Pittsburgh Decision on Warrant Requirement for Historical Cell Site Records:
A few readers have e-mailed me about the Pittsburgh district court decision requiring a probable cause warrant for stored cell site records. The court affirmed a magistrate's ruling that the Stored Communications Act doesn't govern historical cell site records as a statutory matter, and thus that the Rule 41 probable cause warrant requirement applied by default. The court added that this statutory reading was bolstered by the likelihood that the Fourth Amendment also protected the information.
I don't have time for a full post on this, but the Pittsburgh decision is plainly incorrect. Of course the Stored Communications Act covers this: It's a "record concerning an electronic communication service" under 18 U.S.C. 2703(c) which can be compelled with a Terry stop "specific and articulable facts" court order under 18 U.S.C. 2703(c)(2), not a warrant. Also, the notion that the Fourth Amendment protects cell-site info is just pretty clearly wrong under the Supreme Court's decision in Smith v. Maryland. A cell site signal is closely analogous to numbers dialed in Smith: It's a signal that the user sends to the phone company that is necessary for the phone company to deliver the user's calls.
Some have tried to argue that cell site data is different than numbers dialed because (a) some people think that cell phones work by magic, rather than by sending communications to cell towers to let the provider know where the phone is located, and (b) location information is more private than numbers dialed information. But these arguments don't work, I think. First, it's hard to see why the Fourth Amendment should protect a user's failure to have a basic understanding of how technology works. Second, the numbers dialed from a landline phone also give location information: In fact, they tell the police that the person is inside their home, the most private of all places under the Fourth Amendment. The Smith court didn't think this was relevant, though, so under Smith I don't think it's relevant here, either.
Critically, this doesn't mean that historical cell site data should receive no protection. Historical cell site data should and does receive the protection of the Stored Communications Act, which requires a court order based on a showing of specic and articulable facts to believe the information would be relevant and material to an ongoing criminal investigation. But under current law, a warrant shouldn't be required. Anyway, I hope the U.S. will appeal the decision; I would guess the Third Circuit will look at this differently. And for more on the Fourth Amendment issues here, you might be interested in my draft article, The Case for the Third Party Doctrine.
I don't have time for a full post on this, but the Pittsburgh decision is plainly incorrect. Of course the Stored Communications Act covers this: It's a "record concerning an electronic communication service" under 18 U.S.C. 2703(c) which can be compelled with a Terry stop "specific and articulable facts" court order under 18 U.S.C. 2703(c)(2), not a warrant. Also, the notion that the Fourth Amendment protects cell-site info is just pretty clearly wrong under the Supreme Court's decision in Smith v. Maryland. A cell site signal is closely analogous to numbers dialed in Smith: It's a signal that the user sends to the phone company that is necessary for the phone company to deliver the user's calls.
Some have tried to argue that cell site data is different than numbers dialed because (a) some people think that cell phones work by magic, rather than by sending communications to cell towers to let the provider know where the phone is located, and (b) location information is more private than numbers dialed information. But these arguments don't work, I think. First, it's hard to see why the Fourth Amendment should protect a user's failure to have a basic understanding of how technology works. Second, the numbers dialed from a landline phone also give location information: In fact, they tell the police that the person is inside their home, the most private of all places under the Fourth Amendment. The Smith court didn't think this was relevant, though, so under Smith I don't think it's relevant here, either.
Critically, this doesn't mean that historical cell site data should receive no protection. Historical cell site data should and does receive the protection of the Stored Communications Act, which requires a court order based on a showing of specic and articulable facts to believe the information would be relevant and material to an ongoing criminal investigation. But under current law, a warrant shouldn't be required. Anyway, I hope the U.S. will appeal the decision; I would guess the Third Circuit will look at this differently. And for more on the Fourth Amendment issues here, you might be interested in my draft article, The Case for the Third Party Doctrine.
