In the Bear Sterns fraud case, as to defendant Matthew Tannin, the trial judge ruled today that the search warrant to Google for e-mail did not comply with the warrants clause of the Fourth Amendment because it was too general. United States v. Cioffi, 668 F. Supp. 2d 385 (E.D. N.Y. 2009), Doc. 225. An important case on computer searches and extremely interesting reading (I got it free from PACER late this afternoon):

Document searches pose unique Fourth Amendment concerns. As the Supreme Court noted more than 30 years ago:

[T]here are grave dangers inherent in executing a warrant authorizing a search and seizure of a person’s papers that are not necessarily present in executing a warrant to search for physical objects whose relevance is more easily ascertainable. In searches for papers, it is certain that some innocuous documents will be examined, at least cursorily, in order to determine whether they are, in fact, among those papers authorized to be seized. ... [R]esponsible officials, including judicial officials, must take care to assure that they are conducted in a manner that minimizes unwarranted intrusions upon privacy.

Andresen, 427 U.S. at 482 n.11. The dawn of the Information Age has only heightened those concerns. The risk of exposing intimate (and innocent) correspondence to prying eyes is magnified because “[c]omputers ... often contain significant intermingling of relevant documents with documents that the government has no probable cause to seize.” United States v. Vilar, 2007 WL 1075041, at *35 (S.D.N.Y. 2007) (citation and internal quotation marks omitted); see also United States v. Comprehensive Drug Testing, Inc., 579 F.3d 989, 1005 (9th Cir. 2009) (“Seizure of ... Google’s email servers to look for a few incriminating messages could jeopardize the privacy of millions.”). However, “it is precisely because computer files can by intermingled and encrypted that the computer is a useful criminal tool.” Vilar, 2007 WL 1075041, at *35.

Courts and commentators have wrestled with how best to balance privacy interests and legitimate law-enforcement concerns in the context of computer searches. One approach would require law-enforcement officials to specify a search protocol ex ante and to use, whenever possible, “key word searches ... to distinguish files that fall within the scope of a warrant from files that fall outside the scope of the warrant.” Raphael Winick, Searches and Seizures of Computers and Computer Data, 8 Harv. J.L. & Tech. 75, 108 (1994). Another would require the creation of “firewalls” to prevent investigators and prosecutors from obtaining the results of a computer search until documents within the scope of the warrant had been segregated by a third party. The Ninth Circuit recently endorsed variants of both procedures, among others, to minimize the intrusiveness of computer searches. See Comprehensive Drug Testing, 579 F.3d at 1006 (“Segregation and redaction must be either done by specialized personnel or an independent third party. ... The government’s search protocol must be designed to uncover only the information for which it has probable cause, and only that information may be examined by the case agents.”). As Judge Karas noted in Vilar, however, the majority of courts to have considered the question have not required the government to specify its search protocol in advance. See 2007 WL 1075041, at 37-38 (citing cases); see also Orin S. Kerr, Searches and Seizures in a Digital World, 119 Harv. L. Rev. 531, 572 (2005) (“[T]he [computer] forensics process is too contingent and unpredictable for judges to establish effective ex ante rules.”).

To date, the Second Circuit has not taken sides in the debate on the particularity required for computer searches. There is, however, one form of particularity whose absence the Second Circuit has unequivocally and unqualifiedly condemned: “[A]uthorization to search for ‘evidence of a crime,’ that is to say, any crime, is so broad as to constitute a general warrant.” George, 975 F.2d at 76. “[A] fortiori a warrant not limited in scope to any crime at all is ... unconstitutionally broad.” Id. at 77. George represents not simply a majority view, but the unanimous view of courts across the nation.

