The Volokh Conspiracy

A few days ago, Ryan Lizza had a story in the New Yorker detailing litigation over whether the search warrant to search James Rosen’s e-mail account required that the government notify Rosen of the search. The procedural history is actually kind of complicated, but the key idea is that Magistrate Judges Kay and Facciola thought that the statute required notice to Rosen (which could be delayed on a finding of need) when it obtained the warrant to access his e-mail account. The government appealed that ruling to District Judge Royce Lamberth, who disagreed with the magistrate judges and signed the application allowing the search warrant without notice.

Lizza’s story mostly just reports what happened, without a lot of spin. This morning, however, the Drudge Report has a top-left link to Breitbart’s coverage by Larry O’Connor suggesting that something nefarious is afoot. The Breitbart story says that the DOj went “judge shopping” and that “the effort by the Justice Department to obtain the controversial court order was arduous, contentious, and unsuccessful until finally a third judge acquiesced.” It concludes:

The revelation that two courts denied the secret subpoena before Lamberth finally agreed will damage the narrative that there was nothing extraordinary or out-of-bounds about Holder’s attempt to delve into the private communications of Rosen and his employer.

Putting aside the minor errors in this sentence — it was a warrant not a subpoena, for example — I think it’s important to realize that the magistrate judges were pretty clearly wrong in their interpretation of the law.

The relevant law, the Stored Communications Act, does not require notice to the customer or subscriber when the government obtains a search warrant to search an online account. Under the statute, prior notice is required if the government obtains contents with less process than a warrant. But when the government obtains a warrant, no notice is required. See, e.g., 18 U.S.C. 2703(b)(1) (” A governmental entity may require a provider of remote computing service to disclose the contents of any wire or electronic communication to which this paragraph is made applicable by paragraph (2) of this subsection—(A) without required notice to the subscriber or customer, if the governmental entity obtains a warrant issued using the procedures described in the Federal Rules of Criminal Procedure (or, in the case of a State court, issued using State warrant.”) (emphasis added). Federal Rule of Criminal Procedure 41 has a notice requirement, to be sure, but it is only that the notice be served where the property was taken — which in the case of an e-mail account held by an ISP, means notice to the ISP not the customer. See generally In re Application of the United States for a Search Warrant for the Contents of Electronic Mail, 665 F. Supp.2d 1210 (D. Or. 2009).

Of course, that does not mean that the existing statute strikes the right balance as a matter of policy. There are substantial arguments that Congress should revisit the law and add a notice requirement. Indeed, there are bills pending in Congress that would do just that. But as a matter of current statutory law, I think it’s pretty clear that the notice requirement did not apply and that Magistrate Judges Facciola and Kay simply misread the statute and Rule 41 when they concluded that it did.

Based on past posts on this topic, I gather that some readers will find my legal analysis pedantic and irrelevant. Understanding the law does not change the fact that the government obtained Rosen’s e-mail without giving him notice, and that the government was willing to go to the District Court to ensure that. While that’s true, my sense is that an understanding of the law here does put a different spin on the litigation. The magistrate judges misread the statute, and the government sought review of their error so that the Rosen warrant would be treated like every other e-mail warrant. My sense is that this context makes the story less “extraordinary and out-of-bounds” than it would be if the government’s legal argument had been an aggressive one that pushed the envelope for this specific case.

Congratulations to Sri Srinivasan for being confirmed today to a judgeship on the DC Circuit. The final vote was a squeaker — 97-0.

There is going to be a lot of speculation that Srinivasan may be nominated to the Supreme Court someday, so it’s perhaps worth noting that Justice Ginsburg has indicated that she is not going to retire this coming year. If Ginsburg announces her retirement at the end of the next Supreme Court term in late June 2014, however, Srivinasan would have served slightly over a year on the DC Circuit. By comparison, Justice Thomas served 16 months on that court before being nominated to the Supremes.

Ian Millhiser speculates about future Democratic Supreme Court nominees over at ThinkProgress, naming names. Some of the names Ian mentions strike me as very very plausible; others strike me as very very implausible. But I’m not privy to the thinking of either current or future Democratic administrations, so who knows.

There has been a lot of outrage expressed recently over the contents of an affidavit filed in support of a search warrant to search the e-mail accounts of reporter James Rosen. The government’s affidavit offered the view that Rosen violated the law by aiding and abetting the alleged violations of laws prohibiting the disclosure of classified national security information. Specifically, the affidavit stated, “there is probable cause to believe that the Reporter . . . has committed a violation of 18 U.S.C. 793(d) either as Mr. Kim’s co-conspirator and/or aider and abetter.” To some, the fact that the government would make this argument shows that the Obama Administration is engaging in a War on Journalism. According to this thinking, the Obama Administration is not only trampling on the rights of a free press by going after its sources. Incredibly, they even think of a reporter as a criminal — and are willing to say so in court.

I think you get a different sense of this affidavit if you understand the privacy laws, however, and in particular a federal law called the Privacy Protection Act, 42 U.S.C. 2000aa. It’s a pretty complicated law, so it will take me a minute to explain. But I think I can explain why the affidavit filed in the Rosen search had this language, and why claims that this language reveals a “war on journalism” are based on a misunderstanding.

