Introduction

As previously mentioned, under Article 38(1)b of the Statute of the International Court of Justice (ICJ) the formation of a rule of ‘international custom’ requires a ‘general practice’ by States which is ‘accepted as law’. This chapter examines the first requirement necessary to demonstrate the existence of any customary rule: State practice. Before doing so, I will first briefly discuss the preliminary (and yet fundamental) question of whose practice actually matters for the creation of customary rules. In other words, I will enquire as to whether or not the practice of non-State actors is relevant to the phenomenon of custom creation in both general international law and investor-State arbitration (Section 3.1). The next section will examine the basic requirements for the practice of States to be considered relevant in the process of creating customary norms. I will show that the practice of States needs to be uniform, consistent, extensive and representative for a rule of custom to emerge (Section 3.2). I will specifically examine whether this is the case regarding the fair and equitable treatment (FET) standard found in the vast majority of BITs.

The objective of this chapter is to analyse the different ‘manifestations’ of practice, that is, the types of evidence of State practice that are relevant for the creation of custom in both general international law and investor-State arbitration (Section 3.3). I will start by making a number of introductory remarks on the requirement that the practice be public and whether or not omissions should count as practice (Section 3.3.1). I will also examine the variety of forms of State practice and their relative weight (Section 3.3.2). I will then specifically examine four different categories of ‘manifestations’ of State practice.

First, I will assess the role of treaties (Section 3.3.3). I will observe that treaties can contribute to the formation of new rules of customary international law and that they can also codify existing customary rules. In the context of arbitration, I will examine the strict conditions under which rules contained in BITs can transform into customary rules and the reason why such transformation is very rare in practice. In this context, I will analyse the controversial proposition made by some scholars that all BITs taken together represent a ‘new’ custom.