Is there such a thing as an arbitral precedent? Is it binding to judicial courts? The answer to these questions will depend on the clarification of another set of broader issues relating to the existence of precedents themselves and their inevitable role in shaping today’s possible conceptions of the law. We will analyse these matters and attempt an answer to the initially posed questions.

We have already spoken of this theme in our paper Arbitral Precedent: Once and Again. However, while we previously analysed this matter in an international point of view, now we wish to focus on the arbitral precedent in the context of a national legal framework. We will explain why we believe that arbitral awards may furnish innovative norms, susceptible of being applied by decision-makers in future cases, given that a rational decision-making process is what it takes to arrive to the best decision according to the legal order applied: we believe that formal or institutional criteria which limit and define the precedential value of arbitral and judicial decisions are fictitious, although presenting a practical utility related to judicial uniformization.

For starters, a precedent may be defined, as does Duxbury, “a past event – in law the event is nearly always a decision – which serves as a guide for present action”. Guilherme Rizzo Amaral starts off his study about the possibility of arbitrators being bound by judicial precedents with this quotation followed up by the phrase: “In that sense, one can even maintain the existence of non-judicial precedents”, going on further by stating that “the existence of arbitral precedents could be contended to be somewhere between judicial and non-judicial precedents”.1) Guilherme Rizzo Amaral, Judicial Precedent and Arbitration: Are Arbitrators Bound by Judicial Precedent? 2nd ed, Wildy, Simmonds and Hill Publishing.

The precedent differs from the decision from which it originates. It is the ratio decidendi that may provide elements prone to establish such relevant connections with posterior cases that they present themselves as precedent. There are binding and persuasive precedents and while the former are as such recognized by all courts pertaining to a national legal order – even if only few of them are actually bound by said precedents (might it be because the law says so or because courts themselves take the responsibility of following certain courts’ decisions’ ratio decidendi) – the latter are extracted from whichever decision a decision-maker finds adequate to solve (or help solve) the case in hands: we may, at this point, already conclude that every decision that applies the law of a certain country may be considered a precedent and that it is considered as such whenever another decision-maker applying said law finds so to be adequate. This means the ratio decidendi of a first instance judicial court decision, appellate court decision, as well as arbitral decision may be a precedent, taken into account by a decision-maker in the deciding process. It is, of course, natural for a judicial court judge to decide the same way an arbitrator previously did, whenever the relevant facts (as considered as so by the judge) are analog to those that the arbitrator considered in order to decide a certain way, so far as the judge thinks the conclusion to which the arbitrator arrived is the correct one – if they think so, the judge must be able to explain the thought process leading up to such conclusion, justifying the decision taken.

This said, we understand arbitral decisions may form precedent because arbitrators too apply the law of the country (except when equity is the criterium chosen by the parties for the formation of the decision, which is not a particularity of the arbitral process, since certain judicial court decisions, depending on each country’s rules, are also based on equity and these ones also do not form precedent). W. Mark C. Weidemeier explains it is clear arbitrators cite other arbitrators’ decisions. This is particularly true when it comes to institutionalized arbitration, such as sports arbitration, where the lack of a set of principles or rules applicable means one of the tasks of the arbitrator is to develop the substantive system, which implies that each arbitrator is aware of creating a new rule (which will guide future arbitrators).

Yet, we admit they might not only form precedents for other arbitrators but also for judicial court judges. Arbitration is often seen, as W. Mark C. Weidemeier states, “as an ad hoc forum in which arbitrators do justice (at best) within the confines of particular cases”, but we do not see in this a differentiated characteristic of arbitration, since the judicial court judge also decides the particular case he is faced with: that is what doing justice means. If the arbitrator is set on the goal of doing justice by applying the national Law, then he or she cannot be oblivious to previous decisions, whether rendered by arbitral or judicial courts.