Here’s the background. The Privacy Protection Act was a response to a Supreme Court decision called Zurcher v. Stanford Daily, 436 U.S. 547 (1978). In Zurcher, the Supreme Court held that the Fourth Amendment permits the government to execute a search warrant at a newspaper to recover evidence of a crime that reporters had gathered in the course of investigating the news. The Supreme Court reasoned that warrants can be issued when there is probable cause to believe evidence of a crime is located in the place to be searched, and there is no Fourth Amendment principle that third-parties are exempt from such searches.

You can see the major problem created by Zurcher. If the government is allowed to execute warrants wherever the evidence is, and reporters often have to gather evidence of crime in the course of gathering news, then the government can get lazy and just raid reporters’ homes and offices to find out what the reporters have learned rather than conduct the government’s own investigation.

That problem hadn’t arisen historically because the Supreme Court had earlier adopted the “mere evidence” rule, by which the government was prohibited from obtaining warrants to collect mere evidence; the government only had the power to collect contraband, stolen goods, or the fruits of crime, which necessarily excluded searches of the offices of reporters for evidence of crime. But the Supreme Court had overturned the mere evidence rule in 1967 in Warden v. Hayden, setting up the question in Zurcher eleven years later. So the combination of Hayden and Zurcher meant that the Fourth Amendment allowed the government to target innocent reporters who just happened to to have gathered evidence of crime in the course of gathering the news.

Congress enacted the Privacy Protection Act just two years after Zurcher. The Privacy Protection Act is pretty complicated, and its exceptions have exceptions to its exceptions. But it contains pretty specific guidance for when the government can conduct searches to gather evidence from reporters made in the course of gathering the news in cases involving the disclosure of classified information. The rule is that the government can only do that when “there is probable cause to believe that the person possessing such materials has committed or is committing the criminal offense to which the materials relate.” 42 U.S.C. 2000aa(a). The idea is that the government can’t gather evidence from journalists who happen to have evidence of that crime in their possession except in the rare cases in which there is probable cause to believe that the journalist is involved in the crime, too. In other words, investigators generally can’t take the short-cut of going to the press for the evidence — but when there is probable cause that the press is involved in the crime, the usual Fourth Amendment rules apply.

That brings us to the affidavit filed in the Rosen case. If you read the affidavit, it is pretty clear that the part discussing Rosen’s own potential criminality is designed to show compliance with the Privacy Protection Act. Here’s paragraph 46, the key paragraph that has caused so much concern:

Based on the above, there is probable cause to believe that the Reporter (along with Mr. Kim) has committed a violation of 18 U.S.C. 793(d) either as Mr. Kim’s co-conspirator and/or aider and abetter, and that the evidence of crime is likely to be contained in the [ ]@gmail.com account. Accordingly, the FBI’s request to search the contents of that account falls squarely within section 2000aa(a)’s exception permitting searches of media-related work product materials, even when possessed by a national news reporter because there is “probable cause to believe that the person possessing such materials has committed or is committing the criminal offense to which the materials relate.” 42 U.S.C. 2000aa(a).

The first sentence in that paragraph has caused a lot of anxiety among reporters. But the first sentence is just the set-up for the second sentence, in which the affidavit explains that this matters because it triggers the exception under the Privacy Protection Act. (For a discussion of the underlying criminal law, and in particular the interesting question of whether the laws on disclosing national security information can apply to reporters under the First Amendment, see this excellent post by Eugene.)

I am only making a narrow point, I should stress. I am not arguing that investigators made a good decision when it obtained Rosen’s e-mail. I am not expressing a view about whether the Privacy Protection Act’s exception is too broad, or what the law or policy of criminal investigations involving the media should be. My point is just that the controversial language in the affidavit shouldn’t be read out of context. That language was necessary to demonstrate compliance with a privacy law Congress enacted to provide safeguards for the press.

UPDATE: I tried to stress the narrowness of the post, but it seems I have failed. Let me try again. Different people have voiced different objections to the investigation of Rosen’s e-mails. Some have argued that the investigation is wrong because the government should not obtain evidence from journalists — or at least e-mails — at all. Others have argued that the affidavit is troubling because it reveals DOJ’s apparent view about substantive law that a journalist who asks a source to disclose classified information with the goal of publishing it has committed a federal crime.

This post is responding to a different argument: That the fact of labeling a journalist a criminal in a court filing demonstrates an effort to intimidate the press and further a “war on journalism” by treating journalists like criminals. See, for example, this CBS News story today from Jan Crawford. The argument I’m responding to (as seen in Crawford’s piece) is that labeling a journalist a criminal reflects a hidden motive to intimidate reporters. My sense is that an awareness of the relevant law puts a different perspective on the language, as the motives behind developing a point in an affidavit that is required under the law may be different from developing a point gratuitously when it is not required by law. When a point required by law is developed in an affidavit, it may have been added with the relatively simple motive of complying with the law.

Today Lois Lerner was called to testify before a House Committee about the recent scandal involving the IRS targeting conservative groups for extra scrutiny. Before the Committee, Lerner was invited to make an opening statement. Her opening statement included the following:

On May 14th, the Treasury inspector general released a report finding that the Exempt Organizations field office in Cincinnati, Ohio used inappropriate criteria to identify for further review applications from organizations that planned to engage in political activity, which may mean that they did not qualify for tax exemption.

