The magnitude of the value and quantity of transfers of arms and military technology, the internationalization and commercialization of armaments industry (through vastly increased direct investments, through co-production, licensing and sub-contracting arrangements), the supply of training and technical services as adjuncts of arms supply, the blurring of the dividing line between military and civilian technology - all have made control measures infinitely more complicated and difficult. What compounds the difficulty is the fact that, since an overwhelmingly preponderant portion of arms transfers is made up of government-to-government transactions, control measures must emanate from supplying and/or receiving governments. But even if by some miracle these measures were forthcoming and proved effective, they will have touched only a small part of the problem of disarmament, because the share of international transfers of arms and military technology amounts to only 5–6% of the total world military expenditure. The other, far larger and more intractable, part relates to the staggering stockpiles of both conventional and nuclear weapon systems, almost wholly concentrated in the hands of the two superpowers. Both transfers and stockpiles of armaments are inextricably enmeshed in the existing international structure, epitomized in a dominance-dependency relationship. The paper examines the measures that the supplier nations and recipient nations can take unilaterally, bilaterally and multilaterally to curb arms transfers, and comes to the conclusion that unilateral initiatives, especially on the part of receiving nations, are more feasible. Not to take such initiatives on the ground that they cannot succeed unless taken in concert is only an excuse for doing nothing.