz regime as “an unpredictable jumble,” “a mass of contradictions and obscurities,” “all over the map,” “riddled with inconsistency and incoher­ ence,” “a series of inconsistent and bizarre results that [the Court] has left entirely undefended,” “unstable,” “chameleon-like,” “ ‘notoriously unhelpful,’ ” “a conclusion rather than a starting point for analysis,” “distressingly unmanageable,” “a dismal failure,” “flawed to the core,” “unadorned fiat,” and “inspired by the kind of logic that produced Rube Goldberg’s bizarre contraptions.”10 Even —————— 10 Kugler & Strahilevitz, Actual Expectations of Privacy, Fourth Amendment Doctrine, and the Mosaic Theory, 2015 S. Ct. Rev. 205, 261; Bradley, Two Models of the Fourth Amendment, 83 Mich. L. Rev. 1468 (1985); Kerr, Four Models of Fourth Amendment Protection, 60 Stan. L. Rev. 503, 505 (2007); Solove, Fourth Amendment Pragmatism, 51 Boston College L. Rev. 1511 (2010); Wasserstom & Seidman, The Fourth Amendment as Constitutional Theory, 77 Geo. L. J. 19, 29 (1988); Colb, What Is a Search? Two Conceptual Flaws in Fourth Amendment Doctrine and Some Hints of a Remedy, 55 Stan. L. Rev. 119, 122 (2002); Clancy, The Fourth Amendment: Its History and Interpretation §3.3.4, p. 65 (2008); Minnesota v. Carter, 525 U. S. 83, 97 (1998) (Scalia, J., dissenting); State v. Campbell, 306 Ore. 157, 164, 759 P. 2d 1040, 1044 (1988); Wilkins, Defining the “Reasonable Expectation of Privacy”: an Emerging Tripartite Analysis, 40 Vand. L. Rev. 1077, 1107 (1987); Yeager, Search, Seizure and the Positive Law: Expecta­ tions of Privacy Outside the Fourth Amendment, 84 J. Crim. L. & C. 249, 251 (1993); Thomas, Time Travel, Hovercrafts, and the Framers: 18 CARPENTER v. UNITED STATES

Justice Harlan, four years after penning his concurrence in Katz, confessed that the test encouraged “the substitu­ tion of words for analysis.” United States v. White, 401 U. S. 745, 786 (1971) (dissenting opinion). After 50 years, it is still unclear what question the Katz test is even asking. This Court has steadfastly declined to elaborate the relevant considerations or identify any meaningful constraints. See, e.g., ante, at 5 (“[N]o single rubric definitively resolves which expectations of privacy are entitled to protection”); O’Connor v. Ortega, 480 U. S. 709, 715 (1987) (plurality opinion) (“We have no talisman that determines in all cases those privacy expectations that society is prepared to accept as reasonable”); Oliver, 466 U. S., at 177 (“No single factor determines whether an individual legitimately may claim under the Fourth Amendment that a place should be free of government intrusion”). Justice Harlan’s original formulation of the Katz test appears to ask a descriptive question: Whether a given expectation of privacy is “one that society is prepared to recognize as ‘reasonable.’ ” 389 U. S., at 361. As written, the Katz test turns on society’s actual, current views about the reasonableness of various expectations of privacy. But this descriptive understanding presents several problems. For starters, it is easily circumvented. If, for example, “the Government were suddenly to announce on nationwide television that all homes henceforth would be subject to warrantless entry,” individuals could not realis­ tically expect privacy in their homes. Smith, 442 U. S., at 740, n. 5; see also Chemerinsky, Rediscovering Brandeis’s

James Madison Sees the Future and Rewrites the Fourth Amendment,

80 Notre Dame L. Rev. 1451, 1500 (2005); Rakas v. Illinois, 439 U. S.

128, 165 (1978) (White, J., dissenting); Cloud, Rube Goldberg Meets the

Constitution: The Supreme Court, Technology, and the Fourth

Amendment, 72 Miss. L. J. 5, 7 (2002).

Right to Privacy, 45 Brandeis L. J. 643, 650 (2007) (“[Un­ der Katz, t]he government seemingly can deny privacy just by letting people know in advance not to expect any”). A purely descriptive understanding of the Katz test also risks “circular[ity].” Kyllo, 533 U. S., at 34. While this Court is supposed to base its decisions on society’s expec­ tations of privacy, society’s expectations of privacy are, in turn, shaped by this Court’s decisions. See Posner, The Uncertain Protection of Privacy by the Supreme Court, 1979 S. Ct. Rev. 173, 188 (“[W]hether [a person] will or will not have [a reasonable] expectation [of privacy] will depend on what the legal rule is”). To address this circularity problem, the Court has in­ sisted that expectations of privacy must come from outside its Fourth Amendment precedents, “either by reference to concepts of real or personal property law or to understand­ ings that are recognized and permitted by society.” Rakas v. Illinois, 439 U. S. 128, 144, n. 12 (1978). But the Court’s supposed reliance on “real or personal property law” rings hollow. The whole point of Katz was to “ ‘dis­ credi[t]’ ” the relationship between the Fourth Amendment and property law, 389 U. S., at 353, and this Court has repeatedly downplayed the importance of property law under the Katz test, see, e.g., United States v. Salvucci, 448 U. S. 83, 91 (1980) (“[P]roperty rights are neither the beginning nor the end of this Court’s inquiry [under Katz]”); Rawlings v. Kentucky, 448 U. S. 98, 105 (1980) (“[This Court has] emphatically rejected the notion that ‘arcane’ concepts of property law ought to control the ability to claim the protections of the Fourth Amend­ ment”). Today, for example, the Court makes no mention of property law, except to reject its relevance. See ante, at 5, and n. 1. As for “understandings that are recognized or permitted in society,” this Court has never answered even the most basic questions about what this means. See Kerr, Four 20 CARPENTER v. UNITED STATES

Models of Fourth Amendment Protection, 60 Stan. L. Rev. 503, 504–505 (2007). For example, our precedents do not explain who is included in “society,” how we know what they “recogniz[e] or permi[t],” and how much of society must agree before something constitutes an “understanding.” Here, for example, society might prefer a balanced regime that prohibits the Government from obtaining cell- site location information unless it can persuade a neutral magistrate that the information bears on an ongoing criminal investigation. That is precisely the regime Con­ gress created under the Stored Communications Act and Telecommunications Act. See 47 U. S. C. §222(c)(1); 18 U. S. C. §§2703(c)(1)(B), (d). With no sense of irony, the Court invalidates this regime today—the one that society actually created “in the form of its elected representatives in Congress.” 819 F. 3d 880, 890 (2016). Truth be told, this Court does not treat the Katz test as a descriptive inquiry. Although the Katz test is phrased in descriptive terms about society’s views, this Court treats it like a normative question—whether a particular practice should be considered a search under the Fourth Amend­ ment. Justice Harlan thought this was the best way to understand his test. See White, 401 U. S., at 786 (dissent­ ing opinion) (explaining that courts must assess the “de­ sirability” of privacy expectations and ask whether courts “should” recognize them by “balanc[ing]” the “impact on the individual’s sense of security . . . against the utility of the conduct as a technique of law enforcement”). And a normative understanding is the only way to make sense of this Court’s precedents, which bear the hallmarks of sub­ jective policymaking instead of neutral legal decisionmak­ ing. “[T]he only thing the past three decades have estab­ lished about the Katz test” is that society’s expectations of privacy “bear an uncanny resemblance to those expecta­ tions of privacy that this Court considers reasonable.” Cite as: 585 U. S. ____ (2018) 21

Carter, 525 U. S., at 97 (opinion of Scalia, J.). Yet, “[t]hough we know ourselves to be eminently reasonable, self-awareness of eminent reasonableness is not really a substitute for democratic election.” Sosa v. Alvarez- Machain, 542 U. S. 692, 750 (2004) (Scalia, J., concurring in part and concurring in judgment). * * * In several recent decisions, this Court has declined to apply the Katz test because it threatened to narrow the original scope of the Fourth Amendment. See Grady v. North Carolina, 575 U. S. ___, ___ (2015) (per curiam) (slip op., at 3); Florida v. Jardines, 569 U. S. 1, 5 (2013); Jones, 565 U. S., at 406–407. But as today’s decision demon­ strates, Katz can also be invoked to expand the Fourth Amendment beyond its original scope. This Court should not tolerate errors in either direction. “The People, through ratification, have already weighed the policy tradeoffs that constitutional rights entail.” Luis v. United States, 578 U. S. ___, ___ (2016) (THOMAS, J., concurring in judgment) (slip op., at 10). Whether the rights they rati­ fied are too broad or too narrow by modern lights, this Court has no authority to unilaterally alter the document they approved. Because the Katz test is a failed experiment, this Court is dutybound to reconsider it. Until it does, I agree with my dissenting colleagues’ reading of our precedents. Accordingly, I respectfully dissent. Cite as: 585 U. S. ____ (2018) 1

ALITO, J., dissenting

JUSTICE ALITO, with whom JUSTICE THOMAS joins, dissenting. I share the Court’s concern about the effect of new tech- nology on personal privacy, but I fear that today’s decision will do far more harm than good. The Court’s reasoning fractures two fundamental pillars of Fourth Amendment law, and in doing so, it guarantees a blizzard of litigation while threatening many legitimate and valuable investiga- tive practices upon which law enforcement has rightfully come to rely. First, the Court ignores the basic distinction between an actual search (dispatching law enforcement officers to enter private premises and root through private papers and effects) and an order merely requiring a party to look through its own records and produce specified documents. The former, which intrudes on personal privacy far more deeply, requires probable cause; the latter does not. Treating an order to produce like an actual search, as today’s decision does, is revolutionary. It violates both the original understanding of the Fourth Amendment and more than a century of Supreme Court precedent. Unless it is somehow restricted to the particular situation in the present case, the Court’s move will cause upheaval. Must every grand jury subpoena duces tecum be supported by probable cause? If so, investigations of terrorism, political 2 CARPENTER v. UNITED STATES

corruption, white-collar crime, and many other offenses will be stymied. And what about subpoenas and other document-production orders issued by administrative agencies? See, e.g., 15 U. S. C. §57b–1(c) (Federal Trade Commission); §§77s(c), 78u(a)–(b) (Securities and Ex- change Commission); 29 U. S. C. §657(b) (Occupational Safety and Health Administration); 29 CFR §1601.16(a)(2) (2017) (Equal Employment Opportunity Commission). Second, the Court allows a defendant to object to the search of a third party’s property. This also is revolution- ary. The Fourth Amendment protects “[t]he right of the people to be secure in their persons, houses, papers, and effects” (emphasis added), not the persons, houses, papers, and effects of others. Until today, we have been careful to heed this fundamental feature of the Amendment’s text. This was true when the Fourth Amendment was tied to property law, and it remained true after Katz v. United States, 389 U. S. 347 (1967), broadened the Amendment’s reach. By departing dramatically from these fundamental principles, the Court destabilizes long-established Fourth Amendment doctrine. We will be making repairs—or picking up the pieces—for a long time to come. I Today the majority holds that a court order requiring the production of cell-site records may be issued only after the Government demonstrates probable cause. See ante, at 18. That is a serious and consequential mistake. The Court’s holding is based on the premise that the order issued in this case was an actual “search” within the meaning of the Fourth Amendment, but that premise is inconsistent with the original meaning of the Fourth Amendment and with more than a century of precedent. Cite as: 585 U. S. ____ (2018) 3

