Le quasi-contrat est consacré comme une source autonome d’obligation dans le code civil. Pourtant, l’unité de la notion est problématique. En effet, la loi définit le quasi-contrat comme un fait volontaire dont il résulte un engagement alors que la doctrine le présente comme une notion à mi-chemin entre le contrat et la responsabilité civile extra-contractuelle en raison de l’application du régime contractuel en dehors d’un accord de volontés des parties. Or, en pratique, hormis la gestion d’affaires qui obéit au régime du mandat sans pouvoir exprès, l’enrichissement sans cause, (désormais enrichissement injustifié) et la répétition de l’indu n’obéissent pas à un régime contractuel. Si la doctrine a tenté de proposer le critère du transfert de valeurs dépourvu de cause pour justifier l’unité de la notion, ce critère se trouve mis à mal par la jurisprudence sur les loteries publicitaires. La thèse tente ainsi de rechercher une nouvelle unité, recherche qui s’avère difficile, voire impossible, dans la mesure où chaque quasi-contrat répond à des règles tantôt propres au contrat, tantôt propres à la responsabilité civile extra-contractuelle, ce qui vient alors fausser l’autonomie du quasi-contrat / Quasi-contracts are set as an autonomous part of the civil code, yet the notion answers to no unity. Law defines the notion as a voluntary fact leading to bind its author but scholars are considering quasi-contracts to be halfway between contract law and tort law due to the lack of will leading nevertheless to binding obligations. Practice shows that, except for the business management based on a mandate without any express will, the other quasi-contracts, unjust enrichment and repetition of undue do not follow any contractual regime. Scholars have attempted to propose the transfer of value for explaining the unity of the notion but this criterion is no longer relevant since cases creating a new quasi-contracts: lotteries. This work aims to discover a new criterion to find back unity of quasi-contracts. Impossible task as each quasi-contract shares common aspects of contract law and of tort law

This PhD dissertation is about legal fictions in private law. A legal fiction, broadly, is a false assumption knowingly relied upon by the courts. The main aim of the dissertation is to formulate a test for which fictions should be accepted and which rejected. Subsidiary aims include a better understanding of the fiction as a device and of certain individual fictions, past and present. This research is undertaken, primarily, to establish a rigorous system for the treatment of fictions in English law - which is lacking. Secondarily, it is intended to settle some intractable disputes, which have plagued the scholarship. These theoretical debates have hindered progress on the practical matters which affect litigants in the real world. The dissertation is divided into four chapters. The first chapter is a historical study of common-law fictions. The conclusions drawn thereform are the foundation of the acceptance test for fictions. The second chapter deals with the theoretical problems surrounding the fiction. Chiefly, it seeks precisely to define 'legal fiction', a recurrent problem in the literature. A solution, in the form of a two-pronged definition, is proposed, adding an important element to the acceptance test. The third chapter analyses modern-day fictions and recommends retention or abolition for each fiction. In the fourth chapter, the findings hitherto are synthesised into a general acceptance test for fictions. This test, which is the thesis of this work, is presented as a flowchart. It is the author's hope that this project will raise awareness as to the merits and demerits of legal fictions, de-mystify the debate and bring about reform.