The Volokh Conspiracy - Fourth Amendment Protection in Text Pager Messages:

The Ninth Circuit recently held oral argument in a fascinating case on how the Fourth Amendment protects messages sent and received via pagers. The name of the case is United States v. McCreary, and I have posted the brief for the defense here and the brief for the United States here. You can listen to the audio of the oral argument from mid-October before Judges Hug, W. Fletcher, and Clifton from this link. This potentially is a very important case, so I thought I would blog some relatively detailed thoughts about it. Given the usual pace of such things, the opinion probably will be published in a month or two.

  This appeal involves a string of bank robberies McCreary committed in 2002 along with several co-conspirators. Members of the group communicated with each other largely using MCI/Skytel text pagers. The pagers could send and receive messages in various different ways: first, you could enter in the pin # of the pager from another pager and then type in the message; second, you could go to Skytel's website, enter in the pin, and then send a message over the website; and fourth, you could send an e-mail to the PIN number of the text pager at the "skytel.com" domain. At least at the time, in 2002, MCI/Skytel kept records of all messages sent through its pager system.

  Investigators found out about the group's use of MCI/Skytel pagers and knew several of the pin numbers, so they issued subpoenas to get the records of the calls and the text of the messages sent and received. MCI/Skytel returned records including the actual texts of the pager messages. (I found the record a little unclear here, but that's the basic idea.) The text pager messages proved very helpful to the prosecution, as the members of the conspiracy were very open about what they were doing in their pager messages.

  The legal question in United States v. McCreary is whether the government violated McCreary's Fourth Amendment rights by obtaining the text of the pager messages using a subpoena instead of a search warrant. This big question in turn breaks down into three distinct questions. First, does the legitimate user of a text pager system generally have a Fourth Amendment reasonable expectation of privacy in the contents of his messages? Second, if so, do the facts of McCreary's usage of the pager system fall within that general rule of Fourth Amendment protection? And third, if the Fourth Amendment protects the pager messages, did the subpoena violate the Fourth Amendment?

  Let's take these questions in turn. As to the first question, I think the best answer is yes: the legitimate user of a text pager system generally will have Fourth Amendment protection in the contents of his messages. There are a bunch of reasons why, but the best doctrinal reason is explained in what I wrote on this topic in the forthcoming edition of LaFave, Israel, King & Kerr, Criminal Procedure Ch. § 4.3(c) (3d Ed. forthcoming Jan. 2008). The issue was Fourth Amendment protection in text messages and post-cut-through dialed digits — basically, communications you enter using a telephone or other keypad that are contents rather than dialing of numbers. The same analysis seems to apply to text pager messages.

To read the rest of the post, click here.

In my view, the key question is how to distinguish Smith v. Maryland, which ruled that a person has no Fourth Amendment protection in numbers dialed. Here's how I tackled the issue in the treatise:

  There are two basic ways to interpret Smith v. Maryland, although the existing cases have not drawn out this difference. Under one interpretation, the case is merely a routine application of the party to the communication exception. The phone company was a party to Smith's dialing of the digits and therefore could consent to the monitoring. Because the phone company was a party to the communication, the phone company had control of the numbers dialed and could act in concert with the police much like any other party to the communication. Under a second interpretation, the case stands for the broad proposition that numbers dialed from a telephone do not receive Fourth Amendment protection. Smith dialed numbers from his telephone and the government monitored them; what matters for Fourth Amendment purposes is that the information collected were the numbers dialed. . . .
  In traditional telephone surveillance cases, the difference between these two [interpretations] does not matter. However, the difference matters a great deal to cases involving more modern uses of telephones such as the monitoring of post cut-through dialed digits. These are digits entered after the call has been connected -- that is, "cut through" -- and will generally communicate contents to a recipient other than the telephone company. For example, a person might call up a voicemail account and enter in a password or might call the bank and enter the last four digits of a social security number to access an account. Relatedly, a person might send a message to a pager or might send a text message (based on digits) from one cellular phone to another. Under the first interpretation of Smith, the post cut-through digits are contents of communications between the sender and the recipient of the communication; the phone company is no longer a party to the call, so the government cannot use its facilities to monitor the numbers dialed. Under the second reading, the numbers are still only numbers dialed and receive no protection.
  Although cases have not yet resolved which interpretation is correct, the specific language of Smith v. Maryland suggests that the first limited reading is correct. The reasoning of Smith is based heavily on the notion that the phone company was a party to Smith's communication of the numbers: "The switching equipment that processed those numbers is merely the modern counterpart of the operator who, in an earlier day, personally completed calls for the subscriber." In an earlier day, Smith would have had a conversation with the operator; the numbers dialed would have been the contents of the call between him and the operator. Thus the key to Smith is that the phone company was the intended recipient of the call, not that the numbers dialed have some special status as numbers dialed.
  This analysis should lead courts to hold that post cut-through dialed digits and text messages are not covered by Smith and are instead protected under Katz. Because the phone company is not the intended recipient of post cut-through dialed digits or text messages, there is no difference between the phone company listening in on a call and recording post cut-through dialed digits or texts. Both are protected under the Fourth Amendment just like traditional telephone communications.

