The IAEA has the mandate to promote the peaceful uses of nuclear energy and to verify through safeguards that uses submitted to its safeguards are actually exclusively for peaceful purposes. Why has a safeguards function been laid on the organization? States transferring nuclear techniques, material or equipment that could be of some relevance in a weapons development normally make acceptance of safeguards a condition for the transfer. For the transfer of more important nuclear installations, most suppliers today will even make it a condition that the recipient country accepts safeguards over all its relevant nuclear activities - full-scope or comprehensive safeguards.
Safeguards may thus be seen as a verification imposed by suppliers. They are simply not willing to allow their exports to contribute to nuclear weapons development.
However, safeguards need not be viewed as, or indeed be, an imposition. The verification may be requested by the safeguarded party of imported or, indeed, indigenously developed nuclear techniques, equipment and material, because the State wishes to create confidence that the particular nuclear techniques, equipment or material serve only peaceful purposes. If viewed in this way by the State, the nature of safeguards is one of an international service. Verification by an outsider can provide greater credibility than verification by the party itself.
In a nuclear-weapon-free zone the parties agree between themselves at a minimum that they will not possess, control or acquire nuclear weapons. They may agree on a host of other things as well. For instance they might agree only to operate sensitive nuclear installations jointly, so that they all will have full access to and joint control of such installations. Such agreement may create further confidence that no party is moving ahead in a nuclear weapons acquisition or development. A concrete example is the agreement between the DPRK and the ROK not to operate enrichment or reprocessing plants.
The IAEA has no mandate to discuss these material provisions in a nuclear-weapon-free-zone agreement. But it does have considerable experience in operating verification of the respect for such agreements. It is against this background that the General Conference of the IAEA has asked the Director General several times to consult about the early application of full-scope safeguards to all nuclear activities in the Middle East and to prepare model agreements. It is evident that verification must always be adapted to the material provisions agreed upon. You must know what the parties have agreed to, in order to devise the verification of their respect for the agreed provisions. Verification should fit material rules as a glove fits the hand.
In the case of a nuclear-weapon-free zone in the Middle East we don't know yet what the hand will look like. The material conditions and provisions are yet to be negotiated between the parties. Hence we cannot devise a verification agreement. However, we can thoroughly acquaint ourselves with the various techniques and problems of verification. We can study components that are likely to be of use. In my view this is a very meaningful preparation for the days when the parties hopefully are designing the material conditions of a nuclear-weapon-free zone.
I have discussed these matters with governments in the Middle East during several trips and this workshop is also convoked with the same idea of acquainting the relevant experts with different modalities of applying safeguards. NPT-type safeguards - those defined in INFCIRC/153 - is the most commonly accepted and practised form of verification. Most States in the Middle East have accepted it and its acceptance by all would provide much valuable transparency. However, it is not the only type of comprehensive safeguards that could be of use in the Middle East. You should examine broadly what might be needed.
I should now like to walk with you gently and philosophically into the world of verification and study the landscape.
In order first to obtain some perspective on verification of non-proliferation it may be of interest to think of other areas in which verification occurs within nations or between them.
Within our respective States there are many activities and arrangements that are checked so that we may have confidence that they do not pose various unacceptable risks. Our elevators are usually inspected - perhaps once a year. Cars, too, are inspected regularly in many countries for their safety features, to reduce the number of accidents caused by technical faults. Restaurant kitchens, slaughter houses and many other places may be inspected to check that the level of hygiene is at an acceptable level. These are routine checks. They may occur whether or not there is suspicion of fault. The same applies to some checks against possible terrorist activities. Before we may enter a parliament, for instance, we may have to go through a metal detector and to put our briefcases under X-ray. And at airports we are familiar with the same procedure. In these controls we are all treated in the same way - whether we wear torn jeans or striped pants.
It is not certain that the checker and the enforcer is the same person or authority. If the lady who is looking at the X-ray picture of your briefcase detects a pistol, she would probably quickly call some police or guard. If the inspector who checks the elevator finds a deficiency, he/she may perhaps order the owner of the building to remedy the deficiency, but if the owner ignores the order enforcement authorities will be summoned.
