Theoretically, three basic possibilities can be identified. The first one is that granting interim measures is exclusively allocated to the court. The court would provide the same interim protection to arbitration parties as it does to litigation parties. The opposite extreme is that one shifts interim measures of protection exclusively to the sphere of arbitration and only leaves the enforcement of the arbitrator=s orders to the courts. This would mirror the regular arbitration process. The arbitrator decides the dispute in his award and the courts enforce that award The courts do not interfere with the decision of the arbitrator. finally there is the option of free access to both the court and the arbitrator for interim relief The latter two possibilities call for a legal infrastructure. It must be ensured that the arbitrator can effectively grant interim measures of protection. This requires that his competence is spelled out and that an enforcement mechanism is adopted that suits the special needs of interim measures of protection. The measures must be enforceable very speedily and, where necessary, on an ex parte basis to preserve the element of surprise Today, a small number of countries follow the first possibility when they rule out any arbitrator=s competence to grant interim measures of protection. Examples are Italy and Greece(Sanders, 1996, p. 113 ). The majority follow the last approach as suggested by the Model Law when they provide for court access and the arbitrator=s competence to order interim measures of protection. But only spelling out the arbitrator=s power is not enough to provide a real alternative: a suitable enforcement mechanism must accompany it. The Model Law did not provide a guideline for an enforcement mechanism. Most countries do not provide a truly suitable mechanism Germany and Hong Kong have implemented such a mechanism. No country has yet adopted the second possibility. The English court-subsidiarity model comes close to it in defining precond itions for court access. The arbitrator is the course of first resort the court is the course of last resort with regard to interim measures It is a concept favoured and coined by german authors as the court-subsidiarity model (see Schlosser, 1989, pp. 306-307). Nevertheless, the new German law follows the free-choice approach of the Model Law. But a subsid arity model is reflected in the approach the ICC fosters. Article 23(2)of the ICC Rules(in force as from 1 January 1998)states that an arbitration party can only in >appropriate circumstances= apply to the courts for interim measures of protection once the file has been transmitted to the arbitral tribunal. With the >appropriate= test, the ICC Rules set up a hurdle that needs to be taken before a party can obtain court protection. This mechanism mirrors what can be regarded as a policy of court subsidiarity. However, it should be noted that article 23 (2)ICC Rules is wider than its predecessor, article 8(5). Article 8(5) referred to >exceptional circumstanc 3.1.2 Access to the courts If the lex arbitri does not clearly state that a party can have recourse to the courts for orders of interim protection, a dispute may arise as to whether seeking such recourse to the courts constitutes a waiver of the arbitration agreement but it can also have as a consequence that the courts refuse to aid arbitration parties, as has been the case in New York. In the McCreary decision, the Third Circuit held that the New york
Theoretically, three basic possibilities can be identified. The first one is that granting interim measures is exclusively allocated to the court. The court would provide the same interim protection to arbitration parties as it does to litigation parties. The opposite extreme is that one shifts interim measures of protection exclusively to the sphere of arbitration and only leaves the enforcement of the arbitrator=s orders to the courts. This would mirror the regular arbitration process. The arbitrator decides the dispute in his award and the courts enforce that award. The courts do not interfere with the decision of the arbitrator. Finally, there is the option of free access to both the court and the arbitrator for interim relief. The latter two possibilities call for a legal infrastructure. It must be ensured that the arbitrator can effectively grant interim measures of protection. This requires that his competence is spelled out and that an enforcement mechanism is adopted that suits the special needs of interim measures of protection. The measures must be enforceable very speedily and, where necessary, on an ex parte basis to preserve the element of surprise. Today, a small number of countries follow the first possibility when they rule out any arbitrator=s competence to grant interim measures of protection. Examples are Italy and Greece (Sanders, 1996, p. 113). The majority follow the last approach as suggested by the Model Law when they provide for court access and the arbitrator=s competence to order interim measures of protection. But only spelling