In fact, this discussion only makes sense when dealing with Civil Law, since it is clear the arbitrator in Common Law orders could never only rely on written laws. And it is when considering Civil Law systems that it is most pertinent (yet paradoxical) to assert that arbitration does not mean a positivistic view of the Law: it is not a field in which decision-makers are only to consider written laws. This observation may seem bizarre, since it contradicts the most common and intuitive thoughts about arbitration, often pictured as a more open and resourceful environment to solve a conflict; yet, when seen as an instance where the arbitrator is to read a law and interpret it without resorting to other interpretations, the result is precisely a positivistic approach. We mean to distance ourselves from it.

Both arbitrators and judges pertain to the Law, since they are the highest expression of its interpretation and application, and, therefore, creation of such Law – as Jhering declares, what only exists in the parliamentary statutes and in paper is nothing but a phantom of Law.

All we have said can lead us to a disturbing conclusion: if arbitral decisions were not to be considered by judicial courts, the State would be failing at one of its most crucial duties – delivering justice through its courts, for they would be reporting to incomplete, stale laws (for their application is also their formation), thus creating a parallel legal context in which former solutions are to be ignored, while the reasoning of solutions found without using all possible tools would be legally binding.

In other words, and consequently, we do not think that the contractual nature of arbitral courts should imply that they be excluded from the idea of jurisdictional system, which consistency depends – even more so if we consider the Common Law family – on the concept of preceding rulings being, at some extent, taken into account by decision-makers.

The definition of precedent has already made us arrive to these conclusions, which will be the starting point to the second part of this paper where we will ponder the weight of the traditional distinction made between binding and persuasive precedents and its usefulness, in an attempt to understand which nature an arbitral precedent (which we have just now clarified as existent) may have. As for now, we will take a short look into how precedents are framed by the legal families of Civil and Common Law.

According to the traditional fons iuris theories, only judicial courts decisions could form binding precedents. In the Common Law tradition, the stare decisis principle implies that previous decisions, rendered by the same or by an hierarchically higher court – therefore, either operating horizontally or vertically – should define not only a pattern to subsequent ones, but peremptory criteria for case resolution. Civil Law orders face the same challenges regarding predictability, equality and, ultimately, the deliverance of Justice in the form of just decisions. However, and although past judgments (more precisely, their reasoning) are presented as relevant arguments both by the parties and the judge, they are challengeable and can almost always be diverted from, for they are not equivalent to parliamentary legal dispositions, save for rare exceptions. Nonetheless a remark is to be made: in both families, the formation of precedents – both binding and persuasive – is taken as a consequence of the hierarchic position of the court that renders the decisions, often sustained by assumptions. They regard the quality of the decision-maker (thus implying another set of assumptions), the jurisdiction of the court or even its composition; more profoundly, and considering that, if binding, court decisions are equivalent to written law – which making power belongs to the State – arbitral decisions are not usually accepted as able to form precedent.

We are left – and leave the reader – with this question: according to the aims of the judgment and its generally accepted conditions of validity, does it make sense to justify the binding force of only some precedents based on a classic fons iuris theory when it only concerns institutional legitimacy? This question is one we will attempt to answer from the premise we started this one off with: every decision that applies the Law of a certain country may be considered a precedent and it is considered as such whenever another decision-maker applying said Law finds so to be adequate (which is a decision he or she will have to justify, as well as the decision to dismiss a previous decision’s ratio decidendi when it concerns a similar case).

In the introductory part to this blog post, we took a stance on what a precedent is – a past decision which serves as a guide for present action – and considered that arbitral awards may create self-standing rules with precedential value for both arbitrators and judges deciding a case. We concluded this by considering that the contractual nature of the arbitration agreement does not exclude arbitral decisions from the jurisdictional system and the Law of the country, when the arbitrator is bound to decide by applying such Law: in such case, arbitrators both interpret, apply and create Law. What this means is that state court judges must also consider arbitral decisions when deciding a case. However, we do not think these conclusions can sustain themselves solely based on the definition of precedent and on the nature of arbitration. We must now better understand what deciding is, what a valid and legitimate decision is, what differs persuasive from binding precedents and why States only consider binding the preceding decisions coming from high courts.

