Evidence in International Arbitration: Criteria for Admission and Evaluation
Arbitration rules give broad authority to arbitrators regarding the consideration of evidence. They usually do not set any formal procedure of admission and evaluation of evidence and say little if anything about the criteria for such admission and evaluation. The UNCITRAL Arbitration Rules (as revised in 2010) art.27(4) provide that once a party offers evidence to prove the facts it relies on, the tribunal is required to “determine the admissibility, relevance, materiality, and weight of the evidence offered”.
The American Arbitration Association (AAA) International Arbitration Rules art.20(6) provide that “[the] tribunal shall determine the admissibility, relevance, materiality and weight of the evidence offered by any party”.
Likewise, the London Court of International Arbitration (LCIA) Arbitration Rules art.22.1(f) empower the tribunal
“to decide whether or not to apply any strict rules of evidence (or any other rules) as to the admissibility, relevance or weight of any material tendered by a party on any matter of fact or expert opinion”.
The vast majority of national arbitration laws also recognise the discretion of arbitrators in these issues. Thus, there is wide recognition of the arbitral tribunals’ discretion to admit any relevant evidence they deem to have probative value, as well as of their power to reject evidence that is irrelevant or repetitious or unsuitable to prove the facts it purports to prove.
On the other hand, in international arbitration the parties are free to submit any evidence in order to prove the facts necessary to establish their cases. However the growing importance of the time and cost efficiency of arbitration makes arbitrators believe that they should not consider everything that each party has to submit or request.
Though arbitral tribunals both ad hoc and institutional regularly consider evidentiary questions, especially those concerning relevance, reliability and the distinction between admissibility (in the broad sense) and evidentiary weight, a degree of confusion in the application of those criteria still exists. This article aims to assist in distinguishing what, from the assessment of evidential value or weight, is required for admissibility, an issue to be decided at the end of the proceedings in light of all of the evidence.
A party seeking to challenge an award because a tribunal refused to admit evidence, and by doing so negatively affected the party’s right to present the case, may succeed. However, it is far more difficult to convince a court that an arbitration tribunal erred when it admitted the evidence but failed to properly evaluate its significance. In countries which based their arbitration laws on the UNCITRAL Model Law or other arbitration-friendly legislation and court practice it is hardly possible to challenge an award on the grounds that the arbitrator improperly evaluated the evidence. This is resulting in arbitral tribunals often agreeing to admit most of the evidence submitted by the parties, but giving serious consideration to the weight that is to be attached to that evidence.
Indeed, the concept of the general admissibility of relevant evidence is recognised in international arbitration. It was largely taken from the common law tradition (e.g. the US evidence law with respect to admissibility establishes one seemingly simple rule: all relevant evidence is generally admissible, evidence which is not relevant is not admissible). Thus, generally speaking all relevant evidence is admissible in arbitration, except as otherwise provided by mandatory rules, or by agreement of the parties.
The concept of deciding to “admit” or “exclude” evidence gives the broad meaning to the term “admissibility”; that includes the evaluation and assessment of evidence in deciding the case. Arbitrators admit evidence; that is why admissibility is the most general condition for evidence to be admitted. In theory, the tribunal shall not consider evidence ruled irrelevant, immaterial or inadmissible sensu stricto. That is, evidence is admissible sensu lato if the criteria of relevance, materiality and admissibility sensu stricto are met. We believe that when arbitration rules and arbitration laws refer to “admissibility” as the specific criterion of evidence they use it mostly in the specific narrow sense which will be discussed further below.
The inadmissibility of evidence may serve as a ground for it to be refused in admission or in the ordering of production, or excluded from evidence if already admitted. Before addressing the main focus of this article, it is necessary to distinguish between refusal or exclusion on purely procedural grounds (e.g. non-compliance with the terms established by the tribunal for submissions) and exclusion on the grounds of inadmissibility. The parties can agree or the tribunal can determine that evidence must be submitted in a timely fashion;
in that respect, the tribunal can set a specific deadline for submission and can refuse any evidence submitted after that deadline. Non-compliance with the deadline by the submitting party does not directly affect the properties of the evidence and shall be dealt with as if it were a procedural issue, that is, evidence may be admitted if procedural fairness is not prejudiced. A similar approach can be taken if a party requests leave to exclude documents which were not exchanged. The arbitrators should not consider those documents to be automatically inadmissible. Where documents were not exchanged in accordance with the rules of procedure, the arbitrators may adjourn the hearing to afford the disadvantaged party a fair opportunity to examine and comment on the documents.