Thus we find that in not so few activities is checking on respect for standards, rules, bans in the hands of special inspectors, but the enforcement - the sanctions - are placed in the hands of police or other armed units.
Safeguards inspectors, checking the absence of weapons-related nuclear activities, are not policemen who can enforce - seize a weapon or stop an activity. They are checkers, verifiers. They have to report and the power of enforcement - to the extent that it exists - lies elsewhere. I shall revert in a moment to that question.
Of course, we must also note that many verification activities in our national societies are directly placed in the hands of the enforcement authorities, e.g. the search for illegal possession of drugs or weapons. But here we can notice a difference to the airport controls or the checking of our cars. The police does not systematically check every house and person for drugs or weapons. They mostly check where they have some reason to think that there may be a violation. Indeed, they will normally have to show reason for a search of a house. They cannot enter at random.
If they find illegal weapons or drugs, they intervene themselves - they seize - since they are the enforcement agency.
It is interesting to keep these various models in mind as we discuss international verification and enforcement.
International verification is much rarer than verification in the national sphere. There is, for instance, no international agency verifying the safe construction and maintenance of aircraft or ships. This is done by national authorities and they may base themselves on national norms, which, however, may be modelled upon internationally agreed standards - perhaps even conventions. There is, as we well know, no international authority that verifies the safe construction and maintenance of nuclear power reactors - no international licensing authority for nuclear plants - but there are internationally agreed safety standards and as a matter of service - not mandatory checking - there are visits by international expert teams to nuclear installations, notably power reactors, to observe, advise and report on their safety standards.
As individuals and citizens we are jealous of our personal integrity and the inviolability of our homes. We don't want to confer a right on the police to search our houses and apartments for weapons or drugs or our own clothes and bodies - except for good reasons. The exceptions we make are few - at airports, on the road to check all motorists for possible excessive alcohol use.
States are even more jealous of their personal and bodily integrity - their sovereignty. We like to be masters in our own houses. We don't like to have agents of any foreign State in our countries checking on our conduct. But we have slowly come to realize that in some situations it may be in our own interest nevertheless at any rate to have such agents or agents of international organizations for some checking tasks.
This realization has come slowly. In the early part of this century so-called international sanitary councils were established in various cities where large numbers of pilgrims passed and where for that reason epidemic control had to be exercised. Despite the strong rationale for such institutions the concept of sovereignty long inhibited acceptance. (See Eagleton, International Government, 3rd ed. 1957, p. 175).
Many of the obligations which States accept among themselves do not call for any international or mutual verification. For instance, if States agree to combat the traffic in drugs, it is assumed that their respective national authorities will do the best they can and no international verification is needed of the efforts. The supervision of an agreed international regime may, in this manner, be as it were delegated to national authorities - a construction which certainly facilitates the enforcement, as enforcement measures are far less problematic at the national than at the international level.
In other areas international verification may have been found unnecessary because the violation of rules have been assumed to be easily visible by all. Thus the 1925 ban on the use of bacteriological and chemical weapons did not have any verification clause. Nevertheless, as we remember, in the war between Iraq and Iran the UN appointed an international team to verify the use of chemical weapons. The ban on the production and use of biological weapons has also not been complemented by any verification machinery. But the recent convention against chemical weapons has an elaborate verification system.
However, it is obvious that more and more international verification is seen as a common need and interest - internationally, regionally and bilaterally and the traditional and conceptual objections to it are weakening
Peace-keeping forces have often only the role to observe and report. And it is becoming increasingly common to call in international observers at national elections. Indeed, I think Costa Rica has made it a consistent practice. Here, obviously, national authorities which may have a strong interest in a given outcome may not have maximum credibility to verify fairness. International teams will be more credible. The respect for human rights standards is also increasingly subjected to international verification.
In hardly any area could the importance of full respect for a commitment and credible verification of this respect be greater than in disarmament and arms control and in this field no sector could be more vital than the nuclear one. Even the cheating on one single weapon could have the most dramatic consequences.