This case, however, is not about search terms or firewalls. It is, rather, about the fundamental and venerable prohibition on general warrants. Since “it is obvious that a general warrant authorizing the seizure of evidence without mentioning a particular crime or criminal activity to which the evidence must relate is void under the Fourth Amendment,” “no reasonably well-trained officer could believe otherwise.” George, 975 F.2d at 77. Before Groh, an executing officer might reasonably (and correctly) believe that his or her knowledge of the probable-cause affidavit would, under Bianco, cure a facially overbroad warrant. But Groh has been on the books since 2004. Indeed, the Supreme Court believed no reasonable officer could question its holding even then. Since Groh was a civil case, the Court was not called upon to directly apply Leon’s good-faith exception; however, it did address – and reject – the defendant’s claim of qualified immunity, the doctrine upon which Leon was based:

Given that the particularity requirement is set forth in the text of the Constitution, no reasonable officer could believe that a warrant that plainly did not comply with that requirement was valid. ... [A]s we observed in [Leon], ‘a warrant may be so facially deficient – i.e., in failing to particularize the place to be searched or the things to be seized – that the executing officers cannot reasonably presume it to be valid.’ This is such a case.

540 U.S. at 564-65 (quoting Leon, 468 U.S. at 923).

IV

The government has also invoked the “inevitable discovery” doctrine, under which unlawfully seized evidence is not suppressed if “the prosecution can establish by a preponderance of the evidence that the information ultimately or inevitably would have been discovered by lawful means.” Nix v. Williams, 467 U.S. 431, 444 (1984). The Second Circuit has held that “[t]he exception requires the district court to determine, viewing affairs as they existed at the instant before the unlawful search, what would have happened had the unlawful search never occurred.” United States v. Eng, 997 F.2d 987, 990 (2d Cir. 1993) (emphasis in original). The government does not -- and could not -- argue that the November 23rd Email would have been discovered by lawful means had the Warrant never been executed. Rather, it argues that the email could lawfully be obtained through another, more particularized warrant now that the Warrant has been invalidated. This distinction is dispositive because for purposes of the doctrine, timing is everything.

. . .

Even if I were to restrict the government to the facts set forth in the Affidavit, a more fundamental problem would remain. The “inevitable discovery” doctrine is essentially a rule of harmless error. See Nix, 467 U.S. at 443 n.3 (“The ultimate or inevitable discovery exception to the exclusionary rule is closely related in purpose to the harmless-error rule[.]”). No court has ever endorsed the view that it would allow the government to retroactively cure a Fourth Amendment violation, and it is easy to see why. The violation is not the discovery of incriminating evidence, but the invasion of the searchee’s privacy. Since such a violation is inherently irremediable, the exclusionary rule exists to deter violations in the first instance. See Elkins v. United States, 364 U.S. 206, 217 (1960) (“The [exclusionary] rule is calculated to prevent, not to repair. Its purpose is to deter--to compel respect for the constitutional guaranty in the only effectively available way--by removing the incentive to disregard it.”). It is one thing to say that there is no need to deter by suppressing evidence that would have been discovered regardless of unconstitutional conduct; it is quite another to allow the government a second chance to fix a problem that never should have arisen.

V

In concluding, I note that my observation in Cohan--that, in light of Groh, a facially invalid warrant cannot be salvaged unless a particularized affidavit is attached to the warrant and incorporated into the warrant by reference--was technically dicta since I found that the warrant there was sufficiently particular. See 628 F. Supp. 2d at 362. Nonetheless, I thought that this dicta was appropriate to alert the government that if it did not heed Groh’s holding in the future, it would risk the suppression of evidence that it wished to introduce at trial.

I have been the assigned judge on this case since its inception over a year ago. Cohan was issued this past June, two weeks before the Warrant was sought. The government was, therefore, on explicit notice that my understanding of the Supreme Court’s holding in Groh would undoubtedly oblige me strike down a facially invalid warrant in the absence of incorporation and attachment of the supporting affidavit.

CONCLUSION

Tannin’s motion to suppress is granted. The government is barred from introducing the November 23rd Email into evidence in its case-in-chief.

Newslinks: Judge Deals Huge Blow To Prosecution In Bear Stearns Hedge Fund Case on Business Insider; Tannin ‘Blow Up’ E-Mail Won’t Be Seen by Fraud Jury on Bloomberg.com; NYTimes.com: Judge in Bear Trial Deals Setback to Prosecutors.

The decision is criticized in Volokh Conspiracy as unnecessary because the government could have gotten the information by subpoena without risking an overbreadth challenge, or would it? What about standing?

Up until now, I would have thought that a person had no standing in his e-mail records held by the provider, here G-mail, under United States v. Miller (1976). I agree that one should have an expectation of privacy in his or her e-mail accounts.