out the arbitrator=s power is not enough to provide a real alternative: a suitable enforcement mechanism must accompany it. The Model Law did not provide a guideline for an enforcement mechanism. Most countries do not provide a truly suitable mechanism. Germany and Hong Kong have implemented such a mechanism. No country has yet adopted the second possibility. The English court-subsidiarity model comes close to it in defining preconditions for court access. The arbitrator is the course of first resort; the court is the course of last resort with regard to interim measures. It is a concept favoured and coined by German authors as the court-subsidiarity model (see Schlosser, 1989, pp. 306-307). Nevertheless, the new German law follows the free-choice approach of the Model Law. But a subsidiarity model is reflected in the approach the ICC fosters. Article 23 (2) of the ICC Rules (in force as from 1 January 1998) states that an arbitration party can only in >appropriate circumstances= apply to the courts for interim measures of protection once the file has been transmitted to the arbitral tribunal. With the >appropriate= test, the ICC Rules set up a hurdle that needs to be taken before a party can obtain court protection. This mechanism mirrors what can be regarded as a policy of court subsidiarity. However, it should be noted that article 23 (2) ICC Rules is wider than its predecessor, article 8 (5). Article 8 (5) referred to >exceptional circumstances=. 3.1.2 Access to the courts If the lex arbitri does not clearly state that a party can have recourse to the courts for orders of interim protection, a dispute may arise as to whether seeking such recourse to the courts constitutes a waiver of the arbitration agreement. But it can also have as a consequence that the courts refuse to aid arbitration parties, as has been the case in New York. In the McCreary decision, the Third Circuit held that the New York
Convention >forbids the courts of a contracting state from entertaining a suit which violates an agreement to arbitrate. Thus the contention that arbitration is merely another method of trial, to which state provisional remedies should equally apply, is unavailable [.]=. The reasoning of McCreary is, however, not followed by all American courts. In Carolina Power& Light Company v. Uranex it was held: >There is no indication in either the text or the apparent policies of the [New York Convention that resort to prejud gment attachment was to be precluded[. = As the Supreme Court had not yet an opportunity to rule on the question, American jurisprudence differs on the availability of interim measures of protection from the courts(see Born, 1994, pp. 772-773). A provision provid ing for both access to courts and the compatibility of court-ordered interim relief with arbitration is essential to do justice to the arbitration parties= cause and to prevent uncertainty If an arbitrator could equally efficiently order all measures of interim relief that the court provides, there would hardly arise any need and be any justification for court applications by arbitration parties. But as the arbitrator=s jurisdiction is limited, the court=s assistance is needed The first limitation of the arbitrator=s jurisdiction inherent in the arbitral process The arbitrator derives his power from the arbitratio agreement. The arbitration agreement is a contract that only binds the parties who entered into it. It has no effect on third parties. A Mareva injunction ordered by an arbitrator could thus only bind the arbitration parties, not their banks where the assets are held. Secondly, an arbitrator lacks any enforcement power. As a consequence, the Mareva injunction ordered by him would not be fortified with the threat of contempt of court, but would only have contractual effect between the parties. Taking these two limitations into account, it becomes evident how important access to the courts is fo certain kinds of interim relief in arbitration. Even if enforcement of arbitrator-granted interim relief is provided for, there will always be one step added to the procedure if enforcement is at stake. A court needs to step in to grant permission to enforce the arbitrator-granted order. This causes delay But it is not only the need for efficiency and speed that demands access to the courts for arbitration parties, it is also administrative peculiarities of arbitration. There is no permanent arbitral tribunal to which international commercial arbitrations are submitted. a tribunal has to be established for each requested arbitration This takes time. As long as the file has not been transferred to the arb itrator, there is no one available to grant interim measures of protection. Litigation practice shows that interim measures of protection are regularly in the highest demand before the case proceeds to trial. This observation is equally valid for arbitration. The greatest need for provisional remedies arises at a point when the tribunal has not yet been established. Arbitral institutions try to remedy this problem. An attempt undertaken by the International Chamber of Commerce in 1990 with its >Pre-Arbitral Referee Procedure= is, however, not successful (Schwartz, 1993, p. 64). Other institutions such as the World Intellectual Property Organization(WIPO)with its progressive draft >WIPO Emergency Relief Rules= look for alternatives. The London Court of International Arbitration (LCia)dropped its proposal in 1997 The above shows how essential court involvement is under certain circumstances but there are also voices against it. For instance, Boesch(1989, p. 8 )warns vehemently about the risks plication for interim relief to the courts can carry for arbitration