The order in this case was the functional equivalent of a subpoena for documents, and there is no evidence that these writs were regarded as “searches” at the time of the founding. Subpoenas duces tecum and other forms of compulsory document production were well known to the founding generation. Blackstone dated the first writ of subpoena to the reign of King Richard II in the late 14th century, and by the end of the 15th century, the use of such writs had “become the daily practice of the [Chan- cery] court.” 3 W. Blackstone, Commentaries on the Laws of England 53 (G. Tucker ed. 1803) (Blackstone). Over the next 200 years, subpoenas would grow in prominence and power in tandem with the Court of Chancery, and by the end of Charles II’s reign in 1685, two important innova- tions had occurred. First, the Court of Chancery developed a new species of subpoena. Until this point, subpoenas had been used largely to compel attendance and oral testimony from witnesses; these subpoenas correspond to today’s subpoe- nas ad testificandum. But the Court of Chancery also improvised a new version of the writ that tacked onto a regular subpoena an order compelling the witness to bring certain items with him. By issuing these so-called sub- poenas duces tecum, the Court of Chancery could compel the production of papers, books, and other forms of physi- cal evidence, whether from the parties to the case or from third parties. Such subpoenas were sufficiently common- place by 1623 that a leading treatise on the practice of law could refer in passing to the fee for a “Sub pœna of Ducas tecum” (seven shillings and two pence) without needing to elaborate further. T. Powell, The Attourneys Academy 79 (1623). Subpoenas duces tecum would swell in use over the next century as the rules for their application became ever more developed and definite. See, e.g., 1 G. Jacob, The Compleat Chancery-Practiser 290 (1730) (“The Sub- 4 CARPENTER v. UNITED STATES

poena duces tecum is awarded when the Defendant has confessed by his Answer that he hath such Writings in his Hands as are prayed by the Bill to be discovered or brought into Court”). Second, although this new species of subpoena had its origins in the Court of Chancery, it soon made an appear- ance in the work of the common-law courts as well. One court later reported that “[t]he Courts of Common law . . . employed the same or similar means . . . from the time of Charles the Second at least.” Amey v. Long, 9 East. 473, 484, 103 Eng. Rep. 653, 658 (K. B. 1808). By the time Blackstone published his Commentaries on the Laws of England in the 1760’s, the use of subpoenas duces tecum had bled over substantially from the courts of equity to the common-law courts. Admittedly, the transi- tion was still incomplete: In the context of jury trials, for example, Blackstone complained about “the want of a compulsive power for the production of books and papers belonging to the parties.” Blackstone 381; see also, e.g., Entick v. Carrington, 19 State Trials 1029, 1073 (K. B. 1765) (“I wish some cases had been shewn, where the law forceth evidence out of the owner’s custody by process. [But] where the adversary has by force or fraud got pos- session of your own proper evidence, there is no way to get it back but by action”). But Blackstone found some com- fort in the fact that at least those documents “[i]n the hands of third persons . . . can generally be obtained by rule of court, or by adding a clause of requisition to the writ of subpoena, which is then called a subpoena duces tecum.” Blackstone 381; see also, e.g., Leeds v. Cook, 4 Esp. 256, 257, 170 Eng. Rep. 711 (N. P. 1803) (third-party subpoena duces tecum); Rex v. Babb, 3 T. R. 579, 580, 100 Eng. Rep. 743, 744 (K. B. 1790) (third-party document production). One of the primary questions outstanding, then, was whether common-law courts would remedy the “defect[s]” identified by the Commentaries, and allow Cite as: 585 U. S. ____ (2018) 5

parties to use subpoenas duces tecum not only with respect to third parties but also with respect to each other. Black- stone 381. That question soon found an affirmative answer on both sides of the Atlantic. In the United States, the First Con- gress established the federal court system in the Judiciary Act of 1789. As part of that Act, Congress authorized “all the said courts of the United States . . . in the trial of actions at law, on motion and due notice thereof being given, to require the parties to produce books or writings in their possession or power, which contain evidence per- tinent to the issue, in cases and under circumstances where they might be compelled to produce the same by the ordinary rules of proceeding in chancery.” §15, 1 Stat. 82. From that point forward, federal courts in the United States could compel the production of documents regard- less of whether those documents were held by parties to the case or by third parties. In Great Britain, too, it was soon definitively estab- lished that common-law courts, like their counterparts in equity, could subpoena documents held either by parties to the case or by third parties. After proceeding in fits and starts, the King’s Bench eventually held in Amey v. Long that the “writ of subpœna duces tecum [is] a writ of com- pulsory obligation and effect in the law.” 9 East., at 486, 103 Eng. Rep., at 658. Writing for a unanimous court, Lord Chief Justice Ellenborough explained that “[t]he right to resort to means competent to compel the produc- tion of written, as well as oral, testimony seems essential to the very existence and constitution of a Court of Com- mon Law.” Id., at 484, 103 Eng. Rep., at 658. Without the power to issue subpoenas duces tecum, the Lord Chief Justice observed, common-law courts “could not possibly proceed with due effect.” Ibid. The prevalence of subpoenas duces tecum at the time of the founding was not limited to the civil context. In crim- 6 CARPENTER v. UNITED STATES

inal cases, courts and prosecutors were also using the writ to compel the production of necessary documents. In Rex v. Dixon, 3 Burr. 1687, 97 Eng. Rep. 1047 (K. B. 1765), for example, the King’s Bench considered the propriety of a subpoena duces tecum served on an attorney named Sam- uel Dixon. Dixon had been called “to give evidence before the grand jury of the county of Northampton” and specifi- cally “to produce three vouchers . . . in order to found a prosecution by way of indictment against [his client] Peach . . . for forgery.” Id., at 1687, 97 Eng. Rep., at 1047– 1048. Although the court ultimately held that Dixon had not needed to produce the vouchers on account of attorney- client privilege, none of the justices expressed the slightest doubt about the general propriety of subpoenas duces tecum in the criminal context. See id., at 1688, 97 Eng. Rep., at 1048. As Lord Chief Justice Ellenborough later explained, “[i]n that case no objection was taken to the writ, but to the special circumstances under which the party possessed the papers; so that the Court may be considered as recognizing the general obligation to obey writs of that description in other cases.” Amey, supra, at 485, 103 Eng. Rep., at 658; see also 4 J. Chitty, Practical Treatise on the Criminal Law 185 (1816) (template for criminal subpoena duces tecum). As Dixon shows, subpoenas duces tecum were routine in part because of their close association with grand juries. Early American colonists imported the grand jury, like so many other common-law traditions, and they quickly flourished. See United States v. Calandra, 414 U. S. 338, 342–343 (1974). Grand juries were empaneled by the federal courts almost as soon as the latter were estab- lished, and both they and their state counterparts actively exercised their wide-ranging common-law authority. See R. Younger, The People’s Panel 47–55 (1963). Indeed, “the Founders thought the grand jury so essential . . . that they provided in the Fifth Amendment that federal prosecution Cite as: 585 U. S. ____ (2018) 7

for serious crimes can only be instituted by ‘a presentment or indictment of a Grand Jury.’ ” Calandra, supra, at 343. Given the popularity and prevalence of grand juries at the time, the Founders must have been intimately familiar with the tools they used—including compulsory process— to accomplish their work. As a matter of tradition, grand juries were “accorded wide latitude to inquire into viola- tions of criminal law,” including the power to “compel the production of evidence or the testimony of witnesses as [they] conside[r] appropriate.” Ibid. Long before national independence was achieved, grand juries were already using their broad inquisitorial powers not only to present and indict criminal suspects but also to inspect public buildings, to levy taxes, to supervise the administration of the laws, to advance municipal reforms such as street repair and bridge maintenance, and in some cases even to propose legislation. Younger, supra, at 5–26. Of course, such work depended entirely on grand juries’ ability to access any relevant documents. Grand juries continued to exercise these broad inquisi- torial powers up through the time of the founding. See Blair v. United States, 250 U. S. 273, 280 (1919) (“At the foundation of our Federal Government the inquisitorial function of the grand jury and the compulsion of witnesses were recognized as incidents of the judicial power”). In a series of lectures delivered in the early 1790’s, Justice James Wilson crowed that grand juries were “the peculiar boast of the common law” thanks in part to their wide- ranging authority: “All the operations of government, and of its ministers and officers, are within the compass of their view and research.” 2 J. Wilson, The Works of James Wilson 534, 537 (R. McCloskey ed. 1967). That reflected the broader insight that “[t]he grand jury’s investigative power must be broad if its public responsibility is ade- quately to be discharged.” Calandra, supra, at 344. Compulsory process was also familiar to the founding 8 CARPENTER v. UNITED STATES

generation in part because it reflected “the ancient propo- sition of law” that “ ‘ “the public . . . has a right to every man’s evidence.” ’ ” United States v. Nixon, 418 U. S. 683, 709 (1974); see also ante, at 10 (KENNEDY, J., dissenting). As early as 1612, “Lord Bacon is reported to have declared that ‘all subjects, without distinction of degrees, owe to the King tribute and service, not only of their deed and hand, but of their knowledge and discovery.’ ” Blair, supra, at 279–280. That duty could be “onerous at times,” yet the Founders considered it “necessary to the administration of justice according to the forms and modes established in our system of government.” Id., at 281; see also Calandra, supra, at 345. B Talk of kings and common-law writs may seem out of place in a case about cell-site records and the protections afforded by the Fourth Amendment in the modern age. But this history matters, not least because it tells us what was on the minds of those who ratified the Fourth Amendment and how they understood its scope. That history makes it abundantly clear that the Fourth Amendment, as originally understood, did not apply to the compulsory production of documents at all. The Fourth Amendment does not regulate all methods by which the Government obtains documents. Rather, it prohibits only those “searches and seizures” of “persons, houses, papers, and effects” that are “unreasonable.” Consistent with that language, “at least until the latter half of the 20th century” “our Fourth Amendment juris- prudence was tied to common-law trespass.” United States v. Jones, 565 U. S. 400, 405 (2012). So by its terms, the Fourth Amendment does not apply to the compulsory production of documents, a practice that involves neither any physical intrusion into private space nor any taking of property by agents of the state. Even Justice Brandeis—a Cite as: 585 U. S. ____ (2018) 9

stalwart proponent of construing the Fourth Amendment liberally—acknowledged that “under any ordinary con- struction of language,” “there is no ‘search’ or ‘seizure’ when a defendant is required to produce a document in the orderly process of a court’s procedure.” Olmstead v. United States, 277 U. S. 438, 476 (1928) (dissenting opinion).1 Nor is there any reason to believe that the Founders intended the Fourth Amendment to regulate courts’ use of compulsory process. American colonists rebelled against the Crown’s physical invasions of their persons and their property, not against its acquisition of information by any and all means. As Justice Black once put it, “[t]he Fourth Amendment was aimed directly at the abhorred practice of breaking in, ransacking and searching homes and other buildings and seizing people’s personal belongings without warrants issued by magistrates.” Katz, 389 U. S., at 367 (dissenting opinion). More recently, we have acknowl- edged that “the Fourth Amendment was the founding generation’s response to the reviled ‘general warrants’ and ‘writs of assistance’ of the colonial era, which allowed —————— 1 Any other interpretation of the Fourth Amendment’s text would run

into insuperable problems because it would apply not only to subpoenas duces tecum but to all other forms of compulsory process as well. If the Fourth Amendment applies to the compelled production of documents, then it must also apply to the compelled production of testimony—an outcome that we have repeatedly rejected and which, if accepted, would send much of the field of criminal procedure into a tailspin. See, e.g., United States v. Dionisio, 410 U. S. 1, 9 (1973) (“It is clear that a subpoena to appear before a grand jury is not a ‘seizure’ in the Fourth Amendment sense, even though that summons may be inconvenient or burdensome”); United States v. Calandra, 414 U. S. 338, 354 (1974) (“Grand jury questions . . . involve no independent governmental invasion of one’s person, house, papers, or effects”). As a matter of original understanding, a subpoena duces tecum no more effects a “search” or “seizure” of papers within the meaning of the Fourth Amendment than a subpoena ad testificandum effects a “search” or “seizure” of a person. 10 CARPENTER v. UNITED STATES