  The next question is whether McCreary in particular has a reasonable expectation of privacy in the messages. The difference between the general case and the specific one is critically important. For example, a person generally has a reasonable expectation of privacy in his car. However, he won't have such protection if it's a stolen car, or he obtained the car by fraud, or (in most circuits) it's a rental car and his name is not on the rental contract. This reflects a broad principle in Fourth Amendment law: Your reasonable expectation of privacy generally requires a particular relationship with the protected space or information, and whether access violates a reasonable expectation of privacy is contingent on how access is obtained. See, e.g., Minnesota v. Carter (while there is an REP in homes generally, an invited guest lacks an REP when he is in the home only temporarily to bag drugs).

  I couldn't quite tell McCreary's factual relationship with the pager. It sounds like it was a borrowed pager; as long as he had the permission of the owner, he should still have a reasonable expectation of privacy in the messages. If it was a stolen pager, it could be a different story. Also, he should have an REP whether he is the sender or receiver; both retain the same rights, cf. United States v. Villarreal, 963 F.2d 770, 774 (5th Cir. 1992).

  At the oral argument, the argument focused in part on (a) the role of the Stored Communications Act, and (b) the role of Skytel's billing practices. In my view, the Stored Communications Act is just irrelevant. It is a statutory privacy law designed to supplement any existing constitutional protections, but its provisions neither add to nor can take away from existing constitutional protection. It's possible to come up with arguments that the Act actually creates protection, and it's also possible to come up with arguments that the Act should be read to discourage findings of Fourth Amendment protection. Indeed there are snippets of authority for both sides — see pages 435-36 of my Computer Crime casebook for a discussion if you're interested. But in my view neither of these arguments work: the argument that the statute creates protection fails because the statute intentionally creates only lesser protections, and the argument that the statute takes away protection fails because it's ultimately an argument about the passive virtues rather than constitutional meaning. The constitutional and statutory questions must be kept separate.

  The argument about Skytel's billing practices is relevant only insofar as it relates to Skytel's rights to permit third party consent or the private search doctrine. If Skytel had access to all the messages, or even publicized them, then that would impact Skytel's rights to consent to police seeing the text messages or (if the messages were made public, such as posting them on a public website) whether Skytel had conducted a "private search" that exposed the messages to all. Those issues may be indeed relevant in a future case, but I don't see how they are directly relevant here: here the government obtained a subpoena and wasn't seeking third party consent.

  The third and final issue is the trickiest: If McCreary has Fourth Amendment rights in the messages, does a subpoena suffice to compel the information from Skytel? At first blush, the issue seems sort of odd: You're probably thinking, hey, doesn't Fourth Amendment protection mean a warrant? However, the answer turns out to be much more complicated. Courts generally treat subpoenas quite differently as a sort of parallel authority based on reasonableness instead of probable cause. The question is one I discussed extensively in my amicus brief in United States v. Bach in 2002 (see pages 15-24, for the three of you still reading): If the government obtains Fourth Amendment protected information from third parties, is a subpoena sufficient or is probable cause required?

  I have written about this in my Computer Crime Law casebook (pages 437-445), without ever resolving it: I think it's actually maddeningly hard. To see how hard it is, just take a glimpse at the cases involving subpoenas for blood samples. The government wants a blood sample from you, and they cut a subpoena to get it. You want to challenge the subpoena; what's the standard? Obviously it's some kind of search to force you to give up your blood. But what cause is required? State and federal courts are all over the map on this question: some say that a usual reasonableness subpoena is required, others say that a subpoena must also have reasonable suspicion, and others say that the subpoena must be accompanied by probable cause. The problem is hard because it clashes two principles: the well-established low legal regulation of subpoenas, and the general high legal regulation of collecting evidence from inside the body. Which wins out?