Convention >forbids the courts of a contracting state from entertaining a suit which violates an agreement to arbitrate. Thus the contention that arbitration is merely another method of trial, to which state provisional remedies should equally apply, is unavailable [. . .]=. The reasoning of McCreary is, however, not followed by all American courts. In Carolina Power & Light Company v. Uranex it was held: >There is no indication in either the text or the apparent policies of the [New York] Convention that resort to prejudgment attachment was to be precluded [. . .].= As the Supreme Court had not yet an opportunity to rule on the question, American jurisprudence differs on the availability of interim measures of protection from the courts (see Born, 1994, pp. 772-773). A provision providing for both access to courts and the compatibility of court-ordered interim relief with arbitration is essential to do justice to the arbitration parties= cause and to prevent uncertainty. If an arbitrator could equally efficiently order all measures of interim relief that the court provides, there would hardly arise any need and be any justification for court applications by arbitration parties. But as the arbitrator=s jurisdiction is limited, the court=s assistance is needed. The first limitation of the arbitrator=s jurisdiction is inherent in the arbitral process. The arbitrator derives his power from the arbitration agreement. The arbitration agreement is a contract that only binds the parties who entered into it. It has no effect on third parties. A Mareva injunction ordered by an arbitrator could thus only bind the arbitration parties, not their banks where the assets are held. Secondly, an arbitrator lacks any enforcement power. As a consequence, the Mareva injunction ordered by him would not be fortified with the threat of contempt of court, but would only have contractual effect between the parties. Taking these two limitations into account, it becomes evident how important access to the courts is for certain kinds of interim relief in arbitration. Even if enforcement of arbitrator-granted interim relief is provided for, there will always be one step added to the procedure if enforcement is at stake. A court needs to step in to grant permission to enforce the arbitrator-granted order. This causes delay. But it is not only the need for efficiency and speed that demands access to the courts for arbitration parties, it is also administrative peculiarities of arbitration. There is no permanent arbitral tribunal to which international commercial arbitrations are submitted. A tribunal has to be established for each requested arbitration. This takes time. As long as the file has not been transferred to the arbitrator, there is no one available to grant interim measures of protection. Litigation practice shows that interim measures of protection are regularly in the highest demand before the case proceeds to trial. This observation is equally valid for arbitration. The greatest need for provisional remedies arises at a point when the tribunal has not yet been established. Arbitral institutions try to remedy this problem. An attempt undertaken by the International Chamber of Commerce in 1990 with its >Pre-Arbitral Referee Procedure= is, however, not successful (Schwartz, 1993, p. 64). Other institutions such as the World Intellectual Property Organization (WIPO) with its progressive draft >WIPO Emergency Relief Rules= look for alternatives. The London Court of International Arbitration (LCIA) dropped its proposal in 1997. The above shows how essential court involvement is under certain circumstances. But there are also voices against it. For instance, Boesch (1989, p. 8) warns vehemently about the risks an application for interim relief to the courts can carry for arbitration