British officers to rummage through homes in an unre- strained search for evidence of criminal activity.” Riley v. California, 573 U. S. ___, ___ (2014) (slip op., at 27). General warrants and writs of assistance were noxious not because they allowed the Government to acquire evidence in criminal investigations, but because of the means by which they permitted the Government to acquire that evidence. Then, as today, searches could be quite invasive. Searches generally begin with officers “mak[ing] nonconsensual entries into areas not open to the public.” Donovan v. Lone Steer, Inc., 464 U. S. 408, 414 (1984). Once there, officers are necessarily in a position to observe private spaces generally shielded from the public and discernible only with the owner’s consent. Private area after private area becomes exposed to the officers’ eyes as they rummage through the owner’s property in their hunt for the object or objects of the search. If they are search- ing for documents, officers may additionally have to rifle through many other papers—potentially filled with the most intimate details of a person’s thoughts and life— before they find the specific information they are seeking. See Andresen v. Maryland, 427 U. S. 463, 482, n. 11 (1976). If anything sufficiently incriminating comes into view, officers seize it. Horton v. California, 496 U. S. 128, 136–137 (1990). Physical destruction always lurks as an underlying possibility; “officers executing search warrants on occasion must damage property in order to perform their duty.” Dalia v. United States, 441 U. S. 238, 258 (1979); see, e.g., United States v. Ramirez, 523 U. S. 65, 71–72 (1998) (breaking garage window); United States v. Ross, 456 U. S. 798, 817–818 (1982) (ripping open car upholstery); Brown v. Battle Creek Police Dept., 844 F. 3d 556, 572 (CA6 2016) (shooting and killing two pet dogs); Lawmaster v. Ward, 125 F. 3d 1341, 1350, n. 3 (CA10 1997) (breaking locks). Compliance with a subpoena duces tecum requires none Cite as: 585 U. S. ____ (2018) 11

of that. A subpoena duces tecum permits a subpoenaed individual to conduct the search for the relevant docu- ments himself, without law enforcement officers entering his home or rooting through his papers and effects. As a result, subpoenas avoid the many incidental invasions of privacy that necessarily accompany any actual search. And it was those invasions of privacy—which, although incidental, could often be extremely intrusive and damag- ing—that led to the adoption of the Fourth Amendment. Neither this Court nor any of the parties have offered the slightest bit of historical evidence to support the idea that the Fourth Amendment originally applied to subpoe- nas duces tecum and other forms of compulsory process. That is telling, for as I have explained, these forms of compulsory process were a feature of criminal (and civil) procedure well known to the Founders. The Founders would thus have understood that holding the compulsory production of documents to the same standard as actual searches and seizures would cripple the work of courts in civil and criminal cases alike. It would be remarkable to think that, despite that knowledge, the Founders would have gone ahead and sought to impose such a require- ment. It would be even more incredible to believe that the Founders would have imposed that requirement through the inapt vehicle of an amendment directed at different concerns. But it would blink reality entirely to argue that this entire process happened without anyone saying the least thing about it—not during the drafting of the Bill of Rights, not during any of the subsequent ratification debates, and not for most of the century that followed. If the Founders thought the Fourth Amendment applied to the compulsory production of documents, one would imag- ine that there would be some founding-era evidence of the Fourth Amendment being applied to the compulsory pro- duction of documents. Cf. Free Enterprise Fund v. Public Company Accounting Oversight Bd., 561 U. S. 477, 505 12 CARPENTER v. UNITED STATES

(2010); Printz v. United States, 521 U. S. 898, 905 (1997). Yet none has been brought to our attention. C Of course, our jurisprudence has not stood still since 1791. We now evaluate subpoenas duces tecum and other forms of compulsory document production under the Fourth Amendment, although we employ a reasonableness standard that is less demanding than the requirements for a warrant. But the road to that doctrinal destination was anything but smooth, and our initial missteps—and the subsequent struggle to extricate ourselves from their consequences—should provide an object lesson for today’s majority about the dangers of holding compulsory process to the same standard as actual searches and seizures. For almost a century after the Fourth Amendment was enacted, this Court said and did nothing to indicate that it might regulate the compulsory production of documents. But that changed temporarily when the Court decided Boyd v. United States, 116 U. S. 616 (1886), the first—and, until today, the only—case in which this Court has ever held the compulsory production of documents to the same standard as actual searches and seizures. The Boyd Court held that a court order compelling a company to produce potentially incriminating business records violated both the Fourth and the Fifth Amend- ments. The Court acknowledged that “certain aggravating incidents of actual search and seizure, such as forcible entry into a man’s house and searching amongst his pa- pers, are wanting” when the Government relies on com- pulsory process. Id., at 622. But it nevertheless asserted that the Fourth Amendment ought to “be liberally con- strued,” id., at 635, and further reasoned that compulsory process “effects the sole object and purpose of search and seizure” by “forcing from a party evidence against him- self,” id., at 622. “In this regard,” the Court concluded, Cite as: 585 U. S. ____ (2018) 13

“the Fourth and Fifth Amendments run almost into each other.” Id., at 630. Having equated compulsory process with actual searches and seizures and having melded the Fourth Amendment with the Fifth, the Court then found the order at issue unconstitutional because it compelled the production of property to which the Government did not have superior title. See id., at 622–630. In a concurrence joined by Chief Justice Waite, Justice Miller agreed that the order violated the Fifth Amend- ment, id., at 639, but he strongly protested the majority’s invocation of the Fourth Amendment. He explained: “[T]here is no reason why this court should assume that the action of the court below, in requiring a party to pro- duce certain papers . . . , authorizes an unreasonable search or seizure of the house, papers, or effects of that party. There is in fact no search and no seizure.” Ibid. “If the mere service of a notice to produce a paper . . . is a search,” Justice Miller concluded, “then a change has taken place in the meaning of words, which has not come within my reading, and which I think was unknown at the time the Constitution was made.” Id., at 641. Although Boyd was replete with stirring rhetoric, its reasoning was confused from start to finish in a way that ultimately made the decision unworkable. See 3 W. LaFave, J. Israel, N. King, & O. Kerr, Criminal Procedure §8.7(a) (4th ed. 2015). Over the next 50 years, the Court would gradually roll back Boyd’s erroneous conflation of compulsory process with actual searches and seizures. That effort took its first significant stride in Hale v. Henkel, 201 U. S. 43 (1906), where the Court found it “quite clear” and “conclusive” that “the search and seizure clause of the Fourth Amendment was not intended to interfere with the power of courts to compel, through a subpœna duces tecum, the production, upon a trial in court, of documentary evidence.” Id., at 73. Without that writ, the Court recognized, “it would be ‘utterly impossible 14 CARPENTER v. UNITED STATES

to carry on the administration of justice.’ ” Ibid. Hale, however, did not entirely liberate subpoenas duces tecum from Fourth Amendment constraints. While refusing to treat such subpoenas as the equivalent of actual searches, Hale concluded that they must not be unreasonable. And it held that the subpoena duces tecum at issue was “far too sweeping in its terms to be regarded as reasonable.” Id., at 76. The Hale Court thus left two critical questions unanswered: Under the Fourth Amend- ment, what makes the compulsory production of docu- ments “reasonable,” and how does that standard differ from the one that governs actual searches and seizures? The Court answered both of those questions definitively in Oklahoma Press Publishing Co. v. Walling, 327 U. S. 186 (1946), where we held that the Fourth Amendment regulates the compelled production of documents, but less stringently than it does full-blown searches and seizures. Oklahoma Press began by admitting that the Court’s opinions on the subject had “perhaps too often . . . been generative of heat rather than light,” “mov[ing] with vari- ant direction” and sometimes having “highly contrasting” “emphasis and tone.” Id., at 202. “The primary source of misconception concerning the Fourth Amendment’s func- tion” in this context, the Court explained, “lies perhaps in the identification of cases involving so-called ‘figurative’ or ‘constructive’ search with cases of actual search and sei- zure.” Ibid. But the Court held that “the basic distinc- tion” between the compulsory production of documents on the one hand, and actual searches and seizures on the other, meant that two different standards had to be ap- plied. Id., at 204. Having reversed Boyd’s conflation of the compelled production of documents with actual searches and sei- zures, the Court then set forth the relevant Fourth Amendment standard for the former. When it comes to “the production of corporate or other business records,” the Cite as: 585 U. S. ____ (2018) 15

Court held that the Fourth Amendment “at the most guards against abuse only by way of too much indefinite- ness or breadth in the things required to be ‘particularly described,’ if also the inquiry is one the demanding agency is authorized by law to make and the materials specified are relevant.” Oklahoma Press, supra, at 208. Notably, the Court held that a showing of probable cause was not necessary so long as “the investigation is authorized by Congress, is for a purpose Congress can order, and the documents sought are relevant to the inquiry.” Id., at 209. Since Oklahoma Press, we have consistently hewed to that standard. See, e.g., Lone Steer, Inc., 464 U. S., at 414–415; United States v. Miller, 425 U. S. 435, 445–446 (1976); California Bankers Assn. v. Shultz, 416 U. S. 21, 67 (1974); United States v. Dionisio, 410 U. S. 1, 11–12 (1973); See v. Seattle, 387 U. S. 541, 544 (1967); United States v. Powell, 379 U. S. 48, 57–58 (1964); McPhaul v. United States, 364 U. S. 372, 382–383 (1960); United States v. Morton Salt Co., 338 U. S. 632, 652–653 (1950); cf. McLane Co. v. EEOC, 581 U. S. ___, ___ (2017) (slip op., at 11). By applying Oklahoma Press and thereby respect- ing “the traditional distinction between a search warrant and a subpoena,” Miller, supra, at 446, this Court has reinforced “the basic compromise” between “the public interest” in every man’s evidence and the private interest “of men to be free from officious meddling.” Oklahoma Press, supra, at 213. D Today, however, the majority inexplicably ignores the settled rule of Oklahoma Press in favor of a resurrected version of Boyd. That is mystifying. This should have been an easy case regardless of whether the Court looked to the original understanding of the Fourth Amendment or to our modern doctrine. As a matter of original understanding, the Fourth 16 CARPENTER v. UNITED STATES