  The question of Fourth Amendment access to third party records is similarly hard because it also clashes two principles: the well-established low legal regulation of subpoenas, and the general high legal regulation of private papers. In particular, consider the switch in the legal regulation of access from suspects and third parties. If the government wants to subpoena a document from me directly, the Fourth Amendment offers only very limited protection. I have Fourth Amendment protection, sure, but all I have is reasonableness — generally understood as protection against overly broad, overly burdensome, or downright irrelevant requests. In that setting, it's the Fifth Amendment that protects me instead of the Fourth Amendment: the government is unlikely to subpoena my directly because I normally will have a Fifth Amendment privilege against production (see Hubbell). The Fifth Amendment generally blocks government access.

  Now say I hand of my records to a third party, and the government wants the records from the third party. The Fifth Amendment protections evaporate under the Fisher case. So what happens to the Fourth Amendment protection? Does passing off the document to a third party actually raise the protection from reasonableness to full probable cause, sort of filling in for the loss of Fifth Amendment protection? That seems like a pretty weird result. But then isn't it also weird if you have Fourth Amendment protections that are trivial to overcome with a mere grand jury subpoena?

  It will be interesting to see how the Ninth Circuit handles this exceedingly difficult problem if it reaches it. The now-vacated panel opinion in Warshak tried to draw the line at notice: it ruled that the standard is probable cause without notice to the customer but reasonableness with notice to the custmer. Perhaps that's right, although I would like to hear more about why. I've started an article on Fourth Amendment protection for e-mail that will have to address this issue, but to be honest I don't know where I come out yet (haven't worked out that part yet — ah, the luxuries of being an academic). Some of my earlier work suggested that the rule should be reasonableness, and I think that's probably the most doctrinally simple approach, but I'm not entirely sure where I come out today. The sparse legal scholarship has tended to say a warrant is required, see, e.g., Tricia Bellia's GW article, but this strikes me as influenced by a significant dose of normative preference. Right now I think it's the most difficult and open question in the McCreary case, and I'm very eager to see what the Ninth Circuit will do (and more importantly, how they explain why they did it).