There are indeed risks involved, especially if the application to the court is regarded as a dilatory tactic, a way to get access to a biased forum or as a means to put pressure on the other party. Threatening to go to a public court with a dispute that should be solved confidentially can do this but such threats come from a party that does not ac in good faith. This kind of behaviour should not be taken as a reason to bar a good faith party from access to the most efficient remedy, especially if efficiency is decisive to secure assets without which the whole process of dispute settlement is rendered superfluous. Arbitration is there to serve the rights of the parties and is not an end in itself that needs to be protected at the expense of the bona fide party against attacks from the mala fide party. There are other means to prevent such attacks, for instance by stipulating that provisional remedies from courts in certain countries-the courts of which are deemed to be biased or bribable b are unavailable. Whether the courts will recognize such an opting-out of court protection is, however, doubtful English courts do(section 44, subsection 1 of the Arbitration Act 1996), German ones do not(LG Frankfurt). Further, it could be stipulated contractually that a party needs the arbitrator=s permission to apply to the courts. This can be regarded as a means to filter mala fide applications. But it could give rise to dispute concerning the stipulation and thereby hamper the whole process. The task to dismiss prima fa dilatory or tactical applications is best entrusted to experienced judges without imposing restraints on court access by the parties. The relationship between courts and arbitration should be one of mutual trust, respect and support. Modern trend is gradually moving in this direction. A centralization of arbitration matters at one court preferably a hierarchically higher one -would serve this goal 3.2 The arbitrator=s competence to order interim measures of protection The lex arbitri should also provide for the arbitrators competence to order interim measures of protection. It is not only necessary to ind icate that the state recognizes such a power, which is not yet the case in all countries, such as in Italy(article 818 Code of Civil Procedure), but it is also helpful to give the arbitrator explicit assurance of his power. This will especially be the case in institutional arbitrations because the contractual power conferred to them in the arbitration rules(e.g. article 23, paragraph I of the 1998 ICC Rules or rule 25(g) and rule 25 of the 1997 Arbitration Rules of the Singapore International Arbitration Centre-SIAC) is backed up in the lex arbitri of the seat of arbitration then there is no risk involved that the award will be set aside or will not be enforced on the ground that the >arbitral procedure[. was not in accordance with the law of the country where the arbitration took place=(article V paragraph 1(d)of the New York Convention for non-enforcement ground; compare article 34, paragraph 2(a)(iv) of the UNCITRAL Model Law for setting aside ground ) Arbitrators who are reluctant to grant provisional remed ies might find relief in such an assurance 3.3 The enforcement of arbitrator-granted remedies An arbitrator has no imperium, >the right [. to enforce the laws=(Black, 1891, p 594). Thus the state needs to step in with its enforcement machinery to lend the arbitrators= order the necessary authority to ensure compliance. Whether the enforcement machinery will be invoked is a different matter. Indeed Craig, Park and Paulsson point out that >parties do not ord inarily flout procedural orders made by
There are indeed risks involved, especially if the application to the court is regarded as a dilatory tactic, a way to get access to a biased forum or as a means to put pressure on the other party. Threatening to go to a public court with a dispute that should be solved confidentially can do this. But such threats come from a party that does not act in good faith. This kind of behaviour should not be taken as a reason to bar a goodfaith party from access to the most efficient remedy, especially if efficiency is decisive to secure assets without which the whole process of dispute settlement is rendered superfluous. Arbitration is there to serve the rights of the parties and is not an end in itself that needs to be protected at the expense of the bona fide party against attacks from the mala fide party. There are other means to prevent such attacks, for instance by stipulating that provisional remedies from courts in certain countries - the courts of which are deemed to be biased or bribable B are unavailable. Whether the courts will recognize such an opting-out of court protection is, however, doubtful: English courts do (section 44, subsection 1 of the Arbitration Act 1996), German ones do not (LG Frankfurt). Further, it could be stipulated contractually that a party needs the arbitrator=s permission to apply to the courts. This can be regarded as a means to filter mala fide applications. But it could give rise to dispute concerning the stipulation and thereby hamper the whole process. The task to dismiss prima facie dilatory or tactical applications is best entrusted to experienced judges without imposing restraints on court access by the parties. The relationship between courts and arbitration should be one of mutual trust, respect and support. Modern trend is gradually moving in this direction. A centralization of arbitration matters at one court - preferably a hierarchically higher one - would serve this goal. 3.2 The arbitrator=s competence to order interim measures of