While digitalisation for a long time carried the expectation of a more participatory, more horizontal public, its effects on democracies have recently been more often discussed under headings such as polarization and fragmentation. No matter whether one takes a critical or an apologetic stance it is pretty straightforward that the realities of many-to-many-communication have a strong impact on how actual democracies function. In this paper I will discuss what obligations do arise from the recent shifts in our media landscape and whether former theories of democracy and their conceptualization of the public sphere still hold. If democracies have to rely on a certain kind of public discourse in how far do they have to provide the infrastructure that allows democratic discourse to flourish? Do they have to counter fake news or hate speech? Or do they just have to ensure plurality? Do they have to have certain instances of public, quality-controlled broadcasting? What kind of communication is covered by this obligations and what layer of communication infrastructures is the right one to wield effective as well as legitimate influence? And how to deal with algorithmic sorting, transnational publics , etc. The paper will present an attempt to revitalize the discourse on the shaping of the public sphere as an obligation of democracies and ask whether ideas of an active democratic shaping of the public sphere can still play a role in a public sphere that is structured totally different than the mass public spheres of the 20th century.
