The quest for regional security mechanisms has been part of the Israeli-Arab peace process since the 1991 Madrid Peace Conference. This process was born out of the positive experience of the Cold War and security forums from other regions. From the outset, such a translation of principles from the relations between superpower blocs to individual states was problematic. The lessons of other regions where such attempts had been made were also rarely relevant. In other areas of the world, regional security setups were constructed in a context of an existing peace, whereas in the Middle East they were discussed as part of a peace process within a state of war.
The discussion of regional security in the 1990s took place within a very different regional context than that which exists today. The region was still in the throes of the trauma of the Iraqi occupation of Kuwait and the debt to the United States for having saved the Gulf states from their own Arab brethren. The radical camp was in retreat: Syria had found itself without a superpower patron; the PLO was shamed by its support of Iraq; Saddam Hussein was restricted by sanctions, weapons inspections, no-fly zones in the north and the south and coalition military power. The general vector seemed to be toward stability.
There is a tendency to identify tensions primarily with the Israeli-Arab conflict. This, however, belies the experience of the region. The security events of the last two decades of the 20th century featured the eight-year-long Iraq-Iran war; the Iraqi invasion of Kuwait; periodic flare-ups of crisis in the Kurdish area (either between the Kurds and Iraq or due to Turkish interventions); continued Iraqi support of the Iranian opposition Mujahidin Khalq Organization (MKO) and exchanges of blows between the countries; border conflicts between Iran and its Gulf state neighbors; the Syrian occupation of Lebanon, and Syrian-Turkish tensions over PKK terrorism inter alia. Each of these tensions derived from different sources - none of them linked to the Israeli-Arab conflict. The focus of discussions on regional security on the Israeli-Arab angle could not provide appropriate answers to these other issues. The "War on Terrorism," which has overshadowed the peace process since September 2001, has given new meaning to the link between domestic security and regional security and has introduced new modalities of international involvement in the region.

The Madrid Process
The "multilateral track" of the Madrid conference included the Working Group on Arms Control and Regional Security (ACRS) which held a number of meetings (1993-1995) in which ideas were raised for regional security cooperation (related to such areas as rescue coordination, pre-notification of certain military activities, INCSEA - the Incident at Sea Agreement - and exchange of military information). Some agreements were reached but few were implemented and those that were stopped functioning (e.g. "hotlines" that began to operate in March 1995 based on OSCE [Organization for Security and Cooperation in Europe] end stations but ceased to exist in 1999).
As mentioned above, the early concept of ACRS was strongly affected by the paradigms of Cold War security agreements. However, the very concept of regional security suffered from adaptation pangs to the Middle East. The negotiations reflected fundamentally different approaches by the various parties to the process as to its very raison d'etre. In Israeli (and American) eyes, it was intended to co-opt the countries in the region that were not part of the bilateral negotiations, thus legitimizing relations between Israel and those countries, creating an atmosphere of peace and supporting the bilateral tracks. Practical Confidence and Security Building Measures (CSBM) were supposed to contribute to this end by enhancing the sense of security on all sides.
On the other hand, many of the Arab parties feared that normalization - or even a relaxation of tensions - between Israel and the Arab "periphery" would weaken the support of the Arab world toward the particular demands of the parties to the bilateral track. Egypt, which already had a peace treaty with Israel, feared the process might diminish its unique position - which provided a basis for its campaign against Israel's strategic edge vis-a-vis the Arab states and for promoting ideas for a Nuclear Free Zone (NFZ) in the Middle East.
The Madrid multilateral process petered off in 1995. It became clear that progress in this track was contingent on a comprehensive peace settlement between Israel and its Arab neighbors. Without progress in the bilateral tracks, it appeared the attempt to develop practical regional security and arms control agreements was doomed to failure. At best, it served as a confidence-building measure and a vehicle for facilitating Israeli contacts with that part of the Arab world that was not involved in the bilateral track; at worst, it became a forum for posing unrealistic demands for unilateral Israeli disarmament.