On that same day, the Department of Justice launched an investigation into the matters described in the inspector general’s report. In addition, members of this committee have accused me of providing false information when I responded to questions about the IRS processing of applications for tax exemption.

I have not done anything wrong. I have not broken any laws. I have not violated any IRS rules or regulations, and I have not provided false information to this or any other congressional committee.

And while I would very much like to answer the committee’s questions today, I’ve been advised by my counselto assert my Constitutional right not to testify or answer questions related to the subject matter of this hearing.

After very careful consideration, I’ve decided to follow my counsel’s advice, and not testify or answer any of the questions today.

Because I’m asserting my right not to testify, I know that some people will assume that I’ve done something wrong. I have not.

Under questioning, Lerner then authenticated answers she gave I.G. investigators about what had happened during the investigation. But she refused to make additional statements.

The Chairman of the committee, Darrell Issa, has argued that Lerner’s statement and her authentication of her earlier statements waived her Fifth Amendment right and that he can call her to testify again without Fifth Amendment protection. A lot of people are wondering, is Issa right? Did Lerner waive her rights so she could not then assert them?

I don’t think the answer is clear, as there are no cases quite like it. The general rule is that a witness can’t testify about her version of the facts and then invoke the Fifth Amendment when facing cross examination. Here’s what the Court said in Mitchell v. United States, 526 U.S. 314, 321(1999):

It is well established that a witness, in a single proceeding, may not testify voluntarily about a subject and then invoke the privilege against self-incrimination when questioned about the details. See Rogers v. United States, 340 U.S. 367, 373 (1951). The privilege is waived for the matters to which the witness testifies, and the scope of the “waiver is determined by the scope of relevant cross-examination,” Brown v. United States, 356 U.S. 148, 154—155 (1958). “The witness himself, certainly if he is a party, determines the area of disclosure and therefore of inquiry,” id., at 155. Nice questions will arise, of course, about the extent of the initial testimony and whether the ensuing questions are comprehended within its scope, but for now it suffices to note the general rule.

The tricky part is how to characterize Lerner’s testimony before she invoked the Fifth Amendment. On one hand, if you say that Lerner merely expressed her view that she is innocent but did not actually testify as to any facts, then you could say she did not waive her rights with her statement. Questioning would not be about the details of facts she already testified to, but rather would require her testimony on a subject she declined to testify about. On the other hand, if you say that Lerner’s reciting the allegations and then denying them effectively testified about the allegations, then you could say that she did testify and did waive her rights. From that perspective, she already testified about “the subject” by saying that she did not violate any IRS rules or submit false testimony, and further questioning would be about the details of why she thinks that.

I’m not enough of a Fifth Amendment nerd to have strong views on which side is right. So I posed the question earlier today (based on press reports of what Lerner said, not the full transcript) to a listserv of criminal procedure professors that includes some serious Fifth Amendment experts. Opinions were somewhat mixed, but I think it’s fair to say that the bulk of responders thought that Lerner had not actually testified because she gave no statements about the facts of what happened. If that view is right, Lerner successfully invoked her Fifth Amendment rights and cannot be called again. But this was not a unanimous view, it was not based on the full transcript, and there are no cases that seem to be directly on point. So it’s at least a somewhat open question.

UPDATE: I have updated the post to include more from the transcript of the hearing.

Here’s an interesting new application of Florida v. Jardines (March 26, 2013), which recently held that bringing a drug-sniffing dog up to the front door of a home for a “sniff” of the front door area is a Fourth Amendment search. In this morning’s decision in Powell v. State of Florida, officers went up to the front door of a mobile home and knocked. When there was no answer, they took a step off the front steps and peered through a window at eye level about two feet from the front steps. Peering in, the officers saw marijuana growing under lamps in the home; they later obtained a warrant based on that viewing and searched the home.

The state intermediate court held that peering into the window was a Fourth Amendment search that required a warrant. Going up to the front door and knocking was fine, but peering through the window off of the front steps was not okay under the Jones trespass/intrusion test:

The deputies . . . deviated from established norms by entering upon that portion of the property directly in front of the window. Nothing in their testimony or the record establishes any license to do that. The officers had to step off the front door step, move two feet to the left, and position themselves directly in front of the window, their faces no more than a foot away. At that point they were virtually within the home without breaking its close. Because they physically entered a part of the curtilage where they had no right to be for the purpose of gaining information, the intrusion test is met.

The court notes that the Jones trespass/intrusion test is the easiest way to resolve the case: “Cases involving warrantless searches of the home or curtilage may be somewhat easier to analyze under the intrusion approach because the property interests are generally better defined.” (The court also concludes that the Jones test is best characterized as being about physical “intrusion” not “trespass.” See Footnote 3.) The court also holds that this was a search under the Katz privacy test:

Similarly, the State’s evidence lacked any indicia that the privacy of the mobile home’s kitchen area had been diminished by its occupants. No evidence exists that Powell or Wilbourn knowingly exposed the interior of their home where the plants were located to the public view or impliedly licensed the general public to peer in their front window from a foot away. No evidence was presented that the kitchen area (where the plants were located) could be seen from the public roadway, from the pathway leading to the front door, or from the front door itself. Instead, the plants could only be seen from outside the home by stepping away from the front door, placing officers within a hand’s width of the window pane, casting their view rightward at an acute angle. As in Olivera, Powell and Wilbourn “could reasonably expect that no one would observe or overhear [their] activities” from just outside their window.

