Legal compulsion and self-incriminating passphrase

[Michael Froomkin - U.Miami School of Law]

On Thu, 9 Jul 1998, Caspar Bowden wrote:

> Now I'm really confused. Nicholas Bohm thinks it might hold water, David
> Swarbrick thinks it doesn't depending on whether previous use of the
> passphrase is something that can be searched for (although no trace of it
> remains), and Michael Froomkin says no passphrase can be testimonial

Ok.  I qualified this a tiny bit below...

> (incidentally not sure if you reached the final para of my previous post. A
> passphrase surely can be testimonial in the extreme case that the crime is
> the sending of a piece of e-mail, and the passphrase is a - legally
> binding - digital signature attesting to authorship).

In my original post on this (or was it another?) thread I discussed the
authenticating/identification issue.  Here we agree.  Even the US Justice
Dept. has taken a careful approach to this issue and admitted that use
immiunity might be required. [Note: in this case the knowledge of the
passphrase is surely incriminating, but it doesn't prove that no one else
sent the message -- it just tends to prove it! -- since someone else could
have gotten the passphrase somehow. ]

Part of the confusion may be that I'm opining on US constitutional law (in
the general absence of guidance by the courts), while the English lawyers
here are thinking of PACE ... but not of the supra-national human rights
rules which, I had understood, were to be incorporated into English law in
the life of this government.  Please, someone, is this because they don't
apply, or something about English legal reflexes?

> 
> I still wonder about my original citation from Sergienko
> http://www.urich.edu/~jolt/v2i1/sergienko.html
> 
> "A cryptographic key need not have testimonial content. A key can be any
> word, phrase, or a series of randomly chosen digits. However, one can
> imagine a cryptographic key that has been given an incriminating,
> testimonial content by making it a word or phrase that confesses to a crime.

I suppose one could construct a weird hypo where the passphrase contained
some element of knowledge that only the guilty person could have known
("the knife is buried under the toolshead").  But that's different from a
*confession* ("I killed Jake").  Rightly or wrongly, I think the courts in
the US would usually rule the the confessional password wasn't testimonial
and would so instruct the jury.  For what its worth. 


A. Michael Froomkin        | +1 (305) 284-4285; +1 (305) 284-6506 (fax)
Professor of Law           | 
U. Miami School of Law     | froomkin@law.tm          http://www.law.tm 
P.O. Box 248087            |   
Coral Gables, FL 33124 USA | It's hot here!