Amendment does not regulate the compelled production of documents at all. Here the Government received the relevant cell-site records pursuant to a court order compel- ling Carpenter’s cell service provider to turn them over. That process is thus immune from challenge under the original understanding of the Fourth Amendment. As a matter of modern doctrine, this case is equally straightforward. As JUSTICE KENNEDY explains, no search or seizure of Carpenter or his property occurred in this case. Ante, at 6–22; see also Part II, infra. But even if the majority were right that the Government “searched” Carpenter, it would at most be a “figurative or construc- tive search” governed by the Oklahoma Press standard, not an “actual search” controlled by the Fourth Amend- ment’s warrant requirement. And there is no doubt that the Government met the Oklahoma Press standard here. Under Oklahoma Press, a court order must “ ‘be sufficiently limited in scope, relevant in purpose, and specific in directive so that compliance will not be unreasonably burdensome.’ ” Lone Steer, Inc., supra, at 415. Here, the type of order obtained by the Government almost necessarily satisfies that standard. The Stored Communications Act allows a court to issue the relevant type of order “only if the governmental entity offers specific and articulable facts showing that there are reasonable grounds to believe that . . . the records . . . sough[t] are relevant and material to an ongoing criminal investigation.” 18 U. S. C. §2703(d). And the court “may quash or modify such order” if the provider objects that the “records requested are unusually voluminous in na- ture or compliance with such order otherwise would cause an undue burden on such provider.” Ibid. No such objec- tion was made in this case, and Carpenter does not sug- gest that the orders contravened the Oklahoma Press standard in any other way. That is what makes the majority’s opinion so puzzling. Cite as: 585 U. S. ____ (2018) 17

It decides that a “search” of Carpenter occurred within the meaning of the Fourth Amendment, but then it leaps straight to imposing requirements that—until this point— have governed only actual searches and seizures. See ante, at 18–19. Lost in its race to the finish is any real recognition of the century’s worth of precedent it jeopard- izes. For the majority, this case is apparently no different from one in which Government agents raided Carpenter’s home and removed records associated with his cell phone. Against centuries of precedent and practice, all that the Court can muster is the observation that “this Court has never held that the Government may subpoena third parties for records in which the suspect has a reasonable expectation of privacy.” Ante, at 19. Frankly, I cannot imagine a concession more damning to the Court’s argu- ment than that. As the Court well knows, the reason that we have never seen such a case is because—until today— defendants categorically had no “reasonable expectation of privacy” and no property interest in records belonging to third parties. See Part II, infra. By implying otherwise, the Court tries the nice trick of seeking shelter under the cover of precedents that it simultaneously perforates. Not only that, but even if the Fourth Amendment per- mitted someone to object to the subpoena of a third party’s records, the Court cannot explain why that individual should be entitled to greater Fourth Amendment protec- tion than the party actually being subpoenaed. When parties are subpoenaed to turn over their records, after all, they will at most receive the protection afforded by Okla- homa Press even though they will own and have a reason- able expectation of privacy in the records at issue. Under the Court’s decision, however, the Fourth Amendment will extend greater protections to someone else who is not being subpoenaed and does not own the records. That outcome makes no sense, and the Court does not even attempt to defend it. 18 CARPENTER v. UNITED STATES

We have set forth the relevant Fourth Amendment standard for subpoenaing business records many times over. Out of those dozens of cases, the majority cannot find even one that so much as suggests an exception to the Oklahoma Press standard for sufficiently personal infor- mation. Instead, we have always “described the constitu- tional requirements” for compulsory process as being “ ‘settled’ ” and as applying categorically to all “ ‘subpoenas [of] corporate books or records.’ ” Lone Steer, Inc., 464 U. S., at 415 (internal quotation marks omitted). That standard, we have held, is “the most” protection the Fourth Amendment gives “to the production of corporate records and papers.” Oklahoma Press, 327 U. S., at 208 (emphasis added).2 Although the majority announces its holding in the context of the Stored Communications Act, nothing stops its logic from sweeping much further. The Court has offered no meaningful limiting principle, and none is apparent. Cf. Tr. of Oral Arg. 31 (Carpenter’s counsel admitting that “a grand jury subpoena . . . would be held to the same standard as any other subpoena or subpoena- like request for [cell-site] records”). Holding that subpoenas must meet the same standard as conventional searches will seriously damage, if not destroy, their utility. Even more so than at the founding, today the Government regularly uses subpoenas duces tecum and other forms of compulsory process to carry out its essential functions. See, e.g., Dionisio, 410 U. S., at 11–12 (grand jury subpoenas); McPhaul, 364 U. S., at 382–383 (legislative subpoenas); Oklahoma Press, supra, at 208–209 (administrative subpoenas). Grand juries, for —————— 2 All that the Court can say in response is that we have “been careful

not to uncritically extend existing precedents” when confronting new technologies. Ante, at 20. But applying a categorical rule categorically does not “extend” precedent, so the Court’s statement ends up sounding a lot like a tacit admission that it is overruling our precedents. Cite as: 585 U. S. ____ (2018) 19

example, have long “compel[led] the production of evi- dence” in order to determine “whether there is probable cause to believe a crime has been committed.” Calandra, 414 U. S., at 343 (emphasis added). Almost by definition, then, grand juries will be unable at first to demonstrate “the probable cause required for a warrant.” Ante, at 19 (majority opinion); see also Oklahoma Press, supra, at 213. If they are required to do so, the effects are as predictable as they are alarming: Many investigations will sputter out at the start, and a host of criminals will be able to evade law enforcement’s reach. “To ensure that justice is done, it is imperative to the function of courts that compulsory process be available for the production of evidence.” Nixon, 418 U. S., at 709. For over a hundred years, we have understood that holding subpoenas to the same standard as actual searches and seizures “would stop much if not all of investigation in the public interest at the threshold of inquiry.” Oklahoma Press, supra, at 213. Today a skeptical majority decides to put that understanding to the test. II Compounding its initial error, the Court also holds that a defendant has the right under the Fourth Amendment to object to the search of a third party’s property. This hold- ing flouts the clear text of the Fourth Amendment, and it cannot be defended under either a property-based inter- pretation of that Amendment or our decisions applying the reasonable-expectations-of-privacy test adopted in Katz, 389 U. S. 347. By allowing Carpenter to object to the search of a third party’s property, the Court threatens to revolutionize a second and independent line of Fourth Amendment doctrine. A It bears repeating that the Fourth Amendment guaran- 20 CARPENTER v. UNITED STATES

tees “[t]he right of the people to be secure in their persons, houses, papers, and effects.” (Emphasis added.) The Fourth Amendment does not confer rights with respect to the persons, houses, papers, and effects of others. Its language makes clear that “Fourth Amendment rights are personal,” Rakas v. Illinois, 439 U. S. 128, 140 (1978), and as a result, this Court has long insisted that they “may not be asserted vicariously,” id., at 133. It follows that a “person who is aggrieved . . . only through the introduction of damaging evidence secured by a search of a third per- son’s premises or property has not had any of his Fourth Amendment rights infringed.” Id., at 134. In this case, as JUSTICE KENNEDY cogently explains, the cell-site records obtained by the Government belong to Carpenter’s cell service providers, not to Carpenter. See ante, at 12–13. Carpenter did not create the cell-site records. Nor did he have possession of them; at all rele- vant times, they were kept by the providers. Once Car- penter subscribed to his provider’s service, he had no right to prevent the company from creating or keeping the information in its records. Carpenter also had no right to demand that the providers destroy the records, no right to prevent the providers from destroying the records, and, indeed, no right to modify the records in any way whatso- ever (or to prevent the providers from modifying the rec- ords). Carpenter, in short, has no meaningful control over the cell-site records, which are created, maintained, al- tered, used, and eventually destroyed by his cell service providers. Carpenter responds by pointing to a provision of the Telecommunications Act that requires a provider to dis- close cell-site records when a customer so requests. See 47 U. S. C. §222(c)(2). But a statutory disclosure requirement is hardly sufficient to give someone an ownership interest in the documents that must be copied and disclosed. Many statutes confer a right to obtain copies of documents Cite as: 585 U. S. ____ (2018) 21

without creating any property right.3 Carpenter’s argument is particularly hard to swallow because nothing in the Telecommunications Act precludes cell service providers from charging customers a fee for accessing cell-site records. See ante, at 12–13 (KENNEDY, J., dissenting). It would be very strange if the owner of records were required to pay in order to inspect his own —————— 3 See, e.g., Freedom of Information Act, 5 U. S. C. §552(a) (“Each

agency shall make available to the public information as follows . . .”); Privacy Act, 5 U. S. C. §552a(d)(1) (“Each agency that maintains a system of records shall . . . upon request by any individual to gain access to his record or to any information pertaining to him which is contained in the system, permit him and upon his request, a person of his own choosing to accompany him, to review the record and have a copy made of all or any portion thereof . . .”); Fair Credit Reporting Act, 15 U. S. C. §1681j(a)(1)(A) (“All consumer reporting agencies . . . shall make all disclosures pursuant to section 1681g of this title once during any 12-month period upon request of the consumer and without charge to the consumer”); Right to Financial Privacy Act of 1978, 12 U. S. C. §3404(c) (“The customer has the right . . . to obtain a copy of the record which the financial institution shall keep of all instances in which the customer’s record is disclosed to a Government authority pursuant to this section, including the identity of the Government authority to which such disclosure is made”); Government in the Sunshine Act, 5 U. S. C. §552b(f )(2) (“Copies of such transcript, or minutes, or a tran- scription of such recording disclosing the identity of each speaker, shall be furnished to any person at the actual cost of duplication or transcrip- tion”); Cable Act, 47 U. S. C. §551(d) (“A cable subscriber shall be provided access to all personally identifiable information regarding that subscriber which is collected and maintained by a cable operator”); Family Educational Rights and Privacy Act of 1974, 20 U. S. C. §1232g(a)(1)(A) (“No funds shall be made available under any applica- ble program to any educational agency or institution which has a policy of denying, or which effectively prevents, the parents of students who are or have been in attendance at a school of such agency or at such institution, as the case may be, the right to inspect and review the education records of their children. . . . Each educational agency or institution shall establish appropriate procedures for the granting of a request by parents for access to the education records of their children within a reasonable period of time, but in no case more than forty-five days after the request has been made”). 22 CARPENTER v. UNITED STATES

property. Nor does the Telecommunications Act give Carpenter a property right in the cell-site records simply because they are subject to confidentiality restrictions. See 47 U. S. C. §222(c)(1) (without a customer’s permission, a cell service provider may generally “use, disclose, or permit access to individually identifiable [cell-site records]” only with respect to “its provision” of telecommunications services). Many federal statutes impose similar restrictions on private entities’ use or dissemination of information in their own records without conferring a property right on third parties.4 —————— 4 See, e.g., Family Educational Rights and Privacy Act, 20 U. S. C. §1232g(b)(1) (“No funds shall be made available under any applicable program to any educational agency or institution which has a policy or practice of permitting the release of education records (or personally identifiable information contained therein other than directory infor- mation . . . ) of students without the written consent of their parents to any individual, agency, or organization . . .”); Video Privacy Protection Act, 18 U. S. C. §2710(b)(1) (“A video tape service provider who know- ingly discloses, to any person, personally identifiable information concerning any consumer of such provider shall be liable to the ag- grieved person for the relief provided in subsection (d)”); Driver Privacy Protection Act, 18 U. S. C. §2721(a)(1) (“A State department of motor vehicles, and any officer, employee, or contractor thereof, shall not knowingly disclose or otherwise make available to any person or entity . . . personal information . . .”); Fair Credit Reporting Act, 15 U. S. C. §1681b(a) (“[A]ny consumer reporting agency may furnish a consumer report under the following circumstances and no other . . .”); Right to Financial Privacy Act, 12 U. S. C. §3403(a) (“No financial institution, or officer, employees, or agent of a financial institution, may provide to any Government authority access to or copies of, or the information contained in, the financial records of any customer except in accordance with the provisions of this chapter”); Patient Safety and Quality Im- provement Act, 42 U. S. C. §299b–22(b) (“Notwithstanding any other provision of Federal, State, or local law, and subject to subsection (c) of this section, patient safety work product shall be confidential and shall not be disclosed”); Cable Act, 47 U. S. C. §551(c)(1) (“[A] cable operator shall not disclose personally identifiable information concerning any subscriber without the prior written or electronic consent of the sub- Cite as: 585 U. S. ____ (2018) 23