It’s true that the window was right near the front door. But the fact that the officers stepped off the front porch meant that it exceeded the implied permission of the homeowners:

We cannot agree . . . that stepping off a porch, even a few feet, onto portions of the curtilage where persons are uninvited and then looking into the home at a sharp angle from a hand’s length away from the window pane is anything other than an impermissible intrusion into constitutionally protected space. Whether two feet or twenty, the distance between the door and window matters little given that the officers said they could not see the plants without leaving the front door step and positioning themselves at a spot where they had no right to be.

Does that sound like a fine line? Well, welcome to the fact-sensitive Fourth Amendment:

That said, we can envision front door configurations that have windows incorporated directly into their designs through which a visitor might be able to see the interior using no unusual means or devices. This case simply does not involve such a situation. Under certain circumstances, implicit permission may exist to look through an un-curtained window while standing on a front porch momentarily to see whether the resident is approaching the door, assuming no unreasonable means or devices are used.

Thanks to several Florida lawyers for sending on the opinion — including among them Howard Blumberg, who argued and won Jardines.

On Friday, Jonathan pointed out United States v. Blewett, the new Sixth Circuit decision on the 100-1 crack-cocaine sentencing disparity. Jonathan described the issue in that case as being whether the 2010 Fair Sentencing Act applied retroactively. But the most remarkable part of Blewett actually decides a different question that was neither briefed nor argued: Whether the 100-1 disparity in effect before 2010 was constitutional. And the majority’s argument for why the 100-1 disparity was unconstitutional strikes me as not just wrong but obviously so.

According to the majority opinion signed by Judges Merritt and joined by Judge Martin, the Equal Protection Clause requires judges to disregard bodies of law that have known racially discriminatory effects. That’s the case because applying law that has a known discriminatory impact would be an intentional act of discrimination by judges that the Equal Protection clause forbids. Here’s the key part of the opinion:

In view of the statistical facts and the widespread congressional consensus leading to the adoption of the Fair Sentencing Act’s remedial provisions [replacing the 100-1 ratio in 2010 with an 18-1 ratio], there can be no doubt that the old crack law was racially discriminatory in effect. As a matter of legal doctrine, there is no equal protection violation without discriminatory intent. See Washington v. Davis, 426 U.S. 229 (1976). When the old 100-to-1 crack cocaine statute was adopted, it presumably did not violate the Equal Protection Clause because there was no intent or design to discriminate on a racial basis. Its adoption was simply a mistake. Since 1986, however, we have gained knowledge of the old statute’s devastating effect on blacks. Congress itself acknowledged this problem by enacting the Fair Sentencing Act.

The Fair Sentencing Act was a step forward, but it did not finish the job. The racial discrimination continues by virtue of a web of statutes, sentencing guidelines, and court cases that maintain the harsh provisions for those defendants sentenced before the Fair Sentencing Act. If we continue now with a construction of the statute that perpetuates the discrimination, there is no longer any defense that the discrimination is unintentional. The discriminatory nature of the old sentencing regime is so obvious that it cannot seriously be argued that race does not play a role in the failure to retroactively apply the Fair Sentencing Act. A “disparate impact” case now becomes an intentional subjugation or discriminatory purpose case. Like slavery and Jim Crow laws, the intentional maintenance of discriminatory sentences is a denial of equal protection.

As I understand the reasoning, Judges Merritt and Martin work around the requirement of invidious purpose to discriminate by saying that judicial application of laws with known discriminatory effect forces the judges to have invidious purpose to discriminate when they apply the law. In other words, discriminatory effect plus awareness of it amounts to intentional discrimination in the act of applying the law. And the need to avoid discrimination not only trumps the law but also trumps binding precedents saying that the law is constitutional. The argument doesn’t work on its face, as a judge who applies binding law that may have a discriminatory effect does so not because she wants to achieve a discriminatory result but because that result is what the controlling law requires. But in any event, Judge Gilman’s dissent nicely points out the binding precedent to the contrary. The majority doesn’t even bother with much of a response to Judge Gilman’s dissent: See Footnote 6, which for the most part doesn’t even track forms of legal argument.

I agree that the 100-1 disparity was terrible policy. But the majority’s constitutional analysis strikes me as not just wrong but obviously so.

Over at MotherJones, Kevin Drum has an interesting post on the AP leak investigation: Here’s Why the Government Went Ballistic Over the AP Leak.

Also, over at Slate, Emily Bazelon and former VC blogger Eric Posner debate whether the subpoenas of the AP records were justified. Eric gets the better of the argument, I think, but it’s a helpful exchange either way.

I’ve blogged a few times about the substantial lower court division on whether the police can search a seized cell phone incident to arrest without a warrant under the Fourth Amendment. Today the First Circuit further deepened the split in United States v. Wurie by holding that a warrant is required.

