Further on the Foregone Conclusion Exception to the Fifth Amendment Act of Production Doctrine (6/7/16; 6/8/16)

Professor Oren Kerr has this excellent posting today:  The Fifth Amendment limits on forced decryption and applying the ‘foregone conclusion’ doctrine (The Volokh Conspiracy 6/7/16), here.  He discusses a case on that issue that is teed up in the Third Circuit and perhaps in an off the record amicus brief fashion, points the Third Circuit in the direction that it should hold.

So, what is the issue?.  The issue is whether, after the Government obtains a search warrant for the contents of an encrypted hard disk drive that it is unable to decrypt without a password, the Government can require the person with the password, under penalty of contempt, to provide the password over his Fifth Amendment assertion.  (I suppose the same issue might be presented for any computer related storage that the Government is unable to supply the password for access.)  In Fifth Amendment analysis merely producing documents under some compulsory process (usually a subpoena) requires a testimonial act merely in the act of production.  This has given rise to the Act of Production doctrine whereby a party under compulsory process can claim the Fifth Amendment privilege not as to the documents under compulsion (there is no Fifth Amendment privilege for documents) but as to the testimony inherent in the act of production.  But, perhaps inconsistently with the Fifth Amendment privilege, the Supreme Court has recognized a concept that, if Government can establish that the testimony inherent in the compulsion is a "foregone conclusion" then the testimony in the act of production is irrelevant and the Act of Production doctrine does not apply,  (I think that the way I phrased the "foregone conclusion" concept tilts in favor of the argument Professor Kerr makes, but bear with me here.)  The issue is important for a number of criminal cases.  In the case at issue, it is pornography, but it easily presents itself in tax cases where the documents evidencing the crime may be in encrypted storage.

Professor Kerr sets it up by pointing out the error, a prior Eleventh Circuit opinion, in his opinion, is incorrect.  That opinion is:  In re: Grand Jury Subpoena Duces Tecum Dated March 25, 2011, 670 F.3d 1335 (11th Cir. 2012), here,  (I blogged that case Fifth Amendment Act of Production Privilege and Encrypted Data Files (2/25/12), here; but do not recommend that readers go there because Professor Kerr has a good discussion.)  Professor Kerr picks on the Eleventh Circuit for misapplying the "foregone conclusion" exception to the application of the Fifth Amendment in an Act of Production setting.  Basically (and I urge everyone to read his discussion of the Eleventh Circuit case), he says that the Eleventh Circuit focused its foregone conclusion analysis on the existence of the documents rather than on the testimony being compelled.  Professor Kerr excepts this quote:

[U]nder the “foregone conclusion” doctrine, an act of production is not testimonial — even if the act conveys a fact regarding the existence or location, possession, or authenticity of the subpoenaed materials — if the Government can show with “reasonable particularity” that, at the time it sought to compel the act of production, it already knew of the materials, thereby making any testimonial aspect a “foregone conclusion.”

Professor Kerr's argument, directed not just at criticizing the Eleventh Circuit, but to the Third Circuit in the pending case is that the Eleventh Circuit's analysis misses the point.  It is not the underlying documents that is the focus of the foregone conclusion analysis but the actual testimony being compelled.  When that compulsion is as to a password to a hard drive rather than as to the contents of the hard drive, the only compulsion in issue is the password so that all the foregone conclusion analysis need show is that the compelled party knows the password.  That is all he is compelled to tell.  The compelled person, so the reasoning goes, is giving no testimony beyond the password and the requirement that the Government show that he knows the password with reasonable particularity should satisfy the Fifth Amendment concerns under traditional foregone conclusion analysis.

I think it helpful to quote Professor Kerr's key reasoning:

V. The 11th Circuit’s Error 

In my view, the 11th Circuit’s error was failing to see the big difference between two different kinds of cases. In prior cases like Fisher [Fisher v. United States, 425 U.S. 391 (1976). and Hubbell [United States v. Hubbell, 530 U.S. 27 (2000)], the government had issued subpoenas describing documents that the suspect had to turn over. The testimony implicit in compliance with such subpoenas is very different from the testimony implied by entering in a password. 

Here’s a hypothetical to explain what I mean. Imagine a court order says you must “hand over records of your tax fraud crimes from 2013 to the present.” Assume that if you comply, you will go home and come back with a box of records. There is a lot of testimony implicit in that act. Your implicit testimony includes the following: 

1) You believe that you committed tax fraud in that time window.
2) Each of the records in the box exist and were in your possession.
3) You believe that each of the records you are handing over show that you committed tax fraud in the relevant time window. 

Of these, the third is the most important. When the government issues an order requiring you to hand over a general category of records, you have to go back and decide which of your records fits within the general category and which does not. Handing over the records amounts to testimony that the records you are handing over are within that general class of records sought by the order. 

The 11th Circuit’s error was in applying language from cases compelling disclosure of broad classes of documents to the very different case of an order to enter a password to unlock a computer. The error is subtle but critical. It’s subtle because both cases involve steps that lead to the government accessing a lot of documents. If you look at the cases from 30,000 feet, they look kind of similar. But the error is critical because the testimonial aspects of production in the two cases are vastly different. 

In particular, the idea that the government must have some idea about what files exist and where they are located makes sense when the government has an order requiring the suspect to hand over a described set of files — but it makes no sense when the government is requiring the suspect to enter a password to access those files itself. When the government is relying on the target to go through his stuff and say which files are responsive to a request, the government is obtaining the suspect’s testimony about what files exist and which files are responsive. The suspect has a Fifth Amendment privilege unless that testimony about existence and location of the sought-after files is a foregone conclusion. 

When the court order only compels the suspect to enter a password, on the other hand, the government is not obtaining the suspect’s testimony about what documents exist, where they are, and whether they comply with the court order. The only implicit testimony is, “I know the password.” What files exist, where, and what they say is distinct from that. The government has to find that out on its own. The government has to search the computer and look for the records described in the warrant. It isn’t relying on the defendant’s testimony about what is on the computer because entering in the password does not imply any testimony about that.

I think that, perhaps, Professor Kerr's analysis is too facile.  I say perhaps because he is a lot smarter and more knowledgeable in this area than I am.  So rather than categorically attempting a rebuttal, I will just state my concern.  The password is meaningless unless it is the compelled party's computer or hard drive or at least one that he has sufficient access to to permit a testimonial like implication of some relationship to the contents of the computer or hard drive.  That implication under Act of Production analysis is testimonial it seems to me, and Professor Kerr's analysis simply jumps past that.

I would appreciate comments on the issue to help me think it through.

ADDENDUM 6/8/16 9:00 PM.  I asked for comments, but received none as of this time.  However, there have been some exceptionally good comments on Professor Kerr's blog entry linked above.  So, if you are really interested in this issue, go to that entry and spend some time reviewing both the blog entry and the comments.