	[Orin Kerr, November 30, 2007 at 6:45pm] Trackbacks Fourth Amendment Protection in Text Pager Messages: The Ninth Circuit recently held oral argument in a fascinating case on how the Fourth Amendment protects messages sent and received via pagers. The name of the case is United States v. McCreary, and I have posted the brief for the defense here and the brief for the United States here. You can listen to the audio of the oral argument from mid-October before Judges Hug, W. Fletcher, and Clifton from this link. This potentially is a very important case, so I thought I would blog some relatively detailed thoughts about it. Given the usual pace of such things, the opinion probably will be published in a month or two. This appeal involves a string of bank robberies McCreary committed in 2002 along with several co-conspirators. Members of the group communicated with each other largely using MCI/Skytel text pagers. The pagers could send and receive messages in various different ways: first, you could enter in the pin # of the pager from another pager and then type in the message; second, you could go to Skytel's website, enter in the pin, and then send a message over the website; and fourth, you could send an e-mail to the PIN number of the text pager at the "skytel.com" domain. At least at the time, in 2002, MCI/Skytel kept records of all messages sent through its pager system. Investigators found out about the group's use of MCI/Skytel pagers and knew several of the pin numbers, so they issued subpoenas to get the records of the calls and the text of the messages sent and received. MCI/Skytel returned records including the actual texts of the pager messages. (I found the record a little unclear here, but that's the basic idea.) The text pager messages proved very helpful to the prosecution, as the members of the conspiracy were very open about what they were doing in their pager messages. The legal question in United States v. McCreary is whether the government violated McCreary's Fourth Amendment rights by obtaining the text of the pager messages using a subpoena instead of a search warrant. This big question in turn breaks down into three distinct questions. First, does the legitimate user of a text pager system generally have a Fourth Amendment reasonable expectation of privacy in the contents of his messages? Second, if so, do the facts of McCreary's usage of the pager system fall within that general rule of Fourth Amendment protection? And third, if the Fourth Amendment protects the pager messages, did the subpoena violate the Fourth Amendment? Let's take these questions in turn. As to the first question, I think the best answer is yes: the legitimate user of a text pager system generally will have Fourth Amendment protection in the contents of his messages. There are a bunch of reasons why, but the best doctrinal reason is explained in what I wrote on this topic in the forthcoming edition of LaFave, Israel, King & Kerr, Criminal Procedure Ch. § 4.3(c) (3d Ed. forthcoming Jan. 2008). The issue was Fourth Amendment protection in text messages and post-cut-through dialed digits — basically, communications you enter using a telephone or other keypad that are contents rather than dialing of numbers. The same analysis seems to apply to text pager messages. To read the rest of the post, click here. In my view, the key question is how to distinguish Smith v. Maryland, which ruled that a person has no Fourth Amendment protection in numbers dialed. Here's how I tackled the issue in the treatise: There are two basic ways to interpret Smith v. Maryland, although the existing cases have not drawn out this difference. Under one interpretation, the case is merely a routine application of the party to the communication exception. The phone company was a party to Smith's dialing of the digits and therefore could consent to the monitoring. Because the phone company was a party to the communication, the phone company had control of the numbers dialed and could act in concert with the police much like any other party to the communication. Under a second interpretation, the case stands for the broad proposition that numbers dialed from a telephone do not receive Fourth Amendment protection. Smith dialed numbers from his telephone and the government monitored them; what matters for Fourth Amendment purposes is that the information collected were the numbers dialed. . . . In traditional telephone surveillance cases, the difference between these two [interpretations] does not matter. However, the difference matters a great deal to cases involving more modern uses of telephones such as the monitoring of post cut-through dialed digits. These are digits entered after the call has been connected -- that is, "cut through" -- and will generally communicate contents to a recipient other than the telephone company. For example, a person might call up a voicemail account and enter in a password or might call the bank and enter the last four digits of a social security number to access an account. Relatedly, a person might send a message to a pager or might send a text message (based on digits) from one cellular phone to another. Under the first interpretation of Smith, the post cut-through digits are contents of communications between the sender and the recipient of the communication; the phone company is no longer a party to the call, so the government cannot use its facilities to monitor the numbers dialed. Under the second reading, the numbers are still only numbers dialed and receive no protection. Although cases have not yet resolved which interpretation is correct, the specific language of Smith v. Maryland suggests that the first limited reading is correct. The reasoning of Smith is based heavily on the notion that the phone company was a party to Smith's communication of the numbers: "The switching equipment that processed those numbers is merely the modern counterpart of the operator who, in an earlier day, personally completed calls for the subscriber." In an earlier day, Smith would have had a conversation with the operator; the numbers dialed would have been the contents of the call between him and the operator. Thus the key to Smith is that the phone company was the intended recipient of the call, not that the numbers dialed have some special status as numbers dialed. This analysis should lead courts to hold that post cut-through dialed digits and text messages are not covered by Smith and are instead protected under Katz. Because the phone company is not the intended recipient of post cut-through dialed digits or text messages, there is no difference between the phone company listening in on a call and recording post cut-through dialed digits or texts. Both are protected under the Fourth Amendment just like traditional telephone communications. The next question is whether McCreary in particular has a reasonable expectation of privacy in the messages. The difference between the general case and the specific one is critically important. For example, a person generally has a reasonable expectation of privacy in his car. However, he won't have such protection if it's a stolen car, or he obtained the car by