Relevance is probably the first matter parties and arbitrators have to consider when deciding whether particular materials deserve to be offered as evidence or requested for production.
The term “relevant evidence” in common law generally means evidence having a tendency to make the existence of any fact that is of consequence in the case more probable or less probable than it would be without the evidence. This definition is used in the common law of evidence in which the concept of materiality is merged with relevance. However, in international commercial arbitration these criteria are separated, as the majority of arbitration rules empower arbitrators to decide on relevance and materiality. The International Bar Association (IBA) Rules on the Taking of Evidence in International Arbitration (the IBA Rules) clearly specify relevance to the case and materiality to its outcome as two main criteria for evidence to be admitted or requested for production. Having that in mind we would tend to define the relevance of evidence in international arbitration as having a logical connection with what the evidence purports to prove in the case.
Both common and civil law take the approach that relevance, not being a matter of stricto law, rests upon common sense and a general convention about reasoning. In civil law the connection with the law is more visible: the applicable law helps to determine in general what is relevant (i.e. what needs to be proved for a case to be resolved).
Though relevance is named as the first criterion for admission in practice it is not easy to separate the wheat from the chaff. Arbitrators are reluctant to limit the evidence that can be submitted and normally err toward permitting parties to present evidence, including the introduction of materials of questionable relevance. Arbitrators are mindful of the fact that their award can be set aside if a party was “unable to present the case”. It should be emphasised that parties must only be afforded a fair opportunity for presenting their cases; this does not mean that arbitrators are required to wait until a party actually avails itself of the right to be heard. Thus, a party cannot insist on the admission of evidence that the tribunal considers irrelevant. However, it would be unrealistic to rely fully on that way of thinking. As national arbitration laws usually do not counterbalance that basic principle of arbitral proceedings which protects the party’s right to be heard and do not help arbitrators with any special rules concerning relevance, we may conclude that any irrelevancy-based refusal to admit evidence submitted by a party is associated with significant risk, unless the evidence is manifestly irrelevant.
In common law systems the concept of materiality is merged into the concept of relevance and retains no independent viability. However, in international arbitration practice the materiality criterion is considered mostly in relation to its connection to the outcome of the case, whereas relevance concerns the general relationship between evidence and the case.
The relevance of each element of materiality is a more or less independent category, just as the admission of one piece of evidence that proves the fact in question does not diminish the relevance to the same fact of any other piece of evidence. Materiality is a dependant category: after the admission of one piece of evidence each subsequent piece of evidence or testimony concerning the same fact becomes less material. Materiality is thus ultimately connected with the sufficiency of evidence: after the tribunal is provided with sufficient evidence any other relevant evidence of the same fact is no more material to the outcome of the case. (For example, once the date of the appointment of a company’s officer is confirmed by an extract from a public register there might be no need for the tribunal to receive a board resolution on the appointment.) Tribunals can exclude evidence that is duplicated. Such evidence is excluded as lacking sufficient materiality, but not as lacking relevance. Besides, evidence cannot be both irrelevant and immaterial, as the materiality can be assessed only with respect to the relevant evidence.
The risk associated with immateriality-based refusal to admit evidence may be significant where a tribunal ultimately recognises the non-existence of the facts which a party was aiming to prove with that evidence. It may happen that, after sufficient evidence has been admitted and any further evidence has been refused, the reliability of some of the admitted evidence becomes doubtful. This situation may require the tribunal to invite the parties to present additional evidence.
Admissibility sensu stricto
In contrast to relevance and materiality, admissibility sensu stricto is a purely legal criterion.
Arbitral tribunals possess broad discretion in determining admissibility sensu stricto under the IBA Rules and the majority of arbitration rules. However, this discretion faces a number of limitations in practice, even in those cases when arbitration rules specifically state that evidence need not be admissible in law to be admitted by the tribunal.