protection The lex arbitri should also provide for the arbitrator=s competence to order interim measures of protection. It is not only necessary to indicate that the state recognizes such a power, which is not yet the case in all countries, such as in Italy (article 818 Code of Civil Procedure), but it is also helpful to give the arbitrator explicit assurance of his power. This will especially be the case in institutional arbitrations because the contractual power conferred to them in the arbitration rules (e.g. article 23, paragraph 1 of the 1998 ICC Rules or rule 25 (g) and rule 25 (j) of the 1997 Arbitration Rules of the Singapore International Arbitration Centre - SIAC) is backed up in the lex arbitri of the seat of arbitration. Then there is no risk involved that the award will be set aside or will not be enforced on the ground that the >arbitral procedure [. . .] was not in accordance with the law of the country where the arbitration took place= (article V, paragraph 1 (d) of the New York Convention for non-enforcement ground; compare article 34, paragraph 2 (a) (iv) of the UNCITRAL Model Law for setting aside ground). Arbitrators who are reluctant to grant provisional remedies might find relief in such an assurance. 3.3 The enforcement of arbitrator-granted remedies An arbitrator has no imperium, >the right [. . .] to enforce the laws= (Black, 1891, p. 594). Thus the state needs to step in with its enforcement machinery to lend the arbitrators= order the necessary authority to ensure compliance. Whether the enforcement machinery will be invoked is a different matter. Indeed Craig, Park and Paulsson point out that >parties do not ordinarily flout procedural orders made by
arbitrators under contractually granted powers=(1990, pp 145-146). But as the inclusion of >ord inarily= implies, there are situations in which enforcement is necessary to ensure compliance. In order to be a real alternative to court-ordered interim relief. the orders of the arbitrators must be fortified with a real threat of enforcement. To achieve this end, state support is needed Cross-border enforcement of arbitrator-granted interim relief is an important aspect of the enforcement issue with regard to international commercial arbitration. In the absence of an international regime for cross-border enforcement, a unilateral approach can be seen as a way to solve the problem. The New york Convention is silent on the question of interim measures of protection and their cross-border enforcement(Van den Berg, 1981, p. 144). The Supreme Court of Queensland(see Sanders, 1996, p 115)rejected an attempt to enforce an interim award under the New York Convention in 1993. Berger(1993, p. 345 with further reference)outlines the convincing arguments against the enforcement of interim measures of protection under the New York Convention. In a cross-border context. arbitrator-granted interim relief could become even more efficient than court -ordered relief if at least a unilateral enforcement regime is in place. The importance of cross-border enforcement will be illustrated below with regard to interim measures preventing the dissipation of assets An arbitration will often take place in a third, >neutral= country. This is a place where substantive assets of the parties are most likely not held. If an arbitrator orders interim relief, it is(with the exception of German law)-at best-only enforceable at the seat of arbitration. The effect of the provisional remed ies is therefore rather limited. It will not reach assets in the home countries of the parties. If a court at the seat of arbitration grants an interim order, it can only prevent the dissipation of assets out of its jurisdiction In the absence of a bilateral or multilateral enforcement agreement covering court ordered or arbitrator-granted interim relief, the jurisdictional limitations force an arbitration party to apply to courts in the country or even countries where the assets are held. This is the reason why it is so important to provide for court access even if the seat of arbitration is outside the country where the order is sought. Article 1(2)of the model law caters for this need. If interim relief in arbitration would be cross- border enforceable, one order would be sufficient and no need would arise to apply to several courts for an order under probably different regimes of interim relief. Only enforcement proceedings in different countries would be needed It is unlikely that a country enforces interim relief granted by a court of a foreign country without an agreement providing for reciprocity. To expect a global multilateral enforcement agreement for court-ordered interim relief is illusionary. To establish an international regime for the cross-border enforcement of arbitrator granted interim relief will be difficult to achieve. a possibility would be to amend the New York Convention. In the meantime, unilateral solutions could provide for cross border effectiveness in the sense that countries render their mechanisms for enforcing arbitrator-granted interim relief not only applicable if they are chosen as arbitration venue but also in the case that the seat of arbitration is somewhere outside the country This would reflect the legislator=s commitment to the cause of arbitration as a whole 4. Country reports
