DISCLAIMER

The following is a staff memorandum or other working document 
prepared for the members of the Advisory Committee on Human
Radiation Experiments.  It should not be construed as representing
the final conclusions of fact or interpretation of the issues. 
All staff memoranda are subject to revision based on further 
information and analysis.  For conclusions and recommendations of
the Advisory Committee, readers are advised to consult the Final
Report to be published in 1995.


MEMORANDUM

TO:	Members of the Advisory Committee on Human Radiation
	Experiments

FROM:	Advisory committee Staff

DATE:	March 19, 1995

RE:	Dealing with Classified Information


	Here's my first stab at suggesting how we should deal with
classified materials.  It's certainly not the final word, but we 
should resolve these issues soon, because it now appears that
we will not be able to get some of the information we need without
access to classified documents.  This poses some serious potential
problems:

	Openness.  We are supposed to be undertaking a public
	deliberation.  Our credibility could suffer if we 
	become linked with the system for keeping information
	secret.

	Logistics.  We have no facilities for maintaining
	classified documents on site.  We are not even
	supposed to discuss classofied information outside
	cleared facilities, although we may be able to do
	that downstairs.  We may have difficulties taking
	notes on classified information, and we will certainly
	not be able to bring those notes to our offices or
	write about them using equipment in our offices.

	Review and reporting.  If we learn things of a 
	classified nature that are important for telling our
	story, we will be faced with the choice of 1) submitting
	our report for declassification review or 2) preparing
	a classified annex.  Either would undermine our
	appearance of openness.

	Personal constraints.  Many of us do not want to be in
	the position of knowing things that we cannot talk about. 
	We want to limit our exposure to classified information
	to the minimum amount necessary for the purposes of the
	Committee.

	Here is my outline for a strategy to get what we need from
classified materials while minimizing these problems.  Here are the
steps we should take in each case

1.	Being as specific as possible about facts and documents we
	are interested in, request the declassification of 
	information we need, even if extensive deletions are
	required.  For the Green Run, for example, we could request
	declassification of the basic objective of the release and
	discussions in planning documents on minimizing human risks.

2.	Learn as much as possible about what information is likely
	to be classified and why; learn as much as possible about
	what of this information is already publicly available.  In
	the case of the Los Alamos RaLa releases, for example, the
	source term remains classified, possibly because it would
	reveal information on the amount of high explosive used in

				     1


	an implosion devise.  We should therefore learn how much is
	already known about the amount of high explosive used in
	early nuclear bombs.

3.	Make a targeted request to view documents or portions of
	documents containing information we believe to be relevant.

4.	When reviewing documents, focus on identifying document
	excerpts or specific facts which we want to have 
	declassified in order to tell the story we need to tell.
	One way to request fact declassification is to write a
	summary of the relevant information and request a 
	declassification review.

5.	If these steps fail to produce the declassifications we 
	desire, we must be prepared to state publicly that certain
	pertinent information remains classified in spite of our
	requests for declassification and our belief that
	declassification poses no threat to national security.
	We should be as specific as possible, e.g. "information
	related to radiation safety measures discussed in the
	planning of the Green Run."

6.	The difficult part would come if we uncover information 
	that we believe both a) is important to telling the story
	and b) should remain classified for national security
	reasons.  I do not believe that this will happen and
	John Till agrees, at least with respect to the Los Alamos
	RaLa releases.

	These are the suggested procedures in cases where we have 
identified a specific area where classification stands in the way of
our inquiry.  Some of the examples include the Green Run, the RaLa
implosion tests, and the medical records for the Marshallese, and
perhaps RW.

	In other cases, we have been asked to review large collections
of classified materials for relevant documents, which we will then
ask to have declassified.  The same general considerations apply, but
some will be more important and other less important.  The problem of
individuals being exposed to classified information unrelated to the 
task at hand would be greater, but the risk of a barrier to 
declassification would be less.  These are not collections that
have been identified because they are particularly sensitive; they
are merely too massive to declassify en masse in a timely fashion.

	This from Debby:

	Mark--the points you address at the end of your memo are of
most interest to me -- somebody who might have to wade through
non-relevant classified material to get to the docs that concern us.
On the bright side, the vast majority of what we are interested in
will not be classified even if it is located in a "classified"
collection.  I am moderately hopeful that we will be able to focus
our searches on the basis of folder listings and other inventories
so that we can avoid the problem of seeing things we don't want to.

	This from Don:

	One problem left out is recording Staff surveys of classified
materials where a large amount of material has been identified as
arguably responsive to ACHRE interest, and the suggestion is that
Staff review it for priority declassification, and/or passing over
as probably not useful.  I suggest some kind of log -- which could
be archived downstairs.  Are there other possibilities?


				  2

	This also from Don:

	I'm trying to arrange briefings, specifically targeted for
staff with clearances, on how classified research is dealth with
once it's performed.  Ultimately it may be useful to draw some
comparisons, for institutional case studies or otherwise, between
how open literature science is done and disseminated and how
classified science is done and disseminated.  (Of course, how
much we can go public with such analyses will be an interesting 
issue.)  As a first step, and also to enhance and accelerate
document retrieval, I've asked DOE and DOD for briefings on these
issues from OSTI, which handles these chores for DOE, and DTIC, 
which handles them for DOD.  Let me know if you want to go, and
topics to raise in advance.

	This from David:

	I'd like to know about the limits of merger between 
information about classified information and information about
unclassified information.  For example, logs or registers of
classified documents are typically unclassified documents
themselves, but there are (I think unwritten, but I may be
mistaken) strictures on incuding references to classified
documents in unclassified finding aids, such as library card
catalogs, so that such a register would not be included in
catalog.  This would also prevent listing the unclassified titles
of individual classified documents in such a catalog.  (a) is this
true?  and (b) if os, how do we handle our unclassified notes about
what we find and (c) if not, what is the guidance on this subject?
Debbie has already raised the issue of the classification of our
notes, another good topic. (I suspect our notes about information
objects will not be classified, but those about information may be.)


	We also need to know where and under what circumstances those 
of us with clearances may discuss the classified information contained
in classified documents (I assume there are no strictures on the
discussion of unclassified information about those documents).


				    3