It would be especially strange to hold that the Telecom- munication Act’s confidentiality provision confers a prop- erty right when the Act creates an express exception for any disclosure of records that is “required by law.” 47 U. S. C. §222(c)(1). So not only does Carpenter lack “ ‘the most essential and beneficial’ ” of the “ ‘constituent ele- ments’ ” of property, Dickman v. Commissioner, 465 U. S. 330, 336 (1984)—i.e., the right to use the property to the exclusion of others—but he cannot even exclude the party he would most like to keep out, namely, the Government.5 For all these reasons, there is no plausible ground for maintaining that the information at issue here represents Carpenter’s “papers” or “effects.”6

—————— scriber concerned and shall take such actions as are necessary to prevent unauthorized access to such information by a person other than the subscriber or cable operator”). 5 Carpenter also cannot argue that he owns the cell-site records merely

because they fall into the category of records referred to as “customer proprietary network information.” 47 U. S. C. §222(c). Even assuming labels alone can confer property rights, nothing in this particular label indicates whether the “information” is “proprietary” to the “customer” or to the provider of the “network.” At best, the phrase “customer proprietary network information” is ambiguous, and context makes clear that it refers to the provider’s information. The Telecommunica- tions Act defines the term to include all “information that relates to the quantity, technical configuration, type, destination, location, and amount of use of a telecommunications service subscribed to by any customer of a telecommunications carrier, and that is made available to the carrier by the customer solely by virtue of the carrier-customer relationship.” 47 U. S. C. §222(h)(1)(A). For Carpenter to be right, he must own not only the cell-site records in this case, but also records relating to, for example, the “technical configuration” of his subscribed service—records that presumably include such intensely personal and private information as transmission wavelengths, transport protocols, and link layer system configurations. 6 Thus, this is not a case in which someone has entrusted papers that

he or she owns to the safekeeping of another, and it does not involve a bailment. Cf. post, at 14 (GORSUCH, J., dissenting). 24 CARPENTER v. UNITED STATES

B In the days when this Court followed an exclusively property-based approach to the Fourth Amendment, the distinction between an individual’s Fourth Amendment rights and those of a third party was clear cut. We first asked whether the object of the search—say, a house, papers, or effects—belonged to the defendant, and, if it did, whether the Government had committed a “trespass” in acquiring the evidence at issue. Jones, 565 U. S., at 411, n. 8. When the Court held in Katz that “property rights are not the sole measure of Fourth Amendment violations,” Soldal v. Cook County, 506 U. S. 56, 64 (1992), the sharp boundary between personal and third-party rights was tested. Under Katz, a party may invoke the Fourth Amendment whenever law enforcement officers violate the party’s “justifiable” or “reasonable” expectation of privacy. See 389 U. S., at 353; see also id., at 361 (Harlan, J., concurring) (applying the Fourth Amendment where “a person [has] exhibited an actual (subjective) expectation of privacy” and where that “expectation [is] one that society is prepared to recognize as ‘reasonable’ ”). Thus freed from the limitations imposed by property law, parties began to argue that they had a reasonable expectation of privacy in items owned by others. After all, if a trusted third party took care not to disclose information about the person in question, that person might well have a reasonable expec- tation that the information would not be revealed. Efforts to claim Fourth Amendment protection against searches of the papers and effects of others came to a head in Miller, 425 U. S. 435, where the defendant sought the suppression of two banks’ microfilm copies of his checks, deposit slips, and other records. The defendant did not claim that he owned these documents, but he nonetheless argued that “analysis of ownership, property rights and possessory interests in the determination of Fourth Cite as: 585 U. S. ____ (2018) 25

Amendment rights ha[d] been severely impeached” by Katz and other recent cases. See Brief for Respondent in United States v. Miller, O. T. 1975, No. 74–1179, p. 6. Turning to Katz, he then argued that he had a reasonable expectation of privacy in the banks’ records regarding his accounts. Brief for Respondent in No. 74–1179, at 6; see also Miller, supra, at 442–443. Acceptance of this argument would have flown in the face of the Fourth Amendment’s text, and the Court re- jected that development. Because Miller gave up “domin- ion and control” of the relevant information to his bank, Rakas, 439 U. S., at 149, the Court ruled that he lost any protected Fourth Amendment interest in that information. See Miller, supra, at 442–443. Later, in Smith v. Mary- land, 442 U. S. 735, 745 (1979), the Court reached a simi- lar conclusion regarding a telephone company’s records of a customer’s calls. As JUSTICE KENNEDY concludes, Miller and Smith are thus best understood as placing “necessary limits on the ability of individuals to assert Fourth Amendment interests in property to which they lack a ‘requisite connection.’ ” Ante, at 8. The same is true here, where Carpenter indisputably lacks any meaningful property-based connection to the cell-site records owned by his provider. Because the rec- ords are not Carpenter’s in any sense, Carpenter may not seek to use the Fourth Amendment to exclude them. By holding otherwise, the Court effectively allows Car- penter to object to the “search” of a third party’s property, not recognizing the revolutionary nature of this change. The Court seems to think that Miller and Smith invented a new “doctrine”—“the third-party doctrine”—and the Court refuses to “extend” this product of the 1970’s to a new age of digital communications. Ante, at 11, 17. But the Court fundamentally misunderstands the role of Mil- ler and Smith. Those decisions did not forge a new doc- trine; instead, they rejected an argument that would have 26 CARPENTER v. UNITED STATES

disregarded the clear text of the Fourth Amendment and a formidable body of precedent. In the end, the Court never explains how its decision can be squared with the fact that the Fourth Amendment protects only “[t]he right of the people to be secure in their persons, houses, papers, and effects.” (Emphasis added.) * * * Although the majority professes a desire not to “ ‘embar- rass the future,’ ” ante, at 18, we can guess where today’s decision will lead. One possibility is that the broad principles that the Court seems to embrace will be applied across the board. All subpoenas duces tecum and all other orders compelling the production of documents will require a demonstration of probable cause, and individuals will be able to claim a protected Fourth Amendment interest in any sensitive personal information about them that is collected and owned by third parties. Those would be revolutionary developments indeed. The other possibility is that this Court will face the embarrassment of explaining in case after case that the principles on which today’s decision rests are subject to all sorts of qualifications and limitations that have not yet been discovered. If we take this latter course, we will inevitably end up “mak[ing] a crazy quilt of the Fourth Amendment.” Smith, supra, at 745. All of this is unnecessary. In the Stored Communica- tions Act, Congress addressed the specific problem at issue in this case. The Act restricts the misuse of cell-site rec- ords by cell service providers, something that the Fourth Amendment cannot do. The Act also goes beyond current Fourth Amendment case law in restricting access by law enforcement. It permits law enforcement officers to ac- quire cell-site records only if they meet a heightened standard and obtain a court order. If the American people Cite as: 585 U. S. ____ (2018) 27

now think that the Act is inadequate or needs updating, they can turn to their elected representatives to adopt more protective provisions. Because the collection and storage of cell-site records affects nearly every American, it is unlikely that the question whether the current law requires strengthening will escape Congress’s notice. Legislation is much preferable to the development of an entirely new body of Fourth Amendment caselaw for many reasons, including the enormous complexity of the subject, the need to respond to rapidly changing technology, and the Fourth Amendment’s limited scope. The Fourth Amendment restricts the conduct of the Federal Govern- ment and the States; it does not apply to private actors. But today, some of the greatest threats to individual pri- vacy may come from powerful private companies that collect and sometimes misuse vast quantities of data about the lives of ordinary Americans. If today’s decision en- courages the public to think that this Court can protect them from this looming threat to their privacy, the deci- sion will mislead as well as disrupt. And if holding a provision of the Stored Communications Act to be uncon- stitutional dissuades Congress from further legislation in this field, the goal of protecting privacy will be greatly disserved. The desire to make a statement about privacy in the digital age does not justify the consequences that today’s decision is likely to produce. Cite as: 585 U. S. ____ (2018) 1

GORSUCH, J., dissenting

JUSTICE GORSUCH, dissenting. In the late 1960s this Court suggested for the first time that a search triggering the Fourth Amendment occurs when the government violates an “expectation of privacy” that “society is prepared to recognize as ‘reasonable.’ ” Katz v. United States, 389 U. S. 347, 361 (1967) (Harlan, J., concurring). Then, in a pair of decisions in the 1970s applying the Katz test, the Court held that a “reasonable expectation of privacy” doesn’t attach to information shared with “third parties.” See Smith v. Maryland, 442 U. S. 735, 743–744 (1979); United States v. Miller, 425 U. S. 435, 443 (1976). By these steps, the Court came to conclude, the Constitution does nothing to limit investiga- tors from searching records you’ve entrusted to your bank, accountant, and maybe even your doctor. What’s left of the Fourth Amendment? Today we use the Internet to do most everything. Smartphones make it easy to keep a calendar, correspond with friends, make calls, conduct banking, and even watch the game. Count- less Internet companies maintain records about us and, increasingly, for us. Even our most private documents— those that, in other eras, we would have locked safely in a desk drawer or destroyed—now reside on third party servers. Smith and Miller teach that the police can review all of this material, on the theory that no one reasonably 2 CARPENTER v. UNITED STATES

