In a message dated 98-05-18 17:41:56 EDT, Robin Hanson wrote:
> I agree that there are likely places where secrecy is clearly good, but this
> argument is far too sketchy to convince me that attorney-client priviledge
> is
> one of them. Yes it has some good features, but removing it also may give
> some good features. We have to weigh the two to say what is better on net.
> In fact, it's not even clear to me that the anglo adversarial legal system
> is better than the involved-judge approach popular in the rest of the
world.
> [snip]
> Have you had experience with a legal system where there was not an
attorney-
> client priviledge? If not, I don't see you can have much empirical support
> for one situation over the other.
The problem here is one I've mentioned before, the practical difficulty of
doing real quantitative analysis of legal systems. I'd be curious to hear any
proposal you might make for doing more than a critical logical comparison of
legal policies in this arena.
One major problem in attempting any empirical analysis of legal practices and
policies is the extent to which they are so widely connected to other social
phenomena. Isolating one element of legal policy or practice for analysis or
"experimentation" is extremely difficult. Your comment points at this: The
seemingly narrow policy of attorney-client confidentiality is so intimately
tied up with the extent to which a particular legal regime embodies the
adversary system that separating the effects of one from the many effects of
the other becomes at best a hellishly complex undertaking and, at worst a
practical impossibility.
But to return to your question, I know of no legal system that places
significantly less protection on the attorney-client privilege. However,
there is a recent example of two slightly different approaches to it that may
shed some light on the question. Over the last two decades there have been
two competing theories of the privilege in the case of corporate entities: The
"subject matter" and "control group" tests for the privilege. In the former,
communications between a corporation's lawyer and any employee regarding the
subject matter of the lawyer's representation were privileged. In the latter,
only communications with high-level executives were privileged. The
difference reflected two slightly different views of the legal personhood of
the corporation. Clearly, more communications were privileged under the
former theory than under the latter. Proponents of the former test (mainly
business interests) maintained that the narrower scope of the "control group"
theory curtailed a corporation's ability to obtain legal counsel, since
lawyers could not shield many aspects of their investigation of an issue from
subsequent disclosure. Some U.S. jurisdictions applied one test and others
applied the competing theory. Because of the federal system, enterprises that
carried on a nation-wide business were advised to be guided by the narrower
test, since they might be sued in a jurisdiction applying the "control group"
theory of the privilege.
Of course it is only my own personal anecdotal opinion, but I believe a real
effect could be discerned from the difference between the two theories.
Executives governed by the narrower test seemed to me to be less willing to
engage in broad-ranging reviews of questionable corporate policies, for fear
that the investigations of their lawyers would be discoverable in litigation.
A narrower scope of privilege seemed to discourage what has been called "self-
critical analysis". I wish I could offer this example with the empirical
rigor that would definitively answer the deeper questions about transparency
and privacy but, as I say, this simply isn't possible.
Greg Burch <Gburch1@aol.com>----<burchg@liddellsapp.com>
Attorney ::: Director, Extropy Institute ::: Wilderness Guide
http://users.aol.com/gburch1 -or- http://members.aol.com/gburch1
"Good ideas are not adopted automatically. They must
be driven into practice with courageous impatience."
-- Admiral Hyman G. Rickover
Received on Sat May 23 13:42:45 1998
This archive was generated by hypermail 2.1.8 : Tue Mar 07 2006 - 14:45:30 PST