There has been a procedural revolution over the last twenty years in documentary requests in England, in both litigation and international commercial arbitration. As regards litigation before the English Courts, the civil procedural reforms introduced by Lord Woolf ten years ago substantially altered the old practice of 'automatic' discovery of documentation before trial, which consisted of lists of documents exchanged between the parties followed by successive inspections, additional lists and, almost invariably, further applications for 'specific' discovery.1 The key features of the reforms were the abolition of automatic discovery (as such), the significant restriction on the test for the 'relevance' of documentation (the abolition of the so-called 'Peruvian Guano' rule2) and an overarching test of 'proportionality'. There is no doubt that old-fashioned English discovery, with the aid of the newly-fashioned photocopying machine, had developed into an abusive, costly and delay-prone feature of English commercial litigation. Under the Woolf reforms, the new Part 31 of the Civil Procedure Rules (the 'CPR') was a huge innovation in the field of document production.3

The powers established by the CPR are exercised by the English Courts subject to an 'overriding objective', whereby the courts must deal with every case 'justly'. This term is defined expressly to include, so far as is practicable: '(a) ensuring that the parties are on an equal footing; (b) saving expense; (c) dealing with the case in ways which are proportionate - (i) to the amount of money involved; (ii) to the importance of the case; (iii) to the complexity of the issues; and (iv) to the financial position of each party; (d) ensuring that it is dealt with expeditiously and fairly; and (e) allotting to it an appropriate share of the court's resources, while taking into account the need to allot resources to other cases'. Whilst document production can still be onerous and costly in English commercial litigation, particularly with the advent of emails used in great quantities over long time-periods, it no longer represents the same impediment to a reasonably fair and cost-effective trial in the English courts as was frequently the case previously.4

The Woolf reforms played no significant part, however, in the reforms to the law and practice of international commercial arbitration in London as regards document production. These reforms had already begun independently, following the abolition of court-ordered discovery in arbitrations under amendments to section 12(6)(b) of the Arbitration Act 1950.5 This 'délégalisation'délégalisation' (in Professor [Page58:] Goldman's famous phrase6) permitted further practical reforms to arbitration procedures in London. These reforms were in large part developed pragmatically from international arbitration outside England and were crystallized in the English Arbitration Act 1996. It is not insignificant that the 1996 Act provided its own overriding objectives for arbitration, before the Woolf reforms: section 1 states: "(a) the object of arbitration is to obtain the fair resolution of disputes by an impartial tribunal without unnecessary delay or expense; (b) the parties should be free to agree how their disputes are resolved, subject only to such safeguards as are necessary in the public interest; (c) in matters governed by this Part the court should not intervene except as provided by this Part'. The two sets of statutory guidelines are very similar in substance, although not in form. It was indeed the new practices of international arbitration in London and elsewhere, being largely derived from practical common sense, that influenced the later Woolf reforms.

The procedural liberalism of the 1996 Act allows parties to agree upon any fair procedure they choose, consistent with the statutory objectives established for arbitration in sections 1 and 33 of the Act.7 As regards document production, subject to legal privilege, an arbitral tribunal enjoys a broad and unchartered discretion unless the parties agree otherwise; and sometimes parties will take the initiative in agreeing specific procedures for document production. It is, after all, the parties' arbitration. Nevertheless, arbitrators in London will usually resist any initiative by parties-usually by legal representatives more used to traditional domestic arbitration or English Court practice-to impose Lord Woolf's CPR on international arbitration proceedings. As already noted, English litigation and arbitration may have similar objectives; but the CPR are not designed for arbitrations; and the CPR are in many detailed respects unsuitable for international arbitrations, for they bring in their train a mass of English authorities on the meaning of the CPR's terms and their judicial implementation. The CPR are not a uniform procedural code for all English courts and tribunals.