expects any of it will be kept private. But no one believes that, if they ever did. What to do? It seems to me we could respond in at least three ways. The first is to ignore the problem, maintain Smith and Miller, and live with the consequences. If the confluence of these decisions and modern technology means our Fourth Amendment rights are reduced to nearly nothing, so be it. The second choice is to set Smith and Miller aside and try again using the Katz “reasonable expectation of privacy” jurisprudence that produced them. The third is to look for answers elsewhere. * Start with the first option. Smith held that the govern- ment’s use of a pen register to record the numbers people dial on their phones doesn’t infringe a reasonable expecta- tion of privacy because that information is freely disclosed to the third party phone company. 442 U. S., at 743–744. Miller held that a bank account holder enjoys no reason- able expectation of privacy in the bank’s records of his account activity. That’s true, the Court reasoned, “even if the information is revealed on the assumption that it will be used only for a limited purpose and the confidence placed in the third party will not be betrayed.” 425 U. S., at 443. Today the Court suggests that Smith and Miller distinguish between kinds of information disclosed to third parties and require courts to decide whether to “extend” those decisions to particular classes of information, de- pending on their sensitivity. See ante, at 10–18. But as the Sixth Circuit recognized and JUSTICE KENNEDY ex- plains, no balancing test of this kind can be found in Smith and Miller. See ante, at 16 (dissenting opinion). Those cases announced a categorical rule: Once you dis- close information to third parties, you forfeit any reason- able expectation of privacy you might have had in it. And even if Smith and Miller did permit courts to conduct a Cite as: 585 U. S. ____ (2018) 3

balancing contest of the kind the Court now suggests, it’s still hard to see how that would help the petitioner in this case. Why is someone’s location when using a phone so much more sensitive than who he was talking to (Smith) or what financial transactions he engaged in (Miller)? I do not know and the Court does not say. The problem isn’t with the Sixth Circuit’s application of Smith and Miller but with the cases themselves. Can the government demand a copy of all your e-mails from Google or Microsoft without implicating your Fourth Amendment rights? Can it secure your DNA from 23andMe without a warrant or probable cause? Smith and Miller say yes it can—at least without running afoul of Katz. But that result strikes most lawyers and judges today—me in- cluded—as pretty unlikely. In the years since its adoption, countless scholars, too, have come to conclude that the “third-party doctrine is not only wrong, but horribly wrong.” Kerr, The Case for the Third-Party Doctrine, 107 Mich. L. Rev. 561, 563, n. 5, 564 (2009) (collecting criti- cisms but defending the doctrine (footnotes omitted)). The reasons are obvious. “As an empirical statement about subjective expectations of privacy,” the doctrine is “quite dubious.” Baude & Stern, The Positive Law Model of the Fourth Amendment, 129 Harv. L. Rev. 1821, 1872 (2016). People often do reasonably expect that information they entrust to third parties, especially information subject to confidentiality agreements, will be kept private. Mean- while, if the third party doctrine is supposed to represent a normative assessment of when a person should expect privacy, the notion that the answer might be “never” seems a pretty unattractive societal prescription. Ibid. What, then, is the explanation for our third party doc- trine? The truth is, the Court has never offered a persua- sive justification. The Court has said that by conveying information to a third party you “ ‘assum[e] the risk’ ” it will be revealed to the police and therefore lack a reason- 4 CARPENTER v. UNITED STATES

able expectation of privacy in it. Smith, supra, at 744. But assumption of risk doctrine developed in tort law. It generally applies when “by contract or otherwise [one] expressly agrees to accept a risk of harm” or impliedly does so by “manifest[ing] his willingness to accept” that risk and thereby “take[s] his chances as to harm which may result from it.” Restatement (Second) of Torts §§496B, 496C(1), and Comment b (1965); see also 1 D. Dobbs, P. Hayden, & E. Bublick, Law of Torts §§235–236, pp. 841–850 (2d ed. 2017). That rationale has little play in this context. Suppose I entrust a friend with a letter and he promises to keep it secret until he delivers it to an intended recipient. In what sense have I agreed to bear the risk that he will turn around, break his promise, and spill its contents to someone else? More confusing still, what have I done to “manifest my willingness to accept” the risk that the government will pry the document from my friend and read it without his consent? One possible answer concerns knowledge. I know that my friend might break his promise, or that the govern- ment might have some reason to search the papers in his possession. But knowing about a risk doesn’t mean you assume responsibility for it. Whenever you walk down the sidewalk you know a car may negligently or recklessly veer off and hit you, but that hardly means you accept the consequences and absolve the driver of any damage he may do to you. Epstein, Privacy and the Third Hand: Lessons From the Common Law of Reasonable Expecta- tions, 24 Berkeley Tech. L. J. 1199, 1204 (2009); see W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser & Keeton on Law of Torts 490 (5th ed. 1984). Some have suggested the third party doctrine is better understood to rest on consent than assumption of risk. “So long as a person knows that they are disclosing infor- mation to a third party,” the argument goes, “their choice to do so is voluntary and the consent valid.” Kerr, supra, Cite as: 585 U. S. ____ (2018) 5

at 588. I confess I still don’t see it. Consenting to give a third party access to private papers that remain my prop- erty is not the same thing as consenting to a search of those papers by the government. Perhaps there are excep- tions, like when the third party is an undercover govern- ment agent. See Murphy, The Case Against the Case Against the Third-Party Doctrine: A Response to Epstein and Kerr, 24 Berkeley Tech. L. J. 1239, 1252 (2009); cf. Hoffa v. United States, 385 U. S. 293 (1966). But other- wise this conception of consent appears to be just assump- tion of risk relabeled—you’ve “consented” to whatever risks are foreseeable. Another justification sometimes offered for third party doctrine is clarity. You (and the police) know exactly how much protection you have in information confided to oth- ers: none. As rules go, “the king always wins” is admi- rably clear. But the opposite rule would be clear too: Third party disclosures never diminish Fourth Amendment protection (call it “the king always loses”). So clarity alone cannot justify the third party doctrine. In the end, what do Smith and Miller add up to? A doubtful application of Katz that lets the government search almost whatever it wants whenever it wants. The Sixth Circuit had to follow that rule and faithfully did just that, but it’s not clear why we should. * There’s a second option. What if we dropped Smith and Miller’s third party doctrine and retreated to the root Katz question whether there is a “reasonable expectation of privacy” in data held by third parties? Rather than solve the problem with the third party doctrine, I worry this option only risks returning us to its source: After all, it was Katz that produced Smith and Miller in the first place. Katz’s problems start with the text and original under- 6 CARPENTER v. UNITED STATES

standing of the Fourth Amendment, as JUSTICE THOMAS thoughtfully explains today. Ante, at 5–17 (dissenting opinion). The Amendment’s protections do not depend on the breach of some abstract “expectation of privacy” whose contours are left to the judicial imagination. Much more concretely, it protects your “person,” and your “houses, papers, and effects.” Nor does your right to bring a Fourth Amendment claim depend on whether a judge happens to agree that your subjective expectation to privacy is a “reasonable” one. Under its plain terms, the Amendment grants you the right to invoke its guarantees whenever one of your protected things (your person, your house, your papers, or your effects) is unreasonably searched or seized. Period. History too holds problems for Katz. Little like it can be found in the law that led to the adoption of the Fourth Amendment or in this Court’s jurisprudence until the late 1960s. The Fourth Amendment came about in response to a trio of 18th century cases “well known to the men who wrote and ratified the Bill of Rights, [and] famous throughout the colonial population.” Stuntz, The Substan- tive Origins of Criminal Procedure, 105 Yale L. J. 393, 397 (1995). The first two were English cases invalidating the Crown’s use of general warrants to enter homes and search papers. Entick v. Carrington, 19 How. St. Tr. 1029 (K. B. 1765); Wilkes v. Wood, 19 How. St. Tr. 1153 (K. B. 1763); see W. Cuddihy, The Fourth Amendment: Origins and Original Meaning 439–487 (2009); Boyd v. United States, 116 U. S. 616, 625–630 (1886). The third was American: the Boston Writs of Assistance Case, which sparked colonial outrage at the use of writs permitting government agents to enter houses and business, breaking open doors and chests along the way, to conduct searches and seizures—and to force third parties to help them. Stuntz, supra, at 404–409; M. Smith, The Writs of Assis- tance Case (1978). No doubt the colonial outrage engen- Cite as: 585 U. S. ____ (2018) 7

dered by these cases rested in part on the government’s intrusion upon privacy. But the framers chose not to protect privacy in some ethereal way dependent on judicial intuitions. They chose instead to protect privacy in par- ticular places and things—“persons, houses, papers, and effects”—and against particular threats—“unreasonable” governmental “searches and seizures.” See Entick, supra, at 1066 (“Papers are the owner’s goods and chattels; they are his dearest property; and so far from enduring a sei- zure, that they will hardly bear an inspection”); see also ante, at 1–21 (THOMAS, J., dissenting). Even taken on its own terms, Katz has never been suffi- ciently justified. In fact, we still don’t even know what its “reasonable expectation of privacy” test is. Is it supposed to pose an empirical question (what privacy expectations do people actually have) or a normative one (what expecta- tions should they have)? Either way brings problems. If the test is supposed to be an empirical one, it’s unclear why judges rather than legislators should conduct it. Legislators are responsive to their constituents and have institutional resources designed to help them discern and enact majoritarian preferences. Politically insulated judges come armed with only the attorneys’ briefs, a few law clerks, and their own idiosyncratic experiences. They are hardly the representative group you’d expect (or want) to be making empirical judgments for hundreds of millions of people. Unsurprisingly, too, judicial judgments often fail to reflect public views. See Slobogin & Schumacher, Reasonable Expectations of Privacy and Autonomy in Fourth Amendment Cases: An Empirical Look at “Under- standings Recognized and Permitted by Society,” 42 Duke L. J. 727, 732, 740–742 (1993). Consider just one example. Our cases insist that the seriousness of the offense being investigated does not reduce Fourth Amendment protec- tion. Mincey v. Arizona, 437 U. S. 385, 393–394 (1978). Yet scholars suggest that most people are more tolerant of 8 CARPENTER v. UNITED STATES

police intrusions when they investigate more serious crimes. See Blumenthal, Adya, & Mogle, The Multiple Dimensions of Privacy: Testing Lay “Expectations of Pri- vacy,” 11 U. Pa. J. Const. L. 331, 352–353 (2009). And I very much doubt that this Court would be willing to adjust its Katz cases to reflect these findings even if it believed them. Maybe, then, the Katz test should be conceived as a normative question. But if that’s the case, why (again) do judges, rather than legislators, get to determine whether society should be prepared to recognize an expectation of privacy as legitimate? Deciding what privacy interests should be recognized often calls for a pure policy choice, many times between incommensurable goods—between the value of privacy in a particular setting and society’s interest in combating crime. Answering questions like that calls for the exercise of raw political will belonging to legislatures, not the legal judgment proper to courts. See The Federalist No. 78, p. 465 (C. Rossiter ed. 1961) (A. Hamilton). When judges abandon legal judgment for political will we not only risk decisions where “reasonable expectations of privacy” come to bear “an uncanny resem- blance to those expectations of privacy” shared by Mem- bers of this Court. Minnesota v. Carter, 525 U. S. 83, 97 (1998) (Scalia, J., concurring). We also risk undermining public confidence in the courts themselves. My concerns about Katz come with a caveat. Sometimes, I accept, judges may be able to discern and describe exist- ing societal norms. See, e.g., Florida v. Jardines, 569 U. S. 1, 8 (2013) (inferring a license to enter on private property from the “ ‘habits of the country’ ” (quoting McKee v. Gratz, 260 U. S. 127, 136 (1922))); Sachs, Finding Law, 107 Cal. L. Rev. (forthcoming 2019), online at https://ssrn.com/ abstract=3064443 (as last visited June 19, 2018). That is particularly true when the judge looks to positive law rather than intuition for guidance on social norms. See Cite as: 585 U. S. ____ (2018) 9