It is not surprising therefore that the nuclear field has seen the first large-scale arrangement for international verification. Indeed, even before the IAEA safeguards system was devised verification existed under bilateral agreements between the United States and States which purchased nuclear material from the US. Bilateral verification has not ceased with the establishment of an international system. Verification of nuclear disarmament between the US and Russia is so far exclusively bilateral. The Argentina-Brazilian agreement on a joint accounting and control system is evidently also a bilateral mechanism, but as you will hear, it is more joint than mutual inspection. The agreement between the Democratic People's Republic of Korea (DPRK) and the Republic of Korea (ROK), on the other hand, is more genuinely a bilateral arrangement for mutual inspection.
There are thus no conceptual or practical obstacles to bilateral, trilateral or regional verification. Euratom is, of course, a regional inspection regime. Bilateral inspection offers the parties the means of directly mutually convincing themselves of loyal implementation of non-proliferation pledges. However, reservations about outside inspection might be stronger as regards a bilaterally based system than about a multilateral one, where inspectors will represent a jointly managed international institution rather than a neighbouring State. There is nothing to suggest, lastly, that you could not construct a system that combined international and bilateral inspection.
The rationales for international inspection of non-proliferation pledges may be said to be two-fold: first, verification by outside inspection can achieve a level of credibility that can hardly be attained through measures undertaken by the inspected State itself. Secondly, an international system for global use may accumulate experience and achieve a level of scale which makes it cost effective.
Now let me look for a moment at the enforcement side of non-proliferation commitments. We must be aware that enforcement is a very different thing at the international level from what it is at the level of a State. I mentioned that at the national level many activities are checked by inspection and that enforcement is secured by authorities - police - which have a monopoly on the use of force. The fact that there is such a monopoly has the result that the actual use of force is mostly not needed. The threat is enough.
Between States the situation is very different. There is no police or joint army that can enforce an agreement. States which violate a commitment may, themselves, control vast armed forces. A nuclear-weapon State disregarding a commitment of some kind may not be easy to bring back to order. However, the risk of retaliation or collective reactions from States who wish to uphold the commitment may deter violations and may even bring a violator back to order.
We must recognize that the international community of States is a rudimentary, or a primitive society. In primitive societies - like Europe in the early Middle Ages - citizens and groups of citizens were armed. The family or clan protected its members, if need be, by threat or reliance on blood revenge on rival families and clans. The development of central authorities of power - kings - was slow. When it occurred it brought invariably a monopoly on the use of force. The citizens were disarmed, the central authority exercised the monopoly on the use of force and, at the same time, assumed the duty to protect the citizens against violence - by bringing offenders to court and to punishment.
We have not yet reached this stage in the international community. We still mostly live in clans - alliances - protecting their members by retaliation and we are certainly not disarmed! The risk of retaliation and the balance of power helps to maintain the peace - not yet an overwhelming international force under the control of an international authority. It will take a good deal of time before we get there.
This does not mean that the international community is in a state of complete lawlessness. Most international commitments - like agreements between citizens - are freely entered into and fulfilled because it is in the continuing interest of the party. Pressures by other States and the risk of strong reactions by other States or by the collectivity of the international community may also deter a State from ignoring an international commitment which no longer appears so desirable. But there is no parallel to the almost automatically available enforcement that you have in the state sphere.
An element helpful to achieve effective international reactions to violations of commitments is, however, authoritative impartial establishment that a violation has occurred - and not merely allegations by other interested parties. Here international observers and inspectors have an important role, just as witnesses and court proceedings are indispensible preconditions for the use of enforcement power within States.
I have already spent a good deal of my lecture time to explain the place of international inspection in a primitive international society. However, there is even more to say on this score. First, in a State evidence can be collected, if need be, with the help of the enforcement authorities. You can be subpoenaed to bring your car to inspection. And you know that. In the international community your acceptance of international inspection is based on your explicit consent. If you refuse or limit the right of access of inspectors, there are likely to be reactions, but they are by no means as automatic and predictable as in the domestic sphere. Yet such reactions are most important. Refused or limited access to safeguards inspectors would likely reduce confidence drastically.