The admissibility criterion functions through rules of exclusion which are based either on the assumption that a trier-of-fact may attach undue weight to particular types of evidence (e.g. hearsay) or on the belief that certain values or interests need to be protected (e.g. privileges). Unlike the relevance and materiality criteria, the legal systems in many countries are quite specific about the limits imposed on the admissibility of particular evidence. This might be illustrated by the fact that substantive laws in many countries often contain specific rules which affect the admissibility of evidence in national court proceedings. This often leads to a debate about whether those rules are procedural (and as such part of the lex fori but not part of the lex arbitri), in which case it is not mandatory for arbitrators to follow them; or whether those rules are integral parts of the substantive law and have to be applied by the arbitrators. In the latter case it is within the arbitrator’s discretion how to apply the substantive law. However, it is widely recognised by scholars and arbitration practitioners that the discretion of arbitrators in determining admissibility is subject to the following limitations:
1. Evidence obtained in a manner that is contrary to international public policy (e.g. testimony obtained through torture) shall not be admissible.
2. Evidence may be protected by a privilege or secret (professional privilege, trade secrets, governmental secrecy). Although most arbitration laws and most arbitration rules are silent on this issue, it is generally recognised that arbitrators must take into consideration the above-mentioned privileges and secrets.
It is beyond the scope of this article to analyse problems related to privilege in arbitration.
It is sufficient to say that privilege rules allow a person or party to refuse to disclose certain information, even though that information might be relevant and reliable. At this point we can agree that privilege rules affect the admissibility of evidence.
Thus, while deciding on admissibility sensu stricto arbitrators should take info consideration at least international public policy and the applicable privilege rules. In both cases it is the party that opposes the admission of the evidence who bears the burden of proof that the evidence is inadmissible sensu strictu.
When can questions of admissibility be determined in arbitration?
The practice in international arbitration tribunals is to admit most if not all the evidence and to rely on the arbitrators’ skills in the evaluation of evidence. Arbitrators are akin to professional judges, but not to the jury in common law courts. They do not need to be “protected” from hearing inadmissible evidence or withdrawn till the question of admissibility is argued by the parties and decided by the professional judge. Although in many national court proceedings there is a fundamental distinction between evidentiary objections that affect the admissibility of evidence and those that affect its weight, in international commercial arbitration it is the same person, the arbitrator, who decides the admissibility and the weight of evidence. Thus, the classic concerns of common law courts on the admissibility of evidence do not apply in international arbitration. How does this statement correspond to a situation in which a tribunal determines that it should not review a document of questionable admissibility, and appoints an independent expert to review and report on that document without disclosing its content? This is not a case of arbitrators protecting themselves from considering any potentially inadmissible evidence. The practice of review by an impartial expert is used only because in deciding whether the evidence is admissible the in camera procedure is not generally applicable in international commercial arbitration, for any evidence presented to arbitrators should be presented to the other party as well.
However, if the question of admissibility sensu stricto can be considered before the content of the evidence has reached the tribunal, it may be useful to do so. We may not disagree that “it is not easy to ‘undo’ an impression, even if it is ultimately decreed that the impression should never have been made”.
Thus, it is advisable to raise questions relating to the relevance, materiality or admissibility of evidence at the time when the evidence is submitted or requested for production. In cases where the above questions were not known at the time when the evidence was submitted, they may be raised immediately after the issues do become known.
Different approaches to admitting, ordering production of or excluding evidence
As Patricia Shaughnessy noted, refusing to admit evidence is distinct from, albeit related to, refusing to order the production of evidence. We might even say that arbitrators usually deal differently with the criteria of relevance and materiality when there are questions about the admission of evidence offered by a party, requests for the production of evidence or the exclusion of the admitted evidence to be decided.
This difference is colourfully demonstrated in documents of the International Institute for Conflict Prevention and Resolution in which it is recognised that “since requests for information based on possible relevance are generally incompatible with the need for speed and efficiency, disclosure should be granted only as to items that are relevant and material and for which a party has a substantial, demonstrable need in order to present its position”.
Though materiality has already been mentioned, the “substantial, demonstrable need in order to present its position” which a party might have is nothing more than the materiality criterion. Thus, since the production of evidence requires the expenditure of time and effort, the materiality criterion is the one to emphasise.