Byrd v. United States, 584 U. S. ___, ___–___ (2018) (slip op., at 7–9) (“general property-based concept[s] guid[e] the resolution of this case”). So there may be some occasions where Katz is capable of principled application—though it may simply wind up approximating the more traditional option I will discuss in a moment. Sometimes it may also be possible to apply Katz by analogizing from precedent when the line between an existing case and a new fact pattern is short and direct. But so far this Court has declined to tie itself to any significant restraints like these. See ante, at 5, n. 1 (“[W]hile property rights are often informative, our cases by no means suggest that such an interest is ‘fundamental’ or ‘dispositive’ in deter- mining which expectations of privacy are legitimate”). As a result, Katz has yielded an often unpredictable— and sometimes unbelievable—jurisprudence. Smith and Miller are only two examples; there are many others. Take Florida v. Riley, 488 U. S. 445 (1989), which says that a police helicopter hovering 400 feet above a person’s property invades no reasonable expectation of privacy. Try that one out on your neighbors. Or California v. Greenwood, 486 U. S. 35 (1988), which holds that a person has no reasonable expectation of privacy in the garbage he puts out for collection. In that case, the Court said that the homeowners forfeited their privacy interests because “[i]t is common knowledge that plastic garbage bags left on or at the side of a public street are readily accessible to animals, children, scavengers, snoops, and other members of the public.” Id., at 40 (footnotes omitted). But the habits of raccoons don’t prove much about the habits of the country. I doubt, too, that most people spotting a neighbor rummaging through their garbage would think they lacked reasonable grounds to confront the rummager. Making the decision all the stranger, California state law expressly protected a homeowner’s property rights in discarded trash. Id., at 43. Yet rather than defer to that 10 CARPENTER v. UNITED STATES

as evidence of the people’s habits and reasonable expecta- tions of privacy, the Court substituted its own curious judgment. Resorting to Katz in data privacy cases threatens more of the same. Just consider. The Court today says that judges should use Katz’s reasonable expectation of privacy test to decide what Fourth Amendment rights people have in cell-site location information, explaining that “no single rubric definitively resolves which expectations of privacy are entitled to protection.” Ante, at 5. But then it offers a twist. Lower courts should be sure to add two special principles to their Katz calculus: the need to avoid “arbi- trary power” and the importance of “plac[ing] obstacles in the way of a too permeating police surveillance.” Ante, at 6 (internal quotation marks omitted). While surely lauda- ble, these principles don’t offer lower courts much guid- ance. The Court does not tell us, for example, how far to carry either principle or how to weigh them against the legitimate needs of law enforcement. At what point does access to electronic data amount to “arbitrary” authority? When does police surveillance become “too permeating”? And what sort of “obstacles” should judges “place” in law enforcement’s path when it does? We simply do not know. The Court’s application of these principles supplies little more direction. The Court declines to say whether there is any sufficiently limited period of time “for which the Gov- ernment may obtain an individual’s historical [location information] free from Fourth Amendment scrutiny.” Ante, at 11, n. 3; see ante, at 11–15. But then it tells us that access to seven days’ worth of information does trig- ger Fourth Amendment scrutiny—even though here the carrier “produced only two days of records.” Ante, at 11, n. 3. Why is the relevant fact the seven days of information the government asked for instead of the two days of infor- mation the government actually saw? Why seven days instead of ten or three or one? And in what possible sense Cite as: 585 U. S. ____ (2018) 11

did the government “search” five days’ worth of location information it was never even sent? We do not know. Later still, the Court adds that it can’t say whether the Fourth Amendment is triggered when the government collects “real-time CSLI or ‘tower dumps’ (a download of information on all the devices that connected to a particu- lar cell site during a particular interval).” Ante, at 17–18. But what distinguishes historical data from real-time data, or seven days of a single person’s data from a down- load of everyone’s data over some indefinite period of time? Why isn’t a tower dump the paradigmatic example of “too permeating police surveillance” and a dangerous tool of “arbitrary” authority—the touchstones of the majority’s modified Katz analysis? On what possible basis could such mass data collection survive the Court’s test while collect- ing a single person’s data does not? Here again we are left to guess. At the same time, though, the Court offers some firm assurances. It tells us its decision does not “call into question conventional surveillance techniques and tools, such as security cameras.” Ibid. That, however, just raises more questions for lower courts to sort out about what techniques qualify as “conventional” and why those techniques would be okay even if they lead to “permeating police surveillance” or “arbitrary police power.” Nor is this the end of it. After finding a reasonable expectation of privacy, the Court says there’s still more work to do. Courts must determine whether to “extend” Smith and Miller to the circumstances before them. Ante, at 11, 15–17. So apparently Smith and Miller aren’t quite left for dead; they just no longer have the clear reach they once did. How do we measure their new reach? The Court says courts now must conduct a second Katz-like balancing inquiry, asking whether the fact of disclosure to a third party outweighs privacy interests in the “category of in- formation” so disclosed. Ante, at 13, 15–16. But how are lower courts supposed to weigh these radically different 12 CARPENTER v. UNITED STATES

interests? Or assign values to different categories of information? All we know is that historical cell-site loca- tion information (for seven days, anyway) escapes Smith and Miller’s shorn grasp, while a lifetime of bank or phone records does not. As to any other kind of information, lower courts will have to stay tuned. In the end, our lower court colleagues are left with two amorphous balancing tests, a series of weighty and in- commensurable principles to consider in them, and a few illustrative examples that seem little more than the prod- uct of judicial intuition. In the Court’s defense, though, we have arrived at this strange place not because the Court has misunderstood Katz. Far from it. We have arrived here because this is where Katz inevitably leads. * There is another way. From the founding until the 1960s, the right to assert a Fourth Amendment claim didn’t depend on your ability to appeal to a judge’s per- sonal sensibilities about the “reasonableness” of your expecta- tions or privacy. It was tied to the law. Jardines, 569 U. S., at 11; United States v. Jones, 565 U. S. 400, 405 (2012). The Fourth Amendment protects “the right of the people to be secure in their persons, houses, papers and effects, against unreasonable searches and seizures.” True to those words and their original understanding, the tradi- tional approach asked if a house, paper or effect was yours under law. No more was needed to trigger the Fourth Amendment. Though now often lost in Katz’s shadow, this traditional understanding persists. Katz only “supple- ments, rather than displaces the traditional property- based understanding of the Fourth Amendment.” Byrd, 584 U. S., at ___ (slip op., at 7) (internal quotation marks omitted); Jardines, supra, at 11 (same); Soldal v. Cook County, 506 U. S. 56, 64 (1992) (Katz did not “snuf[f ] out the previously recognized protection for property under Cite as: 585 U. S. ____ (2018) 13

the Fourth Amendment”). Beyond its provenance in the text and original under- standing of the Amendment, this traditional approach comes with other advantages. Judges are supposed to decide cases based on “democratically legitimate sources of law”—like positive law or analogies to items protected by the enacted Constitution—rather than “their own biases or personal policy preferences.” Pettys, Judicial Discretion in Constitutional Cases, 26 J. L. & Pol. 123, 127 (2011). A Fourth Amendment model based on positive legal rights “carves out significant room for legislative participation in the Fourth Amendment context,” too, by asking judges to consult what the people’s representatives have to say about their rights. Baude & Stern, 129 Harv. L. Rev., at 1852. Nor is this approach hobbled by Smith and Miller, for those cases are just limitations on Katz, addressing only the question whether individuals have a reasonable expectation of privacy in materials they share with third parties. Under this more traditional approach, Fourth Amendment protections for your papers and effects do not automatically disappear just because you share them with third parties. Given the prominence Katz has claimed in our doctrine, American courts are pretty rusty at applying the tradi- tional approach to the Fourth Amendment. We know that if a house, paper, or effect is yours, you have a Fourth Amendment interest in its protection. But what kind of legal interest is sufficient to make something yours? And what source of law determines that? Current positive law? The common law at 1791, extended by analogy to modern times? Both? See Byrd, supra, at ___–___ (slip op., at 1–2) (THOMAS, J., concurring); cf. Re, The Positive Law Floor, 129 Harv. L. Rev. Forum 313 (2016). Much work is needed to revitalize this area and answer these questions. I do not begin to claim all the answers today, but (unlike with Katz) at least I have a pretty good idea 14 CARPENTER v. UNITED STATES

what the questions are. And it seems to me a few things can be said. First, the fact that a third party has access to or posses- sion of your papers and effects does not necessarily elimi- nate your interest in them. Ever hand a private document to a friend to be returned? Toss your keys to a valet at a restaurant? Ask your neighbor to look after your dog while you travel? You would not expect the friend to share the document with others; the valet to lend your car to his buddy; or the neighbor to put Fido up for adoption. En- trusting your stuff to others is a bailment. A bailment is the “delivery of personal property by one person (the bailor) to another (the bailee) who holds the property for a certain purpose.” Black’s Law Dictionary 169 (10th ed. 2014); J. Story, Commentaries on the Law of Bailments §2, p. 2 (1832) (“a bailment is a delivery of a thing in trust for some special object or purpose, and upon a contract, expressed or implied, to conform to the object or purpose of the trust”). A bailee normally owes a legal duty to keep the item safe, according to the terms of the parties’ con- tract if they have one, and according to the “implication[s] from their conduct” if they don’t. 8 C. J. S., Bailments §36, pp. 468–469 (2017). A bailee who uses the item in a dif- ferent way than he’s supposed to, or against the bailor’s instructions, is liable for conversion. Id., §43, at 481; see Goad v. Harris, 207 Ala. 357, 92 So. 546, (1922); Knight v. Seney, 290 Ill. 11, 17, 124 N. E. 813, 815–816 (1919); Baxter v. Woodward, 191 Mich. 379, 385, 158 N. W. 137, 139 (1916). This approach is quite different from Smith and Miller’s (counter)-intuitive approach to reasonable expectations of privacy; where those cases extinguish Fourth Amendment interests once records are given to a third party, property law may preserve them. Our Fourth Amendment jurisprudence already reflects this truth. In Ex parte Jackson, 96 U. S. 727 (1878), this Court held that sealed letters placed in the mail are “as Cite as: 585 U. S. ____ (2018) 15