Secondly, international verification must be organized on the basis of equality - like the airport terrorism control. An international inspectorate cannot - like a nation's police - concentrate resources on some parties and largely ignore others. It cannot show more confidence in some countries and less in others. It must impartially inspect all parties and report its findings. If reasons are found to believe that something should have been declared and was not, the inspectorate is expected to explore the matter. This is not a case of discrimination. All parties are subjected to this.
Now let me turn from these basic premises for inspection in the international community to some more specific points in the nuclear non-proliferation field.
What are the possibilities and limitations of inspection?
The objective for a State in joining a non-proliferation treaty and the concomitant verification is clearly to obtain confidence that other parties to the arrangement are respecting their non-proliferation pledge and not be surprised one day to find that a neighbour is a new nuclear-weapon State. As this is a matter of fundamental importance for a State's security, the confidence must be solidly based - or expressed differently the verification must give a very high degree of assurance. If it gives false assurance it may be worse than no verification, because it may lull a party into the belief that all is well, when in reality nuclear weapons are being made.
Perhaps no verification system can be devised that is one hundred percent effective. It is rare in this world that we reduce risks to zero. Seven kilograms of Pu do not take much space, nor does a nuclear warhead, I believe, need to be very big. If a State were to succeed in clandestinely buying one, and that is a big if, it might be able to hide it. Certainly it would be harder to hide enrichment and reprocessing plants. But - Iraq's enrichment installations had not been identified before the Gulf War.
As the reliability of non-proliferation commitments is of such crucial importance, especially in a region like the Middle East, we must analyse which factors and features are likely to result in high - if not 100% reliability.
We must remember, first of all, that we are dealing with sovereign States, controlling their own territories and controlling large resources. Although non-proliferation commitments can be entered into and verification of them can be made for all time, one cannot create absolute guarantees that a State government will not change its mind, and denounce an agreement on one pretext or another. As has often been said, the nuclear genie is out of the bottle. The principles and techniques of enrichment and reprocessing are no longer mysteries. The knowledge of advanced nuclear physics and engineering is also no longer a privilege of a few advanced States - as the case of Iraq showed.
Hence the ultimate guarantee of the continued respect for a non-proliferation agreement lies in the maintenance of a political micro- and macro-climate that gives no incentive to nuclear weapons acquisition. This means peace, extensive trade and interdependence, or integration. After two thousand years of fighting the European countries are now looking to integration as the means of development - and reliable power.
One can, of course, also seek outside security guarantees of various kinds. They may be more or less reliable. European NATO allies have looked to the US for a nuclear umbrella, but France preferred to have one of its own.
Verification cannot guard against the denunciation of non-proliferation commitments - as we see currently in the case of the DPRK. But verification should first, and normally, give confidence that no proliferation is underway and, secondly, give timely warning, if one were underway. Should such a warning signal be heard, there should be time for measures to prevent the proliferation to occur. Both an awareness that verification would discover proliferation - with a high degree of probability - and that the States concerned and the world would take strong measures to thwart proliferation might help to deter such proliferation.
What features are of key importance in verification to ensure a high probability of detection?
One might sum up the essential in one word: transparency. Under safeguards agreements concluded pursuant to the NPT States are obliged to report nuclear material and installations to the Agency. Recently our Board has strengthened the rules concerning the provision of design information and adopted recommendations concerning extensive reporting on the export and import of all nuclear material and on some equipment. There is no reason why States or groups of States cannot go further in the direction of transparency. The maximum nuclear transparency that can be attained between two or more States is presumably the joint operation of installations. Awareness of this was part of the background for the establishment of some European nuclear institutions, like Eurodiff. The joint Argentina-Brazil accountancy and control system does not constitute a joint operation of nuclear installations, but goes a very long way nevertheless in transparency by joint rather than mutual supervision.