With Wurie today and the Florida Supreme Court’s decision in Smallwood a few weeks ago, I would think that Supreme Court review of this legal question is highly likely sometime soon. (Notably, Deputy SG Michael Dreeben argued Wurie for DOJ.)

For my own views on the question, see my short essay Foreword: Accounting for Technological Change, 36 Harv. J. L. & Pub. Pol’y 403 (2013).

Most of the reporting on the DOJ investigation into the leak to the AP has said that the DOJ obtained two months of telephone records. The claim of a two-month period comes from the AP’s own reporting about what DOJ disclosed to the AP in its notice about the collection. The AP story began: “The Justice Department secretly obtained two months of telephone records of reporters and editors for The Associated Press in what the news cooperative’s top executive called a ‘massive and unprecedented intrusion’ into how news organizations gather the news.” (emphasis added)

According to a recent clarification by DOJ, however, that reporting was wrong. In a letter signed by Deputy AG James Cole, DOJ pointed out that although the subpoenas covered calls made during two calendar months, April and May 2012, the subpoenas only covered “a portion of that two-month period.” The DOJ letter doesn’t go into more detail than that, citing the need to keep the investigation confidential. But recall that the government asked the AP to delay running the story on May 2, 2012, and that the AP held the story for a few days until publishing it on May 7, 2012. Given that the action here occurred at the beginning of a month, the fact that the records were obtained relating to calls during two calendar months doesn’t indicate that the records were collected over a long period of time. It might have been a week or two, or even less; we just don’t know. We only know that the start date was some time in the month of April and the end date was some time in the month of May.

It will be interesting to see if press coverage will correct that error from the initial coverage of the story, or if media reports will continue to say that two months of records were collected.

News here.

Attorney General Eric Holder recused himself from approving the AP subpoenas in the CIA operation leak case because he himself had been interviewed in the investigation. But at a press conference today, Holder defended the DOJ investigation on the ground that the nature of the leak required “very aggressive action.” Here’s what Holder said:

This was a very serious — a very serious leak, a very, very serious leak. I’ve been a prosecutor since 1976 and I have to say that this is among — if not the most serious, it is within the top two or three most serious leaks that I’ve ever seen. It put the American people at risk. And that is not hyperbole. It put the American people at risk. And trying to determine who was responsible for that, I think, required very aggressive action.

Why might Holder think that? Some will conclude that Holder is lying, and that as a result he doesn’t really think it’s true. But let’s revisit the May 7, 2012, Associated Press story that started the investigation. Here’s the most relevant excerpt:

The CIA thwarted an ambitious plot by al-Qaeda’s affiliate in Yemen to destroy a U.S.-bound airliner using a bomb with a sophisticated new design around the one-year anniversary of the killing of Osama bin Laden, The Associated Press has learned.

The plot involved an upgrade of the underwear bomb that failed to detonate aboard a jetliner over Detroit on Christmas 2009. This new bomb was also designed to be used in a passenger’s underwear, but this time al-Qaeda developed a more refined detonation system, U.S. officials said.

The FBI is examining the latest bomb to see whether it could have passed through airport security and brought down an airplane, officials said. They said the device did not contain metal, meaning it probably could have passed through an airport metal detector. But it was not clear whether new body scanners used in many airports would have detected it.

The would-be suicide bomber, based in Yemen, had not yet picked a target or bought a plane ticket when the CIA stepped in and seized the bomb, officials said. . . .

The AP learned about the thwarted plot last week but agreed to White House and CIA requests not to publish it immediately because the sensitive intelligence operation was still under way.

The AP story tells us three important things: 1) The CIA knew about the details of the plot during its planning stages; 2) The CIA not only interrupted the plot but actually took the bomb into its possession and then passed it off to the FBI; and 3) the CIA’s work was occurring as recently as a few days before the AP story was published on 5/7/2012.

Based on those three facts, it seems pretty likely that the CIA had people “on the inside” of Al Qaeda’s Yemen affiliate who took possession of the bomb. As covert agents, they would have covered up their removal of the bomb by making it seem like something else had happened. That matters, I think, because it means that the Al Qaeda affiliates in Yemen have a ready way to find the CIA plant(s). Just trace back what happened to the bomb, and specifically find the folks who claim to have seen it last and who came up with some story about what happened to it around the beginning of May. Chances are, that would bring you to the guys working for the CIA. And that discovery probably means no more CIA plants working on the inside the next time, which may take away the person(s) who otherwise could disrupt future plots or tip off U.S. authorities to a future attack.

Of course, we can’t be sure that this is the case. But a close read of the original AP story does suggest reasons why this would have been a particularly dangerous leak. And, if so, that may explain why AG Holder thinks that this leak justifies a “very aggressive” investigation.

UPDATE: Helpful comments point me to this New York Times story from May 8, 2012, that added considerably more details after the initial AP story was out:

The suicide bomber dispatched by the Yemen branch of Al Qaeda last month to blow up a United States-bound airliner was actually an intelligence agent for Saudi Arabia who infiltrated the terrorist group and volunteered for the mission, American and foreign officials said Tuesday.