fully guarded from examination and inspection, except as to their outward form and weight, as if they were retained by the parties forwarding them in their own domiciles.” Id., at 733. The reason, drawn from the Fourth Amend- ment’s text, was that “[t]he constitutional guaranty of the right of the people to be secure in their papers against unreasonable searches and seizures extends to their pa- pers, thus closed against inspection, wherever they may be.” Ibid. (emphasis added). It did not matter that letters were bailed to a third party (the government, no less). The sender enjoyed the same Fourth Amendment protec- tion as he does “when papers are subjected to search in one’s own household.” Ibid. These ancient principles may help us address modern data cases too. Just because you entrust your data—in some cases, your modern-day papers and effects—to a third party may not mean you lose any Fourth Amend- ment interest in its contents. Whatever may be left of Smith and Miller, few doubt that e-mail should be treated much like the traditional mail it has largely supplanted— as a bailment in which the owner retains a vital and pro- tected legal interest. See ante, at 13 (KENNEDY, J., dis- senting) (noting that enhanced Fourth Amendment protec- tion may apply when the “modern-day equivalents of an individual’s own ‘papers’ or ‘effects’ . . . are held by a third party” through “bailment”); ante, at 23, n. 6 (ALITO, J., dissenting) (reserving the question whether Fourth Amendment protection may apply in the case of “bail- ment” or when “someone has entrusted papers he or she owns . . . to the safekeeping of another”); United States v. Warshak, 631 F. 3d 266, 285–286 (CA6 2010) (relying on an analogy to Jackson to extend Fourth Amendment protection to e-mail held by a third party service provider). Second, I doubt that complete ownership or exclusive control of property is always a necessary condition to the assertion of a Fourth Amendment right. Where houses 16 CARPENTER v. UNITED STATES

are concerned, for example, individuals can enjoy Fourth Amendment protection without fee simple title. Both the text of the Amendment and the common law rule support that conclusion. “People call a house ‘their’ home when legal title is in the bank, when they rent it, and even when they merely occupy it rent free.” Carter, 525 U. S., at 95– 96 (Scalia, J., concurring). That rule derives from the common law. Oystead v. Shed, 13 Mass. 520, 523 (1816) (explaining, citing “[t]he very learned judges, Foster, Hale, and Coke,” that the law “would be as much disturbed by a forcible entry to arrest a boarder or a servant, who had acquired, by contract, express or implied, a right to enter the house at all times, and to remain in it as long as they please, as if the object were to arrest the master of the house or his children”). That is why tenants and resident family members—though they have no legal title—have standing to complain about searches of the houses in which they live. Chapman v. United States, 365 U. S. 610, 616–617 (1961), Bumper v. North Carolina, 391 U. S. 543, 548, n. 11 (1968). Another point seems equally true: just because you have to entrust a third party with your data doesn’t necessarily mean you should lose all Fourth Amendment protections in it. Not infrequently one person comes into possession of someone else’s property without the owner’s consent. Think of the finder of lost goods or the policeman who impounds a car. The law recognizes that the goods and the car still belong to their true owners, for “where a person comes into lawful possession of the personal prop- erty of another, even though there is no formal agreement between the property’s owner and its possessor, the pos- sessor will become a constructive bailee when justice so requires.” Christensen v. Hoover, 643 P. 2d 525, 529 (Colo. 1982) (en banc); Laidlaw, Principles of Bailment, 16 Cor- nell L. Q. 286 (1931). At least some of this Court’s deci- sions have already suggested that use of technology is Cite as: 585 U. S. ____ (2018) 17

functionally compelled by the demands of modern life, and in that way the fact that we store data with third parties may amount to a sort of involuntary bailment too. See ante, at 12–13 (majority opinion); Riley v. California, 573 U. S. ___, ___ (2014) (slip op., at 9). Third, positive law may help provide detailed guidance on evolving technologies without resort to judicial intui- tion. State (or sometimes federal) law often creates rights in both tangible and intangible things. See Ruckelshaus v. Monsanto Co., 467 U. S. 986, 1001 (1984). In the context of the Takings Clause we often ask whether those state- created rights are sufficient to make something someone’s property for constitutional purposes. See id., at 1001– 1003; Louisville Joint Stock Land Bank v. Radford, 295 U. S. 555, 590–595 (1935). A similar inquiry may be appropriate for the Fourth Amendment. Both the States and federal government are actively legislating in the area of third party data storage and the rights users enjoy. See, e.g., Stored Communications Act, 18 U. S. C. §2701 et seq.; Tex. Prop. Code Ann. §111.004(12) (West 2017) (defining “[p]roperty” to include “property held in any digital or electronic medium”). State courts are busy expounding common law property principles in this area as well. E.g., Ajemian v. Yahoo!, Inc., 478 Mass. 169, 170, 84 N. E. 3d 766, 768 (2017) (e-mail account is a “form of property often referred to as a ‘digital asset’ ”); Eysoldt v. ProScan Imaging, 194 Ohio App. 3d 630, 638, 2011–Ohio– 2359, 957 N. E. 2d 780, 786 (2011) (permitting action for conversion of web account as intangible property). If state legislators or state courts say that a digital record has the attributes that normally make something property, that may supply a sounder basis for judicial decisionmaking than judicial guesswork about societal expectations. Fourth, while positive law may help establish a person’s Fourth Amendment interest there may be some circum- stances where positive law cannot be used to defeat it. 18 CARPENTER v. UNITED STATES

Ex parte Jackson reflects that understanding. There this Court said that “[n]o law of Congress” could authorize letter carriers “to invade the secrecy of letters.” 96 U. S., at 733. So the post office couldn’t impose a regulation dictating that those mailing letters surrender all legal interests in them once they’re deposited in a mailbox. If that is right, Jackson suggests the existence of a constitu- tional floor below which Fourth Amendment rights may not descend. Legislatures cannot pass laws declaring your house or papers to be your property except to the extent the police wish to search them without cause. As the Court has previously explained, “we must ‘assur[e] preser- vation of that degree of privacy against government that existed when the Fourth Amendment was adopted.’ ” Jones, 565 U. S., at 406 (quoting Kyllo v. United States, 533 U. S. 27, 34 (2001)). Nor does this mean protecting only the specific rights known at the founding; it means protecting their modern analogues too. So, for example, while thermal imaging was unknown in 1791, this Court has recognized that using that technology to look inside a home constitutes a Fourth Amendment “search” of that “home” no less than a physical inspection might. Id., at 40. Fifth, this constitutional floor may, in some instances, bar efforts to circumvent the Fourth Amendment’s protec- tion through the use of subpoenas. No one thinks the government can evade Jackson’s prohibition on opening sealed letters without a warrant simply by issuing a sub- poena to a postmaster for “all letters sent by John Smith” or, worse, “all letters sent by John Smith concerning a particular transaction.” So the question courts will con- front will be this: What other kinds of records are suffi- ciently similar to letters in the mail that the same rule should apply? It may be that, as an original matter, a subpoena requir- ing the recipient to produce records wasn’t thought of as a Cite as: 585 U. S. ____ (2018) 19

“search or seizure” by the government implicating the Fourth Amendment, see ante, at 2–12 (opinion of ALITO, J.), but instead as an act of compelled self-incrimination implicating the Fifth Amendment, see United States v. Hubbell, 530 U. S. 27, 49–55 (2000) (THOMAS, J., dissent- ing); Nagareda, Compulsion “To Be a Witness” and the Resurrection of Boyd, 74 N. Y. U. L. Rev. 1575, 1619, and n. 172 (1999). But the common law of searches and sei- zures does not appear to have confronted a case where private documents equivalent to a mailed letter were entrusted to a bailee and then subpoenaed. As a result, “[t]he common-law rule regarding subpoenas for docu- ments held by third parties entrusted with information from the target is . . . unknown and perhaps unknowable.” Dripps, Perspectives on The Fourth Amendment Forty Years Later: Toward the Realization of an Inclusive Regu- latory Model, 100 Minn. L. Rev. 1885, 1922 (2016). Given that (perhaps insoluble) uncertainty, I am content to adhere to Jackson and its implications for now. To be sure, we must be wary of returning to the doctrine of Boyd v. United States, 116 U. S. 616. Boyd invoked the Fourth Amendment to restrict the use of subpoenas even for ordinary business records and, as JUSTICE ALITO notes, eventually proved unworkable. See ante, at 13 (dissenting opinion); 3 W. LaFave, J. Israel, N. King, & O. Kerr, Crim- inal Procedure §8.7(a), pp. 185–187 (4th ed. 2015). But if we were to overthrow Jackson too and deny Fourth Amendment protection to any subpoenaed materials, we would do well to reconsider the scope of the Fifth Amend- ment while we’re at it. Our precedents treat the right against self-incrimination as applicable only to testimony, not the production of incriminating evidence. See Fisher v. United States, 425 U. S. 391, 401 (1976). But there is substantial evidence that the privilege against self- incrimination was also originally understood to protect a person from being forced to turn over potentially incrimi- 20 CARPENTER v. UNITED STATES

nating evidence. Nagareda, supra, at 1605–1623; Rex v. Purnell, 96 Eng. Rep. 20 (K. B. 1748); Slobogin, Privacy at Risk 145 (2007). * What does all this mean for the case before us? To start, I cannot fault the Sixth Circuit for holding that Smith and Miller extinguish any Katz-based Fourth Amendment interest in third party cell-site data. That is the plain effect of their categorical holdings. Nor can I fault the Court today for its implicit but unmistakable conclusion that the rationale of Smith and Miller is wrong; indeed, I agree with that. The Sixth Circuit was powerless to say so, but this Court can and should. At the same time, I do not agree with the Court’s decision today to keep Smith and Miller on life support and supplement them with a new and multilayered inquiry that seems to be only Katz- squared. Returning there, I worry, promises more trouble than help. Instead, I would look to a more traditional Fourth Amendment approach. Even if Katz may still supply one way to prove a Fourth Amendment interest, it has never been the only way. Neglecting more traditional approaches may mean failing to vindicate the full protec- tions of the Fourth Amendment. Our case offers a cautionary example. It seems to me entirely possible a person’s cell-site data could qualify as his papers or effects under existing law. Yes, the tele- phone carrier holds the information. But 47 U. S. C. §222 designates a customer’s cell-site location information as “customer proprietary network information” (CPNI), §222(h)(1)(A), and gives customers certain rights to control use of and access to CPNI about themselves. The statute generally forbids a carrier to “use, disclose, or permit access to individually identifiable” CPNI without the customer’s consent, except as needed to provide the cus- tomer’s telecommunications services. §222(c)(1). It also Cite as: 585 U. S. ____ (2018) 21

requires the carrier to disclose CPNI “upon affirmative written request by the customer, to any person designated by the customer.” §222(c)(2). Congress even afforded customers a private cause of action for damages against carriers who violate the Act’s terms. §207. Plainly, cus- tomers have substantial legal interests in this infor- mation, including at least some right to include, exclude, and control its use. Those interests might even rise to the level of a property right. The problem is that we do not know anything more. Before the district court and court of appeals, Mr. Carpen- ter pursued only a Katz “reasonable expectations” argu- ment. He did not invoke the law of property or any analo- gies to the common law, either there or in his petition for certiorari. Even in his merits brief before this Court, Mr. Carpenter’s discussion of his positive law rights in cell-site data was cursory. He offered no analysis, for example, of what rights state law might provide him in addition to those supplied by §222. In these circumstances, I cannot help but conclude—reluctantly—that Mr. Carpenter for- feited perhaps his most promising line of argument. Unfortunately, too, this case marks the second time this Term that individuals have forfeited Fourth Amendment arguments based on positive law by failing to preserve them. See Byrd, 584 U. S., at ___ (slip op., at 7). Litigants have had fair notice since at least United States v. Jones (2012) and Florida v. Jardines (2013) that arguments like these may vindicate Fourth Amendment interests even where Katz arguments do not. Yet the arguments have gone unmade, leaving courts to the usual Katz hand- waving. These omissions do not serve the development of a sound or fully protective Fourth Amendment jurisprudence.