It sometimes happens, albeit now rarely, that a distinguished but fledgling arbitrator, fresh from the appellate judicial bench with only pre-Woolf experience as a trial judge, can also default to the CPR because 'in choosing London, the parties must have chosen English court procedure'. Nothing could be more wrong: in choosing arbitration, the parties have chosen a procedure other than court procedures. As has been authoritatively said many times, an ICC arbitration in London is not an English arbitration under the ICC Rules, but an international arbitration under the ICC Rules in England.

Yet, more understandably, many English legal practitioners prefer a more explicit procedure for documentary requests than can be found in the broad discretions conferred by the 1996 Act and many institutional rules of arbitration. Section 34(2)(c) of the 1996 Act empowers an arbitral tribunal (unless the parties have agreed otherwise) to decide 'whether any and if so which documents or classes of documents should be disclosed between and produced by the parties and at what stage'. This short reference to document production is little more than one of several items on a checklist of procedural powers in default of the parties' agreement otherwise; and although it deliberately eschews the legal term 'discovery', it says nothing about the procedure under which such a power is exercisable by arbitrators. Indeed, none of the arbitration rules in common use, including the ICC Rules, provide any detailed guidance on the procedure for documentary requests; and it is notorious that Article 24(3) of the UNCITRAL Arbitration Rules provides significantly different procedures for document production to the other party (as distinct to submitting evidence to the tribunal) [Page59:] under the French and English official texts (a mistranslation which newly proposed reforms to these rules might soon correct).8

For these practitioners, it is therefore often necessary for London arbitrators to provide an additional template in order to ensure a comfortable degree of certainty and predictability. For this reason, Article 3 of the IBA Rules on the Taking of Evidence in International Commercial Arbitration (IBA Rules) has performed a useful role, not as a formal contractual addition to the parties' arbitration agreement but rather as a non-binding additional guide to the exercise of discretion by the arbitral tribunal, whether under the 1996 Act or institutional rules of arbitration. It has substantially replaced the ad hocad hoc forms of wording formerly used by many London arbitrators; and in practice it is not difficult to persuade parties to adopt the IBA Rules as a general guide, provided the initiative is made at an early stage of the arbitration, when agreeing the Terms of Reference with the tribunal or at the first procedural meeting. As with so many procedural issues, difficulties can be avoided if early expectations of future procedures are addressed and agreed with the parties before it becomes necessary to resort to such procedures.

Article 3 of the IBA Rules imposes its own discipline as to 'proportionality', by requiring any request to be limited to a document sufficient to identify it (NB in the singular) or 'a description in sufficient detail (including subject matter) of a narrow and specific requested category of documents that are reasonably believed to exist'. This wording excludes not only all 'fishing expeditions' (as did always the old English practice) but also places in the arbitral dustbin any request beginning with the old mantra 'any and all documents howsoever relating to . . .'. In addition, many London arbitrators have implemented a form of schedule to be prepared by the parties to allow easier determination of disputed applications for documentary disclosure under Article 3 of the IBA Rules.9 Each specific request is numbered in the first column for ease of reference; the second column describes the document or documentation requested by the applicant party, the third column describes its relevance by reference to the pleaded issues; the fourth column describes the basis on which possession of the document or documentation is attributed to the respondent party; the fifth column sets out the grounds upon which disclosure is disputed by the respondent party; and the last column is for the arbitrators to indicate their decision. The schedule assumes that the parties have earlier 'met-and-conferred' privately (without the arbitrators) in an attempt to narrow their procedural dispute or at least to understand better why such a dispute exists, all before submitting the remaining dispute over document production to the tribunal for its decision. The schedule can also cross-refer to fuller written submissions on specific issues arising from the applicant's request, which cannot be addressed adequately by an entry into one or more of its columns.

There are, however, two recognized deficiencies in the IBA Rules which London arbitrators seek to mitigate, with supplemental language in appropriate cases. First, the requirement under Article 3(c) of the IBA Rules that the requested documentation not be in the possession of the applicant party can work injustice where the issue on the merits is whether a particular document was or was not at a particular time in the possession of the other party. The applicant party may have acquired the document later; but that should not automatically debar its request.