If regarded as previously developed schemes for the resolution of an ensuing conflict, it is easily understandable that the binding content of precedents does not lie on the operative part of the judgment, for the final decision cannot be separated from the situation that requires it.

Those schemes – the ratio decidendi – result from the intellectual effort of selecting from the reality, as presented to Court, what might be relevant for the final goal of delivering a judgment. That process is not crucial only for the purpose of reasoning the final decision. It is also the exercise that enables the construction of a pattern: identifying the general contours and the specifying elements of any given situation.

The selection process itself involves a certain understanding of the facts and knowledge of the applicable legal dispositions. From the choosing of elements and their – at least simultaneous – logical concatenation, a chain of deductions can be made, ideally syllogistically culminating in one sole conclusion, but often leading to a plurality of potential valid options, from which one is to be chosen by the decision-maker.

“Valid” is not a naively chosen word. The processes we are discussing are eminently rational and it is in this rationality that the crucial condition of legitimacy for the rendition of a judgment is. It may be argued that a court’s legitimacy derives from its powers being assigned by the State (even if an arbitrator may be perceived as to having less of a public legitimacy) – this would be the positivistic approach to the much wider question of knowing what makes everything binding and an understanding of the word “jurisdiction” that we cannot follow.Even if we were to accept this statement, it would be now helpful to recall what we have previously written: it is not the operative part of the judgment that must be applicable to subsequent situations. It is not the previous court’s given order that must be followed, thus the cornerstone of the precedent doctrine cannot be the mere power to order.

Even though it may be true that only certain entities may, by law, validly pronounce judgments, it is not true that the conditions for that political kind of validity apply to the legal and philosophical validity of a judgment. Coherence between premises and conclusion does not depend on any kind of hierarchy or institutional criteria, but on the prevalence of reason over randomness and, consequently, unpredictability, which is the enemy to be fought against by any legal thinker.

Trying to explain the existence and functioning of precedents using the topoi from which we discuss institutional legitimacy is an incomplete manner of analysing such matter and tends to deny the possibility of formation of binding arbitral precedents. In one word: the only acceptable justification for the very existence of binding legal precedents also implies accepting arbitral precedents, if the process to achieve the final judgment is submitted to the same rationalizing elements. What we mean to say is that, while a decision is only legitimate when formed by a rational process which can be communicated, and rendered by a competent entity, there is no competent entity to say whether the ratio decidendi of a decision may form a binding precedent, since it is the necessity of a rational decision able to put to terms the conflict taking place that may oblige the decision-maker to follow a previous set of decisions.

The idea we want to make clear is that the source of validity of every precedent rests in its ratio decidendi, which can only be achieved by the rationalizing process of sorting and applicating facts and legal criteria, a process which must be linguistically expressed and comprehensible: as Castanheira Neves explains, the relevant rationality is not the one that simply aims to achieve a necessary inference or demonstration or a true and universal knowledge or explanation, but a finalistic, practical and relational rationality that aims to solve definitively the conflict in a way that is understandable by the parties and third persons.2) Castanheira Neves, “Entre o ‘legislador’, a ‘sociedade’ e o ‘juiz’ ou entre‘sistema’, ‘função e problema’ – os modelos atualmente alternativos de realização jurisdicional do direito”, Boletim da Faculdade de Direito de Coimbra, v. LXXXIV.