fraud, or (in most circuits) it's a rental car and his name is not on the rental contract. This reflects a broad principle in Fourth Amendment law: Your reasonable expectation of privacy generally requires a particular relationship with the protected space or information, and whether access violates a reasonable expectation of privacy is contingent on how access is obtained. See, e.g., Minnesota v. Carter (while there is an REP in homes generally, an invited guest lacks an REP when he is in the home only temporarily to bag drugs). I couldn't quite tell McCreary's factual relationship with the pager. It sounds like it was a borrowed pager; as long as he had the permission of the owner, he should still have a reasonable expectation of privacy in the messages. If it was a stolen pager, it could be a different story. Also, he should have an REP whether he is the sender or receiver; both retain the same rights, cf. United States v. Villarreal, 963 F.2d 770, 774 (5th Cir. 1992). At the oral argument, the argument focused in part on (a) the role of the Stored Communications Act, and (b) the role of Skytel's billing practices. In my view, the Stored Communications Act is just irrelevant. It is a statutory privacy law designed to supplement any existing constitutional protections, but its provisions neither add to nor can take away from existing constitutional protection. It's possible to come up with arguments that the Act actually creates protection, and it's also possible to come up with arguments that the Act should be read to discourage findings of Fourth Amendment protection. Indeed there are snippets of authority for both sides — see pages 435-36 of my Computer Crime casebook for a discussion if you're interested. But in my view neither of these arguments work: the argument that the statute creates protection fails because the statute intentionally creates only lesser protections, and the argument that the statute takes away protection fails because it's ultimately an argument about the passive virtues rather than constitutional meaning. The constitutional and statutory questions must be kept separate. The argument about Skytel's billing practices is relevant only insofar as it relates to Skytel's rights to permit third party consent or the private search doctrine. If Skytel had access to all the messages, or even publicized them, then that would impact Skytel's rights to consent to police seeing the text messages or (if the messages were made public, such as posting them on a public website) whether Skytel had conducted a "private search" that exposed the messages to all. Those issues may be indeed relevant in a future case, but I don't see how they are directly relevant here: here the government obtained a subpoena and wasn't seeking third party consent. The third and final issue is the trickiest: If McCreary has Fourth Amendment rights in the messages, does a subpoena suffice to compel the information from Skytel? At first blush, the issue seems sort of odd: You're probably thinking, hey, doesn't Fourth Amendment protection mean a warrant? However, the answer turns out to be much more complicated. Courts generally treat subpoenas quite differently as a sort of parallel authority based on reasonableness instead of probable cause. The question is one I discussed extensively in my amicus brief in United States v. Bach in 2002 (see pages 15-24, for the three of you still reading): If the government obtains Fourth Amendment protected information from third parties, is a subpoena sufficient or is probable cause required? I have written about this in my Computer Crime Law casebook (pages 437-445), without ever resolving it: I think it's actually maddeningly hard. To see how hard it is, just take a glimpse at the cases involving subpoenas for blood samples. The government wants a blood sample from you, and they cut a subpoena to get it. You want to challenge the subpoena; what's the standard? Obviously it's some kind of search to force you to give up your blood. But what cause is required? State and federal courts are all over the map on this question: some say that a usual reasonableness subpoena is required, others say that a subpoena must also have reasonable suspicion, and others say that the subpoena must be accompanied by probable cause. The problem is hard because it clashes two principles: the well-established low legal regulation of subpoenas, and the general high legal regulation of collecting evidence from inside the body. Which wins out? The question of Fourth Amendment access to third party records is similarly hard because it also clashes two principles: the well-established low legal regulation of subpoenas, and the general high legal regulation of private papers. In particular, consider the switch in the legal regulation of access from suspects and third parties. If the government wants to subpoena a document from me directly, the Fourth Amendment offers only very limited protection. I have Fourth Amendment protection, sure, but all I have is reasonableness — generally understood as protection against overly broad, overly burdensome, or downright irrelevant requests. In that setting, it's the Fifth Amendment that protects me instead of the Fourth Amendment: the government is unlikely to subpoena my directly because I normally will have a Fifth Amendment privilege against production (see Hubbell). The Fifth Amendment generally blocks government access. Now say I hand of my records to a third party, and the government wants the records from the third party. The Fifth Amendment protections evaporate under the Fisher case. So what happens to the Fourth Amendment protection? Does passing off the document to a third party actually raise the protection from reasonableness to full probable cause, sort of filling in for the loss of Fifth Amendment protection? That seems like a pretty weird result. But then isn't it also weird if you have Fourth Amendment protections that are trivial to overcome with a mere grand jury subpoena? It will be interesting to see how the Ninth Circuit handles this exceedingly difficult problem if it reaches it. The now-vacated panel opinion in Warshak tried to draw the line at notice: it ruled that the standard is probable cause without notice to the customer but reasonableness with notice to the custmer. Perhaps that's right, although I would like to hear more about why. I've started an article on Fourth Amendment protection for e-mail that will have to address this issue, but to be honest I don't know where I come out yet (haven't worked out that part yet — ah, the luxuries of being an academic). Some of my earlier work suggested that the rule should be reasonableness, and I think that's probably the most doctrinally simple approach, but I'm not entirely sure where I come out today. The sparse legal scholarship has tended to say a warrant is required, see, e.g., Tricia Bellia's GW article, but this strikes me as influenced by a significant dose of normative preference. Right now I think it's the most difficult and open question in the McCreary case, and I'm very eager to see what the Ninth Circuit will do (and more importantly, how they explain why they did it).