Second, there is nothing in the IBA Rules which imposes any general obligation upon any party to disclose specific documentation which either supports its opponent's case or detracts from its own case. Such has long been the historical hallmark of English documentary discovery (in litigation and arbitration). In certain [Page60:] cases, it is the only way in which justice can be achieved between disputing parties; for example legal liability insurance where the insurer will usually not know enough about such documentation to frame an appropriately specific request under Article 3 of the IBA Rules. The result on the merits should not be determined by whether or not the insurer happens to know or guess correctly at the significance of a particular 'narrow and specific requested category of documents that are reasonably believed to exist'. Although international arbitration is an adversarial process, the tribunal should not be passive or indifferent to such potential procedural unfairness. Under the 1996 Act and many institutional rules, including the ICC Rules, London arbitrators bear a duty to re-balance procedural inequality between the parties. There are different ways of achieving this result. The simplest way is a procedural order requiring each party's legal representative to certify that to the best of that party's knowledge, after a reasonable search, no document remains undisclosed which either supports its opponent's case or detracts from its own case. Whilst this order can be made in many cases, it would not be appropriate for all arbitrations. (It can also raise, potentially, quite different issues regarding different ethical rules affecting legal representatives from different professional jurisdictions.)

On the other hand, the IBA Rules have elegantly helped to resolve the increasing difficulties facing London arbitrators in adjudicating upon claims for legal privilege and professional confidentiality. These claims can involve several different national legal and professional rules, e.g. the lex loci arbitri, the law applicable to the merits of the dispute, the law applying to the relationship between the service-provider and the client, or a combination of all or none of these. The first part of these difficulties is neatly resolved under Articles 3(6) and 9(2)(b) of the IBA Rules, in a manner which cuts through the Gordian knot of comparative laws and professional regulations:10 the tribunal may recognize a claim for non-disclosure based on 'legal impediment or privilege under legal or ethical rules determined by the [tribunal] to be applicable'. Nonetheless, it leaves unresolved the further difficulties in who determines whether a particular document falls one side or other of this rule. Without the prior written consent of the parties, an English arbitrator (as the final trier of fact) cannot safely inspect a document to determine whether or not it is protected from disclosure on the ground of legal privilege. That exercise, with the parties' consent, is best performed by a disclosure referee, making recommendations to the parties and the tribunal, subject (if necessary) to specific directions or guidelines from the tribunal. Alternatively, it can be done by the parties' respective legal representatives, assuming always the existence of reciprocal professional trust.

In summary, requests for documentary disclosure remain a major procedural event for most international arbitrations in London. As such, these requests can impose significant additional costs and time. Nonetheless, subject to procedural safeguards as regards fairness, cost-effectiveness and overall proportionality, documentary requests are widely seen as an indispensable tool in achieving justice in most arbitrations. The best practices in London arbitration, as regards such documentary requests in international cases, do not require the tribunal to default to the CPR or to other practices of the English courts. The 1996 Arbitration Act is largely permissive, allowing the parties and the tribunal to select the most suitable procedure for the particular dispute between the particular parties and their particular legal representatives. As noted already, the word 'discovery' does not appear in its legislative text for good reasons. The best procedures for document production will therefore differ from case to case, depending upon any of these variables. No international arbitration is the same; and painting by numbers is not [Page61:] appropriate for international arbitrators. In choosing a suitable procedure, the IBA Rules provide an important resource, subject to the overriding discretion of the tribunal and as modified for particular cases. The procedure for document production should fit into the overall procedural timetable with strict deadlines; and good use can be made of schedules to deal with disputed requests.11

The remaining practical problems lie not so much in the concept of documentary requests or the scope of arbitrators' powers, but rather in the timing of such requests. Almost inevitably, a party requests documents early in the arbitration proceedings, at a time when the arbitral tribunal does not or cannot always understand the significance of the documentation requested. When such documentation is refused by the tribunal, it is occasionally a cause for lament at the main hearing (or during subsequent arbitral deliberations) when its relevance is understood by the tribunal, belatedly and for the first time. Although it is wise to subject documentary requests to one or more cut-off dates before the main hearing, it is therefore wiser still to leave room for both the parties and the tribunal to resume the procedure if necessary at a later date, before the award. Lastly, if all this is now less than perfection, or even Panglossian perfection, it is certainly very far from the old imperfections of twenty or more years ago. More significantly, the appropriate powers are now in existence for arbitrators to use; and if they do not or cannot use them, it is not the fault of the 1996 Act.