arbitrators under contractually granted powers= (1990, pp. 145-146). But as the inclusion of >ordinarily= implies, there are situations in which enforcement is necessary to ensure compliance. In order to be a real alternative to court-ordered interim relief, the orders of the arbitrators must be fortified with a real threat of enforcement. To achieve this end, state support is needed. Cross-border enforcement of arbitrator-granted interim relief is an important aspect of the enforcement issue with regard to international commercial arbitration. In the absence of an international regime for cross-border enforcement, a unilateral approach can be seen as a way to solve the problem. The New York Convention is silent on the question of interim measures of protection and their cross-border enforcement (Van den Berg, 1981, p. 144). The Supreme Court of Queensland (see Sanders, 1996, p. 115) rejected an attempt to enforce an interim award under the New York Convention in 1993. Berger (1993, p. 345 with further reference) outlines the convincing arguments against the enforcement of interim measures of protection under the New York Convention. In a cross-border context, arbitrator-granted interim relief could become even more efficient than court-ordered relief if at least a unilateral enforcement regime is in place. The importance of cross-border enforcement will be illustrated below with regard to interim measures preventing the dissipation of assets. An arbitration will often take place in a third, >neutral= country. This is a place where substantive assets of the parties are most likely not held. If an arbitrator orders interim relief, it is (with the exception of German law) - at best - only enforceable at the seat of arbitration. The effect of the provisional remedies is therefore rather limited. It will not reach assets in the home countries of the parties. If a court at the seat of arbitration grants an interim order, it can only prevent the dissipation of assets out of its jurisdiction. In the absence of a bilateral or multilateral enforcement agreement covering courtordered or arbitrator-granted interim relief, the jurisdictional limitations force an arbitration party to apply to courts in the country or even countries where the assets are held. This is the reason why it is so important to provide for court access even if the seat of arbitration is outside the country where the order is sought. Article 1(2) of the Model Law caters for this need. If interim relief in arbitration would be crossborder enforceable, one order would be sufficient and no need would arise to apply to several courts for an order under probably different regimes of interim relief. Only enforcement proceedings in different countries would be needed. It is unlikely that a country enforces interim relief granted by a court of a foreign country without an agreement providing for reciprocity. To expect a global, multilateral enforcement agreement for court-ordered interim relief is illusionary. To establish an international regime for the cross-border enforcement of arbitratorgranted interim relief will be difficult to achieve. A possibility would be to amend the New York Convention. In the meantime, unilateral solutions could provide for crossborder effectiveness in the sense that countries render their mechanisms for enforcing arbitrator-granted interim relief not only applicable if they are chosen as arbitration venue but also in the case that the seat of arbitration is somewhere outside the country. This would reflect the legislator=s commitment to the cause of arbitration as a whole. 4. Country reports
4.1 England 4.1.Ⅰ Background England is an established international arbitration venue. for both institutional and ad hoc arbitrations. The London court of international arbitration is one of the busiest arbitration institutions in the world. It plays in the league of the market leader, the Paris-based International Chamber of Commerce. the American arbitration Association(AAA)and the traditional arbitration centres in Vienna, Stockholm and urIc Prior to the new single Arbitration Act 1996, the English legal framework fo international commercial arbitration comprised not only three different Acts (Arbitration Acts 1950, 1975 and 1979)but also a voluminous body of case law. The diversity of sources made access to the law equally difficult for non-specialists as foreigners. It therefore ill suited international commercial arbitration with its need for a transparent legal regime. But it was not only the law=s presentation which did not keep pace with modern trends, also substantive matters called for reform to defend England=s market share in the arbitration business The Model Law set new standards when it was ad opted by UNCITRAL and recommended by the General assembly of