This is what we meant by writing that the goal of achieving one decision is merely ideal – it is not incompatible with the potential validity of several judgments. The correspondence between a logical conclusion and a court decision is its most fundamental condition of validity. However, it is not because the conclusion derives logically from the premises – as taken to court by the parties – that the judgment is binding, for that would imply an ontological leap: this is where institutional legitimacy plays its most relevant role. It transforms the judiciary syllogism into an order. Indeed, the correctness of an intellectual construction that supports a final illation does not instantly correspond to the reasons of judgment or to the judgment itself. These are different realities, interlocked by institutional powers, assigned to courts by Law. The gap between logic and a declaration that alters a parties’ juridical situation and consubstantiates an enforceable order is both linguistical and logic itself; yet, the chasm is filled by the implicit transformative power that comes along with the power to render a judgment. We are not contradicting our previous statements, for there are two different kinds of legitimacy at presence – the legitimacy derived from a justified and therefore comprehensible reasoning process of deciding (jurisdictional legitimacy, which allows the validity of the decision) and the institutional legitimacy.

There is, in the end, only one optimal decision: the one reasoned (which means it must be driven by practical motives: obtaining the definitive composition of the conflict and the harmonization of the system and all its rules) and institutionally legitimate.

This obliges us to conclude that, institutional legitimacy checked, binding judicial precedents cannot be accepted without also accepting the formation of arbitral precedents and their binding effects. Why? Because the condition of validity for both judgments lies on the same justification: both judge and arbitrator must reach a reasoned decision to stabilize the legal situation of the parties at conflict.

This is why we understand that, in a legal system in which the law aims to reach a definitive solution to a conflict that is not incompatible with any other decisions and legal rules, the difference between binding and merely persuasive precedents ought to not exist, even though it exists in the praxis of both Civil and Common Law. We sustain that, if the mainstay of the precedent doctrine is, as we argued, the validity of the judgment, all decision-makers must try to achieve that rational decision and therefore make use of every source of Law, both formal and informal. Only institutional and political-legislative structures may make use of hierarchic and judicial vs. arbitration differential criteria, for validity criteria are common and equivalent whether they concern arbitral or state court judgments.

This said and if our premises stand, there is no alternative but to perceive both Civil and Common Law as a rational and, therefore, systematic core of legal rules. The word “systematic” might raise some eyebrows, considering that one of the main distinctions made between Civil and Common Law is the systematic nature of Civil Law – however, the path we rationally traced in this paper forces us to conclude that, in Common Law, it is precedents and not legal codes that reveal its systematic nature, preventing contradictory rulings and guaranteeing predictability and equality before the Law.

Traditional – but still taking effect nowadays – constructions only confer binding effects to precedents derived from decisions rendered by higher courts, basing such option on assumptions of various kinds. These constructions stem from the realization of the inexistence of Dworkin’s Judge Hercules and, therefore, the practical impossibility of having a system able of discovering the absolute best (past) decision. The deliverance of Justice is as limited as we are, for the systems chosen are designed by men who have limited amounts of knowledge, experience and time. Yet, we still need to decide; what is most, we need to decide in the best, longest-lasting way possible. Therefore, these constructions are based on assumptions, on the penalty of facing the impossibility to decide in reasonable time. Binding precedents coming only from higher courts are meant to save the judge from the burden of looking into every past decision and deciding which is best, and are meant also to help enforce judicial certainty and predictability. These assumptions aim to help more the system’s continuity than the specified decision taken, for they might not be verified in a certain case. As we understand it, there is no relation of causality between one instance (legal or jurisdictional) saying which precedents (judicial or arbitral) are or are not binding and their actually binding effects, for it is the process of decision making itself that forces a judge to investigate past decisions and juridical opinions to better interpret laws and apply rules and principles. It is the goal – rectius, the judge’s duty – of reaching the optimal decision for the case in hands that might lead him or her to decide according to a past decision.


To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.

Kluwer Arbitration


1 Guilherme Rizzo Amaral, Judicial Precedent and Arbitration: Are Arbitrators Bound by Judicial Precedent? 2nd ed, Wildy, Simmonds and Hill Publishing.
2 Castanheira Neves, “Entre o ‘legislador’, a ‘sociedade’ e o ‘juiz’ ou entre‘sistema’, ‘função e problema’ – os modelos atualmente alternativos de realização jurisdicional do direito”, Boletim da Faculdade de Direito de Coimbra, v. LXXXIV.
This page as PDF

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.