In an extraordinary intelligence coup, the double agent left Yemen last month, traveling by way of the United Arab Emirates, and delivered both the innovative bomb designed for his aviation attack and inside information on the group’s leaders, locations, methods and plans to the Central Intelligence Agency, Saudi intelligence and allied foreign intelligence agencies.

Officials said the agent, whose identity they would not disclose, works for the Saudi intelligence service, which has cooperated closely with the C.I.A. for several years against the terrorist group in Yemen. He operated in Yemen with the full knowledge of the C.I.A. but not under its direct supervision, the officials said.

After spending weeks at the center of Al Qaeda’s most dangerous affiliate, the intelligence agent provided critical information that permitted the C.I.A. to direct the drone strike on Sunday that killed Fahd Mohammed Ahmed al-Quso, the group’s external operations director and a suspect in the bombing of the U.S.S. Cole, an American destroyer, in Yemen in 2000.

He also handed over the bomb, designed by the group’s top explosives expert to be undetectable at airport security checks, to the F.B.I., which is analyzing its properties at its laboratory at Quantico, Va. The agent is now safe in Saudi Arabia, officials said. The bombing plot was kept secret for weeks by the C.I.A. and other agencies because they feared retaliation against the agent and his family — not, as some commentators have suggested, because the Obama administration wanted to schedule an announcement of the foiled plot, American officials said.

Officials said Tuesday night that the risk to the agent and his relatives had now been “mitigated,” evidently by moving both him and his family to safe locations.

But American intelligence officials were angry about the disclosure of the Qaeda plot, first reported Monday by The Associated Press, which had held the story for several days at the request of the C.I.A. They feared the leak would discourage foreign intelligence services from cooperating with the United States on risky missions in the future, said Representative Peter T. King, a New York Republican and chairman of the House Homeland Security Committee.

“We are talking about compromising methods and sources and causing our partners to be leery about working with us,” said Mr. King, who spoke with reporters about the plot on Monday night and Tuesday after he was briefed by counterterrorism officials. Mr. King, who called the bomb plot “one of the most tightly held operations I’ve seen in my years in the House,” said he was told that government officials planned to investigate the source of the original leak. The C.I.A. declined to comment.

The added details complicate the picture somewhat, I think. On one hand, they suggest that the agent’s work may have been done and the information likely was going to become public at least to some extent at some point. On the other hand, the details add the complication about discouraging foreign intel services from working with the U.S., which is a concern that those with intel experience will be able to assess more accurately than I can.

A lot of blogs are expressing outrage at the AP story reporting that the government collected logs of telephone numbers used by the AP. The AP’s story appears to have been written to get people upset. The first sentence of the AP report is not subtle: “The Justice Department secretly obtained two months of telephone records of reporters and editors for The Associated Press in what the news cooperative’s top executive called a ‘massive and unprecedented intrusion’ into how news organizations gather the news.” Yikes, a secret investigation! And a massive and unprecedented intrusion! It sounds like quite the scandal.

But a different picture emerges if you look past the AP’s spin. DOJ is investigating a leak of national security information to AP reporters that culminated in a May 7, 2012 story that disclosed details of a CIA operation in Yemen that stopped a terrorist plot in early 2012. The story had the byline of five AP reporters. DOJ opened an investigation into the leak to the AP, and pursuant to its published special rules on investigations involving the media investigations, issued subpoenas to find out what numbers were dialed from the relevant AP reporters during the months of April and May 2012. Presumably the thinking is that AP reporters called their sources, and the investigators want to trace the phone numbers to see who the sources might be. As far as I can tell, the information collected by the subpoena concerned the work and personal phone numbers of the five reporters and their editor, as well as the general AP office numbers where the reporters were located and for the main number for the AP in the House of Representatives press gallery. The AP knows about this because pursuant to DOJ’s policies found in 28 C.F.R. 50.10, the government was required to give the AP notice that the records were obtained. The AP received that notice in a letter on Friday, and then today (Monday) it released its AP story expressing AP’s outrage. That’s pretty much all we know so far.

Based on what we know so far, then, I don’t see much evidence of an abuse. Of course, I realize that some VC readers strongly believe that everything the government does is an abuse: All investigations are abuses unless there is proof beyond a reasonable doubt to the contrary. To not realize this is to be a pro-government lackey. Or even worse, Stewart Baker. But I would ask readers inclined to see this as an abuse to identify exactly what the government did wrong based on what we know so far. Was the DOJ wrong to investigate the case at all? If it was okay for them to investigate the case, was it wrong for them to try to find out who the AP reporters were calling? If it was okay for them to get records of who the AP reporters were calling, was it wrong for them to obtain the records from the personal and work phone numbers of all the reporters whose names were listed as being involved in the story and their editor? If it was okay for them to obtain the records of those phone lines, was the problem that the records covered two months — and if so, what was the proper length of time the records should have covered?

I get that many people will want to use this story as a generic “DOJ abuse” story and not look too closely at it. And I also understand that those who think leaks are good things will see investigations of leaks as inherently bad. But at least based on what we know so far, I don’t yet see a strong case that collecting these records was an abuse of the investigative process.

It’s here, via How Appealing. The embedding function only has the 1st 20 minutes, which you can watch below (and which will then ask you to click through to the PBS page for the rest).

