Dualities in the classical supergravity limits

Duality symmetries of supergravity theories are powerful tools to restrict the number of possible actions, to link different dimensions and number of supersymmetries and might help to control quantisation. (Hodge-Dirac-)Dualisation of gauge potentials exchanges Noether and topological charges, equations of motion and Bianchi identities, internal rigid symmetries and gauge symmetries, local transformations with nonlocal ones and most exciting particles and waves. We compare the actions of maximally dualised supergravities (ie with gauge potential forms of lowest possible degree) to the non-dualised actions coming from 11 (or 10) dimensions by plain dimensional reduction as well as to other theories with partial dualisations. The effect on the rigid duality group is a kind of contraction resulting from the elimination of the unfaithful generators associated to the (inversely) dualised scalar fields. New gauge symmetries are introduced by these (un)dualisations and it is clear that a complete picture of duality (F(ull)-duality) should include all gauge symmetries at the same time as the rigid symmetries and the spacetime symmetries. We may read off some properties of F-duality on the internal rigid Dynkin diagram: field content, possible dualisations, increase of the rank according to the decrease of space dimension... Some recent results are included to suggest the way towards unification via a universal twisted self-duality (TS) structure. The analysis of this structure had revealed several profound differences according to the parity mod 4 of the dimension of spacetime (to be contrasted with the (Bott) period 8 of spinor properties).