From: "E. Allen Smith" <EALLENSMITH@ocelot.Rutgers.EDU>
Date: Thu, 20 Feb 1997 14:26:46 -0800 (PST)
To: cypherpunks@toad.com
Subject: Re: Constitution and a Right to Privacy
Message-ID: <199702202226.OAA23758@toad.com>
MIME-Version: 1.0
Content-Type: text/plain


From: Dale Thorn <dthorn@gte.net>

>Greg Broiles wrote:
>> At 11:56 PM 2/19/97 -0800, Dale Thorn wrote:
>> >It's good if you don't read the L.A. Times.  One of their lead editor-
>> >ializers (whores), a professor at USC law school named Erwin Chemerinsky,
>> >writes in relation to this subject "The federal government is an inde-
>> >pendent sovereign that cannot have its powers diminished by a state
>> >government's actions." (exact quote, 2/7/97).  Either that's a load of
>> >BS doubletalk, or it's one of the more fascistic commentaries from the
>> >Times, which is usually bad enough.

>> Hey, don't forget to shoot the messenger. Chemerinsky's statement is a
>> concise summary of at least the last 70 years of double jeopardy
>> jurisprudence. The only thing that's unusual about the use of the "dual
>> sovereign" doctrine against the cops who beat King was that it's usually
>> used against ordinary citizens, not cops. Do you suppose it's possible

>I understood clearly the (supposed) intent of the feds in retrying
>the Whites in the South who were beating up on Blacks and getting
>off with White juries - I just believe they would have served the
>people better by declaring mistrials or something instead of using
>the "dual sovereignty" BS, since a study of the Constitution and
>its preparatory papers shows the fathers clearly would have balked
>at this.

	I had understood the basic reasons for the double jeopardy
limit to be two: A. to prevent government from tormenting innocent
individuals with repeated trials; and B. to prevent convictions of
innocents by simple repeated trials until a jury made a mistake. The
latter can be analogized in statistics to the consideration that a
test with a p value of .05 will turn out false positive results 1 time in
20 by the nature of it; thus, if you look at two different studies/trials
with an independent (an oversimplifying assumption for trials, admittedly)
5% chance of an incorrect conviction, the chance is 1-((19/20)(19/20))
or 9.75% for _either_ of them turning out a false positive instead of a
5% chance. In a system partially based on the principle that it's better
to let the guilty go free than to punish the innocent, this result is
not acceptable. While the founding fathers probably didn't realize the
mathematics behind the second, I suspect that they had some intuition
of its nature.
	I can see an argument, however, for using it in the case of
state employees, particularly police - the state judicial/prosecutorial
system can hardly be said not to be biased in favor of them. A better
solution, however, would be to simply do the trial in a federal court
to begin with. (Having trials of federal employees in the courts of the
state of their alleged victims would be a correspondingly good idea,
although multiple possible states for this could be a problem.)
	I'd appreciate comments from persons with more legal knowledge
than I have; I am simply going by logic here.
	-Allen