Watch A More Perfect Union on PBS. See more from Constitution USA with Peter Sagal.

I haven’t blogged recently on judicial decisions considering the mosaic theory of the Fourth Amendment. As regular readers will recall, the “mosaic theory” is a term for the idea that long-term monitoring of a suspect can be a Fourth Amendment search even if short-term monitoring is not. Under this approach, which was suggested by the concurring opinions in United States v. Jones, surveillance and analysis of a suspect is outside the Fourth Amendment until it reaches some point when it has gone on for too long, has created a full picture of a person’s life (the mosaic), and therefore becomes a search that must be justified under the Fourth Amendment. I think the mosaic approach is a misstep for reasons I elaborated on in this article. And the handful of lower courts to have considered the theory since Jones mostly have not adopted it, either because they found it unpersuasive, because they distinguished Jones on the facts, or because they avoided the question under the good-faith exception to the exclusionary rule. See, e.g., United States v. Graham, 846 F.Supp.2d 384 (D.Md. 2012).

In the last week, two district courts have divided on the question: United States v. Rigmaiden (D. Ariz. May 8, 2013), and United States v. Powell, — F.Supp.2d –, 2013 WL 1876761 (E.D. Mich May 3, 2013) In this post, I want to discuss the two rulings, and then offer some critical commentary on Powell at the end.

First, today in United States v. Rigmaiden (D. Ariz. May 8, 2013), Judge Campbell held that the third-party doctrine applies to both cell-site information collected over 38 days and websurfing destination IP address information that collected 1.8 million destination IP addresses. Judge Campbell held that the mosaic theory did not apply when the government obtained historical cell-site records of the target’s location over a 38 day period, which it used to find the defendant:

Defendant argues that the government was able to use the cell-site information to effectively track his aircard from June 10 to July 18, 2008, a period of 38 days, and that this “prolonged surveillance” implicated his reasonable expectation of privacy. Doc. 824 at 215-17. Defendant relies on United States v. Maynard, 615 F.3d 544 (D.C. Cir. 2010), and United States v. Jones, 132 S.Ct. 945 (2012), but those decisions are inapposite. They do not address orders under the SCA, and the Supreme Court in Jones did not adopt the privacy theory advanced by Defendant.

In this case, a government agent, working in his office with the historical cell-site information and using mathematical and triangulation techniques, was able to calculate a general location for Defendant’s aircard during a 38-day period. The calculation narrowed the location of the aircard to one-quarter of a square mile.The Court cannot conclude that such use of cell-site information [over a 38 day period], obtained from a third party under the SCA, is tantamount to attaching a GPS device to a person’s vehicle. Calculations made from the historical cellsite information did not provide minute-by-minute intelligence on Defendant’s precise movements as did the GPS device in Maynard. The calculations merely identified a general area where the aircard was located – and stationary – for 38 days.

Also, in the case of the IP addresses monitored, the volume of surveillance didn’t matter:

Nor did the government violate Defendant’s Fourth Amendment rights by the volume of IP addresses it obtained. Defendant notes that the government obtained 1.8 million IP addresses from Verizon, and argues that the government should have tailored its request more narrowly and obtained only those IP address it had already connected to the tax-refund scheme. Because obtaining IP addresses is akin to obtaining telephone numbers, an act that does not implicate the Fourth Amendment at all, Forrester, 512 F.3d at 510–11, the Court cannot conclude that the government was required to narrowly tailor its request. Moreover,Forrester specifically held that there is “no difference of constitutional magnitude” between obtaining IP addresses and learning the total volume of communications with such addresses, holding that the government’s monitoring of “the total volume of data transmitted to and from [the defendant’s] account” did not violate the Fourth Amendment. Id. at 511. If the government can obtain destination IP addresses and the total volume of such communications without implicating the Fourth Amendment, the Court cannot conclude that the government somehow violated the Fourth Amendment when it obtained 1.8 million IP addresses accessed by Defendant’s computer.

On the other hand, last week Judge Murphy of the United States District Court for the Eastern District of Michigan handed down a decision adopting what amounts to a mosaic approach for cell-phone tracking in United States v. Powell, — F.Supp.2d –, 2013 WL 1876761 (E.D. Mich May 3, 2013). In Powell, eight defendants in a wide-ranging narcotics conspiracy jointly moved to suppress cell-site and GPS information obtained from six cell phones to monitor the location of the conspirators. The FBI had obtained a series of search warrants to obtain the records pursuant to Federal Rules of Criminal Procedure 41. Each of the warrants allowed the government to monitor an individual phone for at least 30 days, and the collection of all the warrants together allowed the agents to monitor the collection of cell phones for about seven months. The monitoring occurred over seven months in 2010, and one warrant in particular revealed very helpful information about the locations of at least one of the conspirators. Fourteen conspirators were charged in the case, and eight defendants moved together to suppress the location monitoring on the ground that the warrants were defective and thus that the fruits of the location monitoring had to be suppressed.

In Judge Murphy’s opinion, he ultimately rules that the good-faith exception to the exclusionary rule applies, so the motion to suppress is denied. But before he gets there, he adopts a few novel holdings about the Fourth Amendment.