Your analysis of the case sounds right to me, but there's something that "feels" strange about the argument I quote above. Yes, I suppose that landline numbers inherently transmit location information. But on the other hand, couldn't a court hold that all that Smith holds is that the police can get the number, and there's no holding re: location information? It's almost like the Catholic principle of double effect-- there might be a difference between incidentally obtaining the location information because there's no way to not receive it when obtaining the number, and actually obtaining location information independent of the number.
Not sure if there's a policy argument in favor of recognizing that distinction, though (other than a general sense that it is more privacy-protective).
This is certainly not the case where the conveyance of information to the third party is not a strict and obvious requirement of the functionality of the technology. In the case of cell-phones, substantially similar functionality could be obtained without cell site information (i.e. only E911 and similar services would be lost). The conveyal of cell site information is mainly for the benefit of the cellphone company's security and efficiency of the network. It's entirely possible that there are alternate technologies, which convey different degrees or kinds of information to the 3rd party, but appear functionally identical tot he end user. The result is the 4th amendment question turns on the specific type of technology, not the end user's reasonable expectations. In many cases the implementation will be unknown or unknowable by the end user (i.e. what degree/type of information is conveyed to the 3rd party is a trade secret).
When the type of information conveyed to third parties is limited in potential expressiveness, there is little opportunity for abuse. When the criminal has no way of knowing the information is sent to the third party, there is no potential for abuse.
Except since the rules for 3rd parties and location are already a mess, no clarity is gained from matching them. When does a guest gain a reasonable expectation of privacy? Probably if staying overnight, but how many hours, and as long as they aren't counting drugs and money? It's pretty much a whatever the court feels like test.
That would be a great argument, except for the fact that even EFF, ACLU, and CDT explicitly rejected it and conceded in their amicus brief (see page 7) the applicability of the Stored Communications Act.
Ian Samuel, Warrantless Location Tracking, 83 N.Y.U. L. REv. (forthcoming October 2008), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1092293.
The short version: it's at least ambiguous whether the electronic surveillance statutes cover this information. The Fourth Amendment argument thus informs the reading of these statutes. Smith just isn't as broad as Professor Kerr suspects. Revealing that someone somewhere is dialing phone numbers from their home is very different indeed from being able to track their movements, as Knotts and Karo firmly establish.
With all due respect, that's gibberish. The statutes say what they say; the meaning of none of the relevant provisions turns on any question of constitutional law.
If what you mean is "the Fourth Amendment protects this information, and the statutes should therefore be read narrowly under the doctrine of avoidance," then say so. Of course, that still obliges you to come up with a cogent Fourth Amendment argument, about which more below.
as Knotts and Karo firmly establish
Actually, Karo holds explicitly that the police can use a tracking device to locate an item in a multi-unit storage facility -- indeed, even track it to a specific row of lockers -- without engaging in a "search" or otherwise infringing the Fourth Amendment, provided they don't track it to a specific locker or other closed private space. In light of that fact, the claim that Knotts and Karo "firmly establish" a rule against tracking someone's movements is pure bunk.
Amicus briefs sure have a lot of precedential weight. I could concede your interest groups, and raise you a more weighty W.D. Pa. opinion.
It's rather simple, excluded info can't be reincorporated.
I believe you're incorrect: Smith v. Maryland was correctly decided, and in fact is a necessary flip side to Katz v. United States. If you think Katz was correct, you should also agree with Smith; both are about ensuring the technological neutrality of the Fourth Amendment. See my article for the details.
Plus, even if you think Smith (or Katz) was incorrect, it's not the job of a trial judge in Pittsburgh to announce that that he is rejecting several decades' of Supreme Court decisions. Lower courts have to follow the Supreme Court, not their personal visions of what they want the Constitution to mean.
I had no idea this case was going on. Interesting!
I don't think anyone else will find it that interesting.
[For the record, Judge Lenihan is very nice and a great teacher. Of course, that has nothing to do with this matter.]
Well, maybe you're right, but I didn't say that they did establish such a rule. You're speaking at much too high a level of generality, anyway. How could there be a "rule against tracking someone's movements" or a "rule allowing tracking someone's movements?" Neither one could be squared with the relevant statutory or constitutional law.
