Multistakeholderism has become one of the fundamental principles of internet governance. It is commonly defined as a model of governance that requires the participation of three stakeholder groups - states, the private sector and civil society - "in their respective role." However, the how, when, and more importantly the why of the stakeholders' respective involvement have not been satisfactorily spelled out. This uncertainty is unsettling. Multistakeholderism is increasingly used to prevent governments from making decisions that may impact the internet pursuant to intergovernmentalism. A result of this pushback against intergovernmental initiatives in the field of the internet is the rarity of internet-specific international law. If multistakeholderism is to prevent the making of international law, its legitimacy should not remain a mystery. Whether we like it or (increasingly) not, traditional international law's legitimacy is firmly grounded in state consent. There is no such clear legitimacy story supporting multistakeholderism. Relying on global constitutionalism, this dissertation addresses this gap. It develops a model for a composite global constituent power, comprised of a state component and a global component, and explains how it may be exercised, whether directly or, under certain conditions, through mediators. The model is then used to assess multistakeholderism. It concludes that the composite global constituent power model legitimizes equal multistakeholderism. In this version of multistakeholderism, states on the one hand, and the private sector and civil society on the other hand, are the mediators of the state component and the global component respectively and are on strictly equal footing.