First, Judge Murphy rules that a warrant was required to conduct as much monitoring as the government conducted. It’s true that in United States v. Skinner, the Sixth Circuit (binding on Judge Murphy) had said that cell-phone tracking of a suspect was outside the Fourth Amendment. But Skinner involved tracking for only a few days, while all of the monitoring collectively in this case involved tracking of the various suspects for seven months. This was a long enough period that it was highly likely that some of the tracking occurred when suspects were inside their homes: “Given the duration and intensity of the tracking, it cannot be reasonably argued that only public-thoroughfare data was collected. The government certainly collected cell-site data emanating from within the defendants’ homes or from another place in which the defendants had a legitimate expectation of privacy.” That matters because United States v. Karo had held that it was a search to monitor the location of a beeper device from inside a home. Under Karo, “[i]f at any point a tracked cell phone signaled that it was inside a private residence (or other location protected by the Fourth Amendment), the only other way for the government to have obtained that information would be by entry into the protected area, which the government could not do without a warrant.” This was bolstered by Kyllo v. United States, Judge Murphy reasoned, because the government couldn’t know if it was monitoring a suspect’s location inside the home:

In Kyllo, the Court rejected the argument that the government could restrict its thermal searches to non-“intimate” details based on the practical observation that “[n]o police officer would be able to know in advance whether his through-the-wall surveillance picks up ‘intimate’ details-and thus would be unable to know in advance whether it is Constitutional .” Id. at 38 (italics added). The same problem is posed by requests for prospective real-time cellsite location data. The March 11, 2010 warrant authorized the DEA to collect real-time cellsite location data for up to thirty days. Under virtually any circumstance, there was no way the DEA could know in advance whether or not the location data collected during that period would come from within a protected area.

Having distinguished Skinner, Judge Murphy then turned to the concurring opinions in Jones:

The Court’s Fourth Amendment concerns also overlap with those expressed by the D.C. Circuit in Maynard, and in the Jones concurrences. See Maynard, 615 F.3d at 562; Jones, 132 S.Ct. at 963–64 (Alito J., concurring in the judgment); Jones, 132 S.Ct. at 954–57 (Sotomayor, J., concurring). Generally speaking, those opinions express the view that warrantless long-term tracking by electronic means violates an individual’s reasonable expectation of privacy, not just because of the potential for tracking into protected areas, because the information obtained through such means is, in the aggregate, so comprehensive.FN7 See, e.g., Jones, 132 S.Ct. at 964 (“[S]ociety’s expectation has been that law enforcement agents and others would not—and indeed, in the main, simply could not—secretly monitor and catalogue every single movement of an individual’s car for a very long period.”) (Alito, J., concurring). The Jones majority found the Fourth Amendment implicated on narrower, property-based grounds, and declined to decide whether surveillance of Jones over thirty days by “electronic means, without an accompanying trespass, is an unconstitutional invasion of privacy.” Id. at 954 (“[T]he present case does not require us to answer that question.”). That basis of decision is not available in this case because no device was installed, and yet the privacy concerns implicated by the tracking seem just as profound.

For these reasons, the Court finds that when the government requests authorization to engage in long-term, real-time tracking of an individual’s movements via his or her cell phone, the situation reaches past the law set forth in Skinner, and Fourth Amendment concerns are implicated.

Judge Murphy reconciles his decision with Skinner as follows:

The cell-phone tracking standard is meant to be read in harmony with the Sixth Circuit’s holding in Skinner. Specifically, if the government seeks to track an individual for a short period of time only, with no foreseeable intrusion into protected areas, the probable-cause showing discussed here would not apply and Skinner would plainly govern. If, for example, the tracking was to be done for a limited purpose on public thoroughfares—like in Skinner, with foreknowledge a suspect was taking a two day cross-country trip—the specific Fourth Amendment concerns addressed here would not be raised. But if the tracking the government seeks to undertake is similar to the “intensive 28–day” monitoring the Sixth Circuit itself distinguished, then the more detailed showing required to meet the cell-phone tracking standard should be made

As I have written about the problems I see with the mosaic theory, I was very interested to know what kind of distinction Judge Murphy was drawing between short term and long term monitoring. Which version of the mosaic theory is Judge Murphy adopting? The DC Circuit’s, Alito’s, Sotomayor’s, or something else? How does it apply? How do you group over six phones and monitoring pursuant to a string of different warrants? Who has standing to challenge a mosaic search? Where are the lines? Judge Murphy responds in Footnote 7:

[The mosaic] view has its critics. See, e.g., Orin Kerr, The Mosaic Theory of the Fourth Amendment, 111 Mich. L.Rev. 311 (2012). The present Order does not adopt any of the “mosaic theory” approaches specifically, nor does it purport to address the questions raised in Kerr’s article regarding the implications of the approach. The Court simply notes that the constitutionality of long-term cell-phone tracking was left open in Jones, submits that the privacy issues presented by such tracking merit a doctrinal response, and finds that the facts presented here fall on the wrong side of the constitutional divide.

Judge Murphy next concludes that such monitoring not only requires a warrant, but that it requires a special kind of warrant:

Continue reading ‘District Judges Divide on Long-Term Cell Phone Tracking Under the Fourth Amendment’ »