This study consisted in undertaking a bibliographical search within the Web of Science Core Collection from January 1900 to December 2020. A total of 637 publications were identified and divided into 9 sections tackling successively the relevance of independence referendums, the biased authorship, the definition of the phenomenon, the technical features of referendums, the elaboration of comparative datasets, the legitimacy of these consultations, the drivers leading to the organisation of independence referendums, the impact of referendums on settling ethnic violence and their capacity to favour state recognition. We affirm that those publications have advanced our knowledge about independence referendums. We also stress the persistence of a high fragmentation of authorship and approaches limiting the adoption of a common vocabulary, validation methods and consistent datasets allowing the accumulation and replication of analyses for establishing robust theories. In conclusion, we indicate some theoretical blind spots which could constitute a future research agenda.