I said that Knotts and Karo establish that tracking someone's movements is different than recording the telephone numbers that they dial. They do so establish. I therefore don't think that the pen register cases are the first and last word on these matters.
Why not? If there is an open question of law and novel facts to which the law would apply, why can't a lower court judge stick his neck out? He is independently appointed, too. He too has an obligation to the Constitution. If he gets it wrong, the Supreme Court can correct him.
It should be noted that the GPS even allows tracking between rooms in your own house. I think that's a violation of your privacy. Records of when you were home making phone calls is of course a violation of your privacy, but not that big of one since the fact that your home is yours generally isn't a secret, and people generally expect a person to be at their own home from time to time. GPS also allows determining what doctor and lawyer you've been visiting. Knowing what examination room your doctor took you to can reveal information about what medical conditions you have. The amount of private information that location tracking can reveal is enormous.
This case isn't about GPS data, or about ongoing/real-time tracking. It's about past records of which cell towers carried particular calls for a given phone. Those records aren't even close to pinpoint accuracy -- more like an approximate range of hundreds of yards.
It's not a tracking device because section 3117 "governs the 'installation' of tracking devices. The 'tracking' of a cell phone does not require the installation of any sort of device." In re Applications, 509 F. Supp. 2d 76, 81 n.11 (D. Mass. 2007); accord In re Application, 405 F. Supp. 2d 435, 449 n.8 (S.D.N.Y. 2005); In re Application, 460 F. Supp. 2d 448, 461 (S.D.N.Y. 2006).
Pages 11-17 of the government's initial brief would seem to address your question at length. Footnote 9 (p. 17) points out why Rule 41 is inapplicable, and in the alternative why the Rule doesn't even impose a probable cause requirement on orders for tracking devices (quoting from the 2006 Advisory Committee notes).
The district judge just rubber-stamped the magistrate's ruling, actually, so we're really critiquing the decision of a magistrate judge. As for why I mention Pittsburgh, I suppose I am being biased by my experience as a Third Circuit clerk a decade ago: I suspect most Third Circuit clerks, past and present, will get its significance.
I recall back in the 1980s when Ruggero Aldisert was Third Circuit Chief Judge - with chambers in Pgh. - he was once overheard complaining to a tier-1 law professor that he wasn't seeing clerkship applicants of nearly as high quality as his brethern in Philthy-delphia. Apparently anti-Yinzer bias and/or residual prejudice stemming from Pgh's. old image as "He** with the lid off" inhibited some from applying for even a really good clerkship opportunity in that town unless they either had a current connection to the city (e.g., attending tier-2 Pitt Law School) or grew up there.
Second, I don't think it ultimately matters, but I was intrigued by the government's claim that cell site data is just a record of phone calls, and therefore records about wire communications. Doesn't cell site data include all of the cell phone's "here I am" chirps to the tower? If so, those are clearly not wire (voice) communications, and then the magistrate judge's argument that they're also excluded from the definition of "electronic communications" kicks in.
I think I buy the government's argument, though, that they would still "pertain to" the subscriber, regardless of whether they concerned electronic communications or not.
No. A relevant passage from the government's brief:
Some random guy,
Just as you cited amicus briefs, I take issue with the portion of the government's brief you cited. Cell phone emit their signals to cell site towers constantly, approximately every seven seconds to ensure the phone has the best signal strength. This occurs regardless of whether a call has been placed. To be sure, this does not occur if the user turns the phone OFF.
Gosh, that's really incredible, given that the FCC accuracy requirement for 95% of calls is 300 meters. (See 47 C.F.R. § 20.18(h)(1).) Who makes up these facts for you?
Cell phone emit their signals to cell site towers constantly, approximately every seven seconds to ensure the phone has the best signal strength.
Whether that's true or not -- and it's pretty unlikely, given the heavy power drain &correspondingly reduced battery life that would result -- is irrelevant because no running record is kept.