~10~- \
revenue for the execution of a treaty, as Venezuela did to the Europcan powers
in 1902.

An interesting form of indirect safcguard is the general exchange of

military and naval attachés as a method of removing fears of unfriendly war

preparations in derogation of treaties of friendship.
The only one of the familiar safeguards which seems to offer any promise in
the international control of atomic energy is that of inspection.

If it were

possible to back up a limitation agreement with a system of disinterested inspection operating on a world-wide basis, the parties to the agreement would have
a way of continuouslyreassuring themselves that no preparations were underway
Within any state to evade the agreement.

But if this were to be the only safein fact as well as in appearance;

guard, it would have to be practically infallible/ otherwise the states living
up to the treaty would be lulled into a sense of false security and the door

opened to easy violation by a potential sroubie“Ganer Furthermore , unless every
state confidently believed in the infallabitity oes

inspection system,

individual nations which had grown suspicious might feel impelled to resort to
secret production of atomic weapons as a precautionary measure.
This type of safeguard has a precedent in the inspection system developed
in connection with the international control of narcotics.3

While this scheme

resulted in bringing to light a number of violations, it was by no means infallible, and was scarcely cffective at all against violations condoned by
national authorities,

.

|

Some scientists impressed by the great technical difficulties in the way of
a really effective inspection system have taken a very gloomy view of the
possibilities of such a safeguard.

Others who are more impressed by the problems

of concealing the large-scale operations involved in the production of atomic
weapons are far less pessimistic.

The information so far made available is not

sufficient to enable the layman to reach a satisfactory conclusion on the

3. This is discussed later in Chapter V, pp, 152-153.

4

Select target paragraph3