Short of such joint operation or supervision, one can imagine arrangements under which States participating in a non-proliferation scheme gave very extensive reports of all their nuclear activities - in research as in industry, in agriculture as in medicine. Safeguards in pursuance of NPT falls far short of such reporting duties. Out of fear that extensive reporting duties and extensive inspection might impede nuclear development and risk revealing commercial or industrial secrets, the reporting and verification has focussed on the nuclear material - without which weapons cannot be made. With the benefit of many years of experience I think one dare now conclude that the fears were exaggerated and that wider reporting duties by the inspected and broader inspection rights could be instituted without any serious drawbacks but with some gains in transparency and assurance. In a nuclear-weapon-free zone the parties would obviously be free to agree on such measures. Inspectors need not be confined to "strategic points" but might be given the right to move freely around in nuclear installations and compounds.
The case of Iraq made all acutely aware of something that was known but was not much in discussion, namely that IAEA inspection was geared to installations and material declared by the inspected State and that the Agency had little or no knowledge of installations and material which the State had not declared.
Today I think there is increasing recognition that the Agency must sharpen its means of assessing whether there are any non-declared installations or there is any non-declared nuclear material.
The case of Iraq has given us a great deal of experience in verification. Much better analytical capacity now exists to identify even minute quantities of nuclear material and samples and smear tests are therefore of increasing value.
Samples of water and air from a large number of points in the territory and waters of a State can also provide most valuable information/assurance. This method has the advantage of not intruding into any installations. Satellite photography, when made available, can be of great help to identify possible undeclared nuclear facilities.
Now let me turn to "special inspections". The Board of Governors has affirmed in general terms in February 1992 and in specific terms relating to the DPRK in February 1993, the right of the Agency to perform special inspections - also to additional non-declared locations - under comprehensive safeguards agreements. The Secretariat has interpreted this right to exist when it has reason to believe that safeguards relevant information or locations exist which have not been declared and about which inadequate or no explanations are given. It has not asserted a right to inspect locations chosen randomly. However, it has taken the view - and been supported by the Board - that neither an assertion by the State that a location is safeguards irrelevant or that it is military, eliminates the Agency's right to special inspection.
In a verification system under a nuclear-weapon-free zone the parties are evidently free to construct inspection rights regarding non-declared locations as they like. Experience gained in the last years, in my view, points to rather wide rights. The Chemical Weapons Convention's arrangements concerning "challenge inspections" are probably the most thoroughly discussed ones. Especially the so-called "managed inspections"of military sites were the subject of much discussion. The Agency's provisions concerning "special inspections" give no specific guidance concerning military locations, but it is evident that some regard must be had to legitimate military-security interests. The Agency has also been ready to show such regard in the case of the DPRK.
There might be many different ways in which bilateral challenge or special inspections can be foreseen in a nuclear-weapon-free zone, whether it is based upon an international verification system or one between the parties alone or a combination of the two.
Let me also make one comment about the right of access, freedom of movement and communication of inspectors. You will find that the safeguards system of the IAEA under INFCIRC/153 provides less far-reaching rights in these regards than does the recent convention against chemical weapons. Under that convention inspectors will have the right to quick access, by their own transport means, if they wish, and they will be able to communicate freely with their home base. They shall be granted visas for multiple visits and designations are deemed automatically accepted, if not expressly rejected. These practical matters have importance for the effective functioning of an inspection system.
I shall conclude by some comments on a difficult but important subject: enrichment and reprocessing plants. Under NPT-type safeguards they are the subject of a considerable amount of inspection as they may contain direct use material. If they are permitted under a nuclear-weapon-free-zone arrangement it may be taken for granted that inspection of them would be extensive. It must be recognized, however, first that there is globally an excess enrichment capacity and that it will be hard, for a long time to come, to justify new enrichment plants - at least unconnected with sizeable nuclear power programmes. Secondly, the need for reprocessing, today, when the breeder reactors are far away, is also questioned by many, at any rate when unconnected with a significant nuclear power programme or part of a closed fuel cycle, waste disposal system. A nuclear-weapon-free zone in the Middle East would have to contain material provisions both as to the acceptability or not of enrichment and reprocessing and of the presence of any direct use material, whether highly enriched uranium or plutonium. Conceivably such capacity and such material might have to be the subject of special transitory provisions.