The IBA Rules also show how the admission of evidence submitted is distinct from ordering the production of evidence. Materiality is paramount in the latter case as the tribunal needs to compare the production of evidence with the burden which has to be placed on the other party or on other resources to be spent. The IBA Rules require the party requesting the production of evidence to state how the documents requested are relevant to the case and material to its outcome. The IBA Rules mention the criterion of materiality also in respect to evidence which is submitted by a party, but in this case it is mentioned rather as a reminder of the ideal qualities of evidence. This is confirmed also by the wording of art.3.11, which requires the party to believe in the relevance and materiality, not to confirm or prove them. Compliance with the lowest standard of relevance and materiality is tested when the tribunal is deciding on the exclusion of evidence. According to the IBA Rules,
any document, statement, testimony or inspection can be excluded from evidence if they lack sufficient relevance or materiality. Arbitration practice shows that cases in which evidence is excluded for lacking relevance or materiality may be rare. Material must be manifestly irrelevant for it to be excluded.
This lowering of the relevance and materiality requirements allows us to distinguish three “standards” which are applied when deciding on the admission of evidence, ordering production and exclusion.
3. Evaluation of Evidence
It is commonly recognised that the admissibility of evidence does not automatically guarantee that the evidence will be considered as having probative value. There are more or less explicit relevance, admissibility and materiality criteria for determining whether a piece of evidence is admissible, whereas the methods for weighing evidence and determining the sufficient level of proof are subjective and somewhat inexplicable. The weight of the evidence usually refers to its persuasive effect on the arbitrator’s mind. It is within the discretion of the tribunal to evaluate the evidence submitted by the parties, though the parties can agree on the sufficiency, as will be discussed later.
The weight of the evidence includes questions of credibility (reliability) and the evaluation of inferences which can be made from the evidence.
The ultimate question for any evidence is whether it constitutes reliable proof of what it is offered to prove. This, of course, turns on a closer inspection of the credibility or reliability (both words are used as synonyms of the other) of the evidence in question. The assessment of credibility is one of the functions of an arbitrator when weighing up evidence. At the admissibility stage the evidence is required to be prima facie credible, that is, it must ave sufficient indicia of reliability and authenticity to establish that it appears to show what it is offered to prove. The above definition is of course circular, in that it refers to the “reliability and authenticity of the evidence” in defining “credibility of the evidence”. To put things more simply we can define “credibility” as the capacity of being worthy of belief or confidence, trustworthy.
When the reliability of a document is under investigation the question of its authenticity is logically prior to the question of reliability. An authentic document—i.e. not a forgery—may or may not be reliable evidence of what it purports to show; and yet, if a document is not authentic it is hard to imagine how it could be helpful. As a result, the authenticity of a document is often considered to be a precondition for its use in an arbitration proceeding.
Depending on the nature of the evidence and the manner in which it is introduced, a number of different indicators (internal or external) might be applied to assess its reliability.
Internal indicators of reliability of a document include elements of the document itself, such as signatures, the form of handwriting, etc. External indicators might rely on other evidence: testimony about how a document was obtained, expert testimony on its authenticity, etc.
There are also many techniques used when deciding on the reliability of material or testimonies that contradict each other (e.g. where the inconsistency is between a current and a prior assertion of fact, the more recent assertion generally is disregarded in favour of the earlier assertion).
The tribunal can admit relevant and admissible evidence even though the credibility is in doubt. This admission does not mean that the tribunal cannot afford that evidence decreased probative value. Ultimately, it is very unlikely that the tribunal would decide to not admit something into evidence because its authenticity has not been established. In fact, it might be better to admit that document as the reliability of evidence can be fully assessed in its entirety with all other evidence. Besides, if a document is revealed to be a forgery, it is the document itself that proves the lack of authenticity, therefore its admission is necessary.
To carry its burden of proof, a party has to offer sufficient evidence. It is probable that it is because this statement is self-evident that the term “sufficiency” is almost never mentioned in arbitration rules. In assessing the sufficiency of evidence the arbitrator must determine whether there is any valid line of reasoning and permissible inferences which could lead a rational person to the conclusion reached by the arbitrator on the basis of the evidence.
Basically, sufficiency is not a criterion of a single piece of evidence, but rather the point at which the standard of proof is met. After reaching that point any additional evidence, though it is relevant and admissible, does not add anything material to the process of proving the fact.