1
Lord Woolf, Access to Justice: Interim Report (June 1995) and Access to Justice: Final Report (July 1996), leading (inter alia) to the Civil Procedure Act 1997. Both reports are available on the website of the Department for Constitutional Affairs, <www.dca.gov.uk/civil/reportfr.htm>.

2
This rule, called after the judgment of the Court of Appeal which expressed it, provided a very broad test of relevance 'as a document which, it is not unreasonable to suppose, may tend to advance the case of the party seeking discovery, or to damage the case of its adversary'; see Cie Financière et Commerciale du Pacifique v. The Peruvian Guano Co. (1882) 11 QBD 55.

3
The Civil Procedure Rules embody the rules and processes applicable to civil actions in courts in England and Wales. Originally introduced in 1998, the CPR are a moving target, subject to continuous drafting changes and practice directions. Their current text, with practice directions and other ancillary documentation is available on the Internet at: <www.dca.gov.uk/civil/procrules_fin/menus/rules.htm>.

4
English litigation does not ordinarily permit pre-trial depositions as part of discovery.

5
This was achieved by section 103 of the Courts and Legal Services Act 1990, following the advice of the Department Advisory Committee on Arbitration, then chaired by Lord Mustill.

6
See e.g. B. Goldman, 'Instance judiciaire et instance arbitrale internationale' in Etudes offertes à Pierre Bellet (Paris: Litec, 1991) 219 at 220ff.

7
Section 33 reads: '(1) The tribunal shall: (a) act fairly and impartially as between the parties, giving each party a reasonable opportunity of putting his case and dealing with that of his opponent, and (b) adopt procedures suitable to the circumstances of the particular case, avoiding unnecessary delay or expense, so as to provide a fair means for the resolution of the matters falling to be determined. (2) The tribunal shall comply with that general duty in conducting the arbitral proceedings, in its decisions on matters of procedure and evidence and in the exercise of all other powers conferred on it.'

8
The two versions of Article 24(3) of the UNCITRAL Arbitration Rules currently read as follows: 'At any time during the arbitral proceedings the arbitral tribunal may require the parties to produce documents, exhibits or other evidence within such a period of time as the tribunal shall determine.A tout moment de la procédure, le tribunal arbitral peut demander aux parties de produire des preuves complémentaires, en leur fixant un délai à cet effet.'

9
It has become called the 'Redfern Schedule' or the 'Rusty Park Schedule' after two distinguished international arbitrators.

10
The literature is vast, but see especially R.M. Mosk & T. Ginsburg, 'Evidentiary Privileges in International Arbitration' (2001) 50 I.C.L.Q. 345; J.H. Rubinstein & B.B. Guerrina, 'The Attorney-Client Privilege and International Arbitration' (2001) 18 J. Int. Arb. 587; F. von Schlabrendorff & A. Sheppard, 'Conflict of Legal Privileges in International Arbitration: An Attempt to Find a Holistic Solution' in G. Aksen, K.-.H Böckstiegel, M.J. Mustill, P.M. Patocchi & A.M. Whitesell (eds.), Global Reflections on International Law, Commerce and Dispute Resolution, Liber Amicorum in honour of Robert Briner (Paris: ICC Publishing, 2005) 743; and G. Burn & Z. Skelton, 'The Problem with Legal Privilege in International Arbitration' (2006) 72 Arbitration 124.

11
See p. 59 above.