November 15, 1971

Mr, A.R. Day

Arms Control and Disarmament Agency
Department of State Building
Washington, D.C. 20451

Dear Pete,

I have read over the CCD material (for which, thanks) and
eSpecially the Swedish remarks about the definition of toxins.
(I had made the identical remarks, even including alcohol, in
discussions with Jim Leonard at Geneva the summer before).

Dragging toxins in by the heels, as substances, and packaging
them with the microbes, has complicated all the definitions - as we
have discussed many times, Kow the Swedes are creeping down the
toxins alley to sneak in as much CW as possible. The mischief is
not in the aim but in the ambiguities it leaves behind.

I agree that toxins should be described in terms of what is
now conventionally classified that way. But we had better nail down
an explicit definition in accordance therewith.

A toxin is 1) a chemical substance
2) of polymeric, high molecular weight composition
3) known to science at least initially as a byproduct
of the growth of a microorganism
4) whose potential use, if any, in war depends on its
toxicity*to man, animals or crops,

 

This will relieve developments like that of penicillin (which was
a military monopoly at first in World War I) or pyrethrin insecticides
or algal foods or other microbial specialty products from being confused
with toxins,

If one admits compounds like tetrodotoxin, you are well on the way
to (unilateral) moratorium on the development of new CW agents, since
the toxicologists can hardly ignore what has been learned of the
molecular specificities of such compounds,

We may well wish to achieve such a moratorium but this is hardly
the route of choice. Furthermore, it will place investigators of such
compounds under a stigma of “working with BW", apart from the formal
prohibitions of the treaty.

W cleats Clervece os Wo wence | over

/ tphoyhad Case Mm flay 6% aw € se few,
foo ! ‘ &

JNM
Qe

 
Mr. A.R. Day -2- 11/15/71

There remains an insoluble problem of the synthetic analogues of
bacterial toxins which are bound to become mich better known in the
next decade. (I foresee at least a tenfold “improvement in toxicity
by the discovery of the “active fragments" of the toxin molecules).
There is no logical boundary between a simple derivative of a natural
toxin, and a completely novel chemical compound discovered by rationa
application of the principles learned from the natural toxins.
Finally, very active work on the synthesis of hormone~-like proteins
will undoubtedly uncover new agents of high potency with uses/abuses
analogous to the insecticides/nerve-gases. Not that insulin is
moderately toxic to normal people (at doses of some few mg.); other
hormones are even more active. A synthetic insulin would readily be
described as “toxin-like"™.

But we can do little better than assert the hope that we will
have perfected better controls on CW before the potential confusions
in this area become material, I do believe we should narrow the zone
of possible confusion by an explicit definition of toxin as indicated
above. There are many examples of a rapid loss of precision where
concepts as opprobrious as war gas or toxins are involved, We may
recall that the history of counting toxins @ BW was to preserve the
credibility of a moratorium on biological production and stockpiling.

Sincerely yours,

Joshua Lederberg
Professor of Genetics

P.S. I hope the above comments are useful to youjabut as I read them over,

I couldn't help but think, “God save us from the specialists!" These details
relate to side issues, and I am sure we have to work them over very carefully
to avoid mischievous confusions later. Also, they represent what I may
individually have to offer from my own expertise, But they are so unrelated
to the real problems, the underlying issues. My paper for the Stanford
journal which I sent you, also touches on the paradoxes of dealing with arms
control in technical detail, and the eventual posture of suffering evil

gladly that must emerge, I suppose, from compromising with the realities
of the world,

Where I feel cut off from reality is in a need for empirical justification
that a particular course of action -- unilateral restraints, or dotting all
the i's and crossing the t's in contractual agreements, for example <-- is
going to make any important difference in the long run. Perhaps insofar as
history all deals with non-reproducible situations, it would be self-illusory
to look for much scientific reassurance, which nags at me when I create
purely speculative anticipations of good or herm,

Anyhow I wanted to say that what any of us knows how to do very well,
is not necessarily the same thing as what is most needed to advance the

human condition; and it would be easy to fall into the trap of confusing
the two,