Although the sufficiency can be finally assessed only in the process of weighing the evidence, the tribunal can take the initiative to manage the case by indicating in advance the evidence that it considers necessary to establish the prima facie proof of certain facts at issue. In doing so arbitrators should adhere to the due process and impartiality principles of arbitration proceedings.
4. Party Autonomy and Arbitrators’ Discretion
In the legal literature analysing the roles of the parties and the tribunal in evidentiary matters, it is often concluded that the rules governing arbitral proceedings make it clear that the admissibility, relevance, materiality and weight of any evidence are for the arbitrator to determine. Sometimes it is considered that the tribunal itself determines the relevance, materiality and probative value of all evidence submitted by the parties, and does not need to hear arguments from the parties concerning these matters. It is thus for the parties to submit the evidence and for the tribunal to evaluate it.
We would rather tend to the approach that allows the parties to retain autonomy and control over the evidentiary issues, even though the parties rarely use their autonomy in that way. The majority of arbitration laws do establish an environment in which the parties can agree on the application of almost any criteria for the evidence. As noted by Holtzmann and Neuhaus the UNCITRAL Model Law art.19(2) specifies that the power conferred on the arbitral tribunal includes the power to determine the admissibility, relevance, materiality and weight of any evidence. Of course, this provision is not mandatory and is therefore subject to the parties’ will. Alternatively, they can agree on particular rules of evidence, fori example that certain evidence should be deemed inadmissible, or that a certain kind of document be sole type of evidence. In this case, the arbitral tribunal should abide by that choice.
Of all the criteria used, the inadmissibility of particular evidence (e.g. special contractual rules which exclude certain materials from evidence) and the issue of sufficiency are the two criteria on which the parties most frequently agree. Though it is the arbitrators’ function to determine whether the evidence in its entirety would rationally support a party’s allegation, the parties may agree on what may constitute sufficient evidence (e.g. by way of indicating in the contract a document that constitutes sufficient evidence of fulfilment of a contractual obligation).
The inadmissibility of illegally obtained evidence is a significant element in compliance with public policy rules. That is why in deciding on the admissibility of evidence the tribunal cannot be as flexible as in any other evidentiary matters. In deciding on the admissibility of evidence the autonomy of the parties and the tribunal may not be contrary to international public policy.
In any case, even if the parties rely on the tribunal in matters related to the application of the criteria for evidence, this only means that the parties do not introduce any specific rules regarding how the relevance, materiality and other criteria should be “measured”.
However, when a party submits evidence the opposing party should be provided with an opportunity to comment on the relevance, materiality, admissibility or probative value of that evidence.
As we have seen the threshold for admitting evidence in international arbitral tribunals is typically quite low, so admissibility, the key question in much of the legal discussion in courts in common law countries, is not a crucial issue in arbitration. Arbitrators focus more on the evidentiary weight they are going to give the evidence.
The basic prerequisites of the admissibility of any kind of evidence in international arbitration are its relevance and materiality. In general, if evidence is shown to be relevant, material, prima facie credible, and is not barred by an exclusionary rule, it is admissible.
The liberal application of these seemingly simple evidentiary rules is necessary to avoid sterile legal debate over admissibility so that the tribunal can concentrate on the pragmatic issues of the case. The tribunal should not, however, be loaded with excessive unnecessary materials. From that perspective it might be reasonable for the tribunal to determine that when introducing documents the submitting party must provide a short description of each document, clearly specifying its relevance, materiality and probative value.
Though in international arbitration the trier-of-fact is not a civilian jury, but a professional fully capable of admitting evidence and rendering objective determinations about the probative value at the end of the proceeding, questions of admissibility should be decided initially, especially if there are claims that any privilege rules apply or if any violation of international public policy is shown. Consideration of any such matters should not lead the procedure into technical formalities.
Three rather different approaches might be applied when deciding on admission of evidence, ordering production and exclusion: the relevance and materiality of the evidence submitted by a party must be shown for it to be admitted; the relevance and significant materiality must be stated by a party and recognised by the tribunal for the production of evidence to be ordered; for evidence already admitted to be excluded it must be manifestly irrelevant.
The above “standards” do not apply when dealing with admissibility sensu stricto, because any inadmissible evidence needs to be refused in admission, excluded or refused in ordering for production.