The Israeli Interests
Nevertheless, one can read between the lines of the positions put forward in these talks some of the guiding principles of an Israeli concept of regional security as it existed at the time. Some of these may be relevant in the future - if and when the circumstances for such a framework arise. First and foremost, the underlying principle of this concept, in Israeli eyes, would be that a clear distinction be made between security regimes for different stages of relations between the states of the region:
1) In which peace has not yet been achieved;
2) In which the countries have an established peace, but it has yet to be cemented;
3) In which peace has become the "natural state" of the region.
According to this principle, no country should be asked to forego an existing strategic advantage until such time as peace becomes cemented and there is no more threat of its being overturned. On the basis of this principle, the goals of regional security should be prioritized: First and foremost there should be mechanisms for prevention of military conflict as a result of mistaken perceptions of mutual intentions and channels for direct communication and consultation to avert crisis and to manage them when they occur; only later can the arms reduction and arms control issue be addressed. Any other method is to put the cart before the horse. The process should focus on developing specific confidence-building measures, such as pre-notification of large-scale military exercises, development of hotlines, crisis prevention mechanisms and verification procedures (the "operational basket").
As the process goes on, the goal of any regional security regime should be to ensure a "win-win" process. In other words, the outcome of the process at any given point should always enhance - and not diminish - the sense of security on all sides. Willingness by a party to give up existing unilateral security-enforcing capabilities must be met with a commensurate compensation, so as to preclude a compromise of net security. In order to meet this goal, a "net security index" of a country should take into account structural vulnerabilities. These vulnerabilities may be rooted in: size and concentration of population (and therefore its exposure to attack); the existence of threats, both within the region and outside it, to the security of the country; the goal of such enemies (limited territorial goals or total destruction); and the existence of coalitions or alliances between threatening countries (or entities) in the region or between them and outside powers.
But who should define the perceived threat and the structural vulnerability? In order to guarantee a true sense of security, every state in the region should define both the threat it considers relevant to its own security and its own relative strength, to which the arms control and the regional security process should provide adequate responses.
A salient question would be how threats such as terrorism could be factored into a regional security structure. A terrorist threat is not counterbalanced by a conventional or non-conventional strategic capability (as terrorists are not deterred by the existence of such a capability) and, therefore, such a threat is not relevant to the "net security index" of a country. The constraints on a country, which is a member of a regional security regime, should distinguish between use of violence in interstate conflicts and response to terrorist threats. On the other hand, the existence of state sponsors of terrorism and the willingness of states in the region to appease terrorists and to allow them an ideological and operational safe haven raises interstate conundrums: Responsibility of all parties to a regional security regime to maintain a monopoly over military power in their respective territories; accountability of states for acts of violence perpetrated or even planned from their territories; rights of "hot pursuit" and arrangements to preclude such actions from precipitating an interstate conflict inter alia. Some may argue that these problems would strengthen the claim that a terrorist threat toward a country entitles it to a "counterbalancing" strategic advantage.
Another equally problematic issue is the impact of a potentially nuclear Iran (with long-range delivery capabilities) on regional security. Iran poses a threat not only to Israel but also to the Gulf states, to Iraq, and past the region to its central Asian neighbors. On one hand, Iran's status as a regional power cannot be ignored; on the other hand, without a sea change in the attitude of Tehran toward Israel and the pro-Western Arab states, any regional security agreements would be an empty shell.
The relation between regional and global security regimes is also a problematic issue. After the blatant failure of the NPT in Iraq, Iran and North Korea, can it be a framework for a regional regime? Or should a regional regime rely on its own mechanisms? What is the role of the United States, with forces in Iraq, the Gulf and Afghanistan and deep involvement in the internal security of a number of the countries in the region? Regional security arrangements must also take into account current initiatives such as the "Greater Middle East," NATO involvement in the Middle East, evolving European defense concepts (in light of a future accession of Turkey to the European Union) and OSCE models. These trends raise the question of the "borders" of the "region" and the distinction between "local" and "international" forces.

A Concept for Phased Regional Security
In practical terms, building a framework for regional security in the Middle East is an incremental task. The stages of such a task could be:
1) A Declaration of Principles of all parties in the region barring all violence as a means of solving conflicts, while preserving a clear definition of the right for self-defense. The declaration should clearly include rejection of terrorism (with an unambiguous definition of terrorism) and violence against civilians, and define moral, financial, logistic and political support of such acts as violations of the principles of the regime. The declaration should be followed by an installation of means for direct communication between all parties of the region. Refusal of a party to join the DOP (e.g. Iran) should entail sanctions by other parties in the region.
2) A period of confidence building in which the means of stage one will be tested and perfected. During this period, there should be a "freezing" of the capabilities of all sides. No party or group of parties should be allowed to offset the security regime by acquisition of new military advantages, but no country will be called upon to forego existing capabilities as a condition of joining the security arrangement. This period should be concomitant with an ongoing bilateral and multilateral peace negotiation process.
3) Conclusion of bilateral peace treaties with all Israel's neighbors and solutions for other outstanding security tensions in the region (border conflicts, Syrian occupation of Lebanon, water, etc.). Benchmarks for the assessment of the durability of the peace should be set. One of these benchmarks will have to be the acceptance of peace by all entities in the region and the reduction of challenges to the peaceful relations. Such benchmarks may be: The continued ideological or religiously based rejection of peace by a state in the region or by political parties within states with which peace treaties have been signed; domestic stability of the states with which peace treaties have been signed, etc.
4) Negotiation of arms reduction along regional lines and gradual implementation of such agreements.
The feasibility of a regime for regional security is based on an assumption that all parties of the region share an interest in stability and conflict prevention and will, therefore, act in good faith to defuse any incipient crisis. As long as escalation and brinkmanship remain a political tool in the hands of some of the regional parties, it is doubtful that even the most modest security regime can be built, and if constructed - survive.

For details on the ill-fated ACRS negotiations, see: Michael D. Yafee, Promoting arms control and regional security, Disarmament Forum The Middle East 2001 no. 2, pp. 9 - 25; Bruce Jentleson, The Middle East Arms Control and Regional Security (ACRS) Talks: Progress, Problems and Prospects, The second ACRS plenary (Moscow, September 1992) was dedicated to the lessons that can be learned for the region from East-West detente. Discussions on Exchange of Military Information (EMI) and "Pre-notification of Certain Military Activities (PCMA) were also based on the Cold War experience.