the United Nations in 1985. It gave impetus to a >Departmental Advisory Committee=(DAC), chaired by Mustill LJ(as he then was ). The task of the Committee was to advise on the adoption of the Model Law in England. The committee rejected adoption on several grounds, among them being the Committee=s finding that the Model Law differed >from established principles of English law=(see Merkin, 1996, p. 2, for quotation from the 1989 DAC Report). Even though the committee rejected adoption of the Model Law, it urged for new legislation and recommended features thereof. One of the features was the restatement of the principles of English arbitration law. But >consideration should also be given that the new statute] should, so far as possible, have the same structure and language as the Model Law, so as to enhance its accessibility to those who are familiar with the Model Law=(see DAC Report, 1996, No. 2 for quotation of paragraph 108(7)of the 1989 DAC Report). The first bill published in 1994 continued to reject adoption of the Model Law, as did the second bill prepared by the DAC in 1995-chaired by Saville L.J. -which followed the abandonment of the 1994 draft The 1995 draft led with mod ifications to the enactment of the arbitration Act 1996. The >DAC Report on the Arbitration Bill= of February 1996 contains a comprehensive commentary on the provisions and points out that >very close regard was paid to the Model Law, and [ that] it will be seen that both the structure and the content of[... the final draft owe much to this model=(DAC Report, 1996, No 4) The Arbitration Act 1996 came into force on 31 January 1997(The Arbitration Act 1996(Commencement No. 1)Order 1996) The new Arbitration Act 1996 was praised in the House of lords and in the House of Commons(see Miller, 1996, p. 252 and p 256)and has been welcomed >as it introduces a new arbitration regime which is clear and workable= (Slaughter and May
4.1 England 4.1.1 Background England is an established international arbitration venue, for both institutional and ad hoc arbitrations. The London Court of International Arbitration is one of the busiest arbitration institutions in the world. It plays in the league of the market leader, the Paris-based International Chamber of Commerce, the American Arbitration Association (AAA) and the traditional arbitration centres in Vienna, Stockholm and Zurich. Prior to the new single Arbitration Act 1996, the English legal framework for international commercial arbitration comprised not only three different Acts (Arbitration Acts 1950, 1975 and 1979) but also a voluminous body of case law. The diversity of sources made access to the law equally difficult for non-specialists as foreigners. It therefore ill suited international commercial arbitration with its need for a transparent legal regime. But it was not only the law=s presentation which did not keep pace with modern trends, also substantive matters called for reform to defend England=s market share in the arbitration business. The Model Law set new standards when it was adopted by UNCITRAL and recommended by the General Assembly of the United Nations in 1985. It gave impetus to a >Departmental Advisory Committee= (DAC), chaired by Mustill L.J. (as he then was). The task of the Committee was to advise on the adoption of the Model Law in England. The committee rejected adoption on several grounds, among them being the Committee=s finding that the Model Law differed >from established principles of English law= (see Merkin, 1996, p. 2, for quotation from the 1989 DAC Report). Even though the committee rejected adoption of the Model Law, it urged for new legislation and recommended features thereof. One of the features was the restatement of the principles of English arbitration law. But >consideration should also be given [that the new statute] should, so far as possible, have the same structure and language as the Model Law, so as to enhance its accessibility to those who are familiar with the Model Law= (see DAC Report, 1996, No. 2 for quotation of paragraph 108 (7) of the 1989 DAC Report). The first bill published in 1994 continued to reject adoption of the Model Law, as did the second bill prepared by the DAC in 1995 - chaired by Saville L.J. - which followed the abandonment of the 1994 draft. The 1995 draft led, with modifications, to the enactment of the Arbitration Act 1996. The >DAC Report on the Arbitration Bill= of February 1996 contains a comprehensive commentary on the provisions and points out that >very close regard was paid to the Model Law, and [that] it will be seen that both the structure and the content of [. . .] the final draft owe much to this model= (DAC Report, 1996, No. 4). The Arbitration Act 1996 came into force on 31 January 1997 (The Arbitration Act 1996 (Commencement No. 1) Order 1996). The new Arbitration Act 1996 was praised in the House of Lords and in the House of Commons (see Miller, 1996, p. 252 and p. 256) and has been welcomed >as it introduces a new arbitration regime which is clear and workable= (Slaughter and May