The criteria for the admission and evaluation of evidence discussed in this article ave very much connected to each other. They may be separated and grouped for convenience only. In practice the necessity for the application of any of the criteria for evaluation may arise at the admissibility stage. The assessment of the materiality of any additional evidence requires the admitted evidence to be evaluated. The categories of relevance, materiality and weight of evidence are dynamic—anything that was irrelevant to the case may become relevant while the case is in progress. The admissibility and reliability of evidence are more static criteria; they are almost exclusively attached to the evidence itself and have little relation to the case and other evidence. If the authenticity and thus the reliability of a document is in doubt it is so irrespective of the essence of the case, even if the authenticity was put in doubt because of contradictions between that document and the set of other materials.
Finally, we must emphasise that party autonomy extends also to evidentiary issues and the application of criteria for evidence, and being a crucial value cannot be negatively affected. The arbitral tribunal should abide by the agreement of the parties unless it is contrary to international public policy.
 In this article the term “evidence” refers to both the information which is supposed to prove a fact in dispute and the means of evidence which contains such information. This term is also used with respect to materials, expert opinions or witness statements which being refused in admission or production, or excluded from evidence or production cannot be considered as evidence in its strict sense if they lack any of the criteria of the admissible evidence.
 See e.g. English Arbitration Act 1996 s.34(1) and (2); German ZPO s.1042(4); Austrian ZPO s.599(1); Ukrainian International Arbitration Act s.19(2).
 US Federal Rules of Evidence r.402.
 George M. Von Mehren and Claudia T. Salomon, “Submitting Evidence in an International Arbitration: The Common Lawyer’s Guide” (2003) 20(3) Journal of International Arbitration 290.
 Among the exceptions is the Swedish Arbitration Act s.25 (SFS 1999:116) that empowers arbitrators to refuse to admit evidence if it is “manifestly irrelevant”.
 See US Federal Rules of Evidence r.401, which deals with relevance and does not mention materiality as a separate criterion.
 See, e.g. Singapore International Arbitration Centre (SIAC) Rules 2013 r.16.2.
 The Ukrainian Civil Code art.218(1) envisages that a court decision cannot rest on witness testimonies if some matters regarding the existence of an agreement are in question.
 The AAA International Arbitration Rules are among the exceptions. They explicitly state that “[t]he tribunal shall take into account applicable principles of legal privilege, such as those involving the confidentiality of communications between a lawyer and client” (art.20).
 See Anne Véronique Schlaepfer and Philippe Bärtsch, “A Few Reflections on the Assessment of Evidence by International Arbitrators” (2010) 3 International Business Law Journal 211. See also the IBA Rules art.9(2)(b) which envisage that the tribunal has the power to exclude any evidence on the grounds of “legal impediment or privilege under the legal or ethical rules determined by the Arbitral Tribunal to be applicable”.
 Richard M. Mosk and Tom Ginsburg, “Evidentiary Privileges in International Arbitration” (2001) 50(2) The International and Comparative Law Quarterly 345.
 Michelle Sindler and Tina Wustermann, “Privilege Across Borders in Arbitration: multijurisdictional nightmare or a storm in a teacup” (2005) 23(4) ASA Bulletin 620.
 Ula Cartwright-Finch and Craig Tevendale, “Privilege in International Arbitration: Is It Time to Recognize the Consensus?” (2009) 26(6) Journal of International Arbitration 834.
 Patricia Shaughnessy, ”Dealing with Privileges in Arbitration” (2007) 51 Scandinavian Studies in Law 457.
 CPR Protocol on Disclosure of Documents and Presentation of Witnesses in Commercial Arbitration art.1(a), available online at: http://www.cpradr.org/Resources/ALLCPRArticles/tabid/265/ID/614/CPR-Protocol-on-Disclosure-of-Documents-and-Presentation-of-Witnesses-in-Commercial-Arbitration.aspx [Accessed February 18, 2014].
 IBA Rules art.3.3(b).
 IBA Rules art.3.11.
 IBA Rules art.9.2(a).
 Audun Jøsang and Viggo A. Bondi, “Legal reasoning with subjective logic” (2001) 8(4) Artificial Intelligence and Law 290.
 H.N. Holtzmann and J.E. Neuhaus, A Guide to the UNCITRAL Model Law on International Commercial Arbitration: Legislative History and Commentary (The Hague: Kluwer Law, 1995), p.566.