Magnetic confinement fusion science leads many other branches of plasma physics in its capacity to predict, interpret and understand the behaviour of energetic particle populations. The range of applications of thiscapability should be extended, for the mutual benefit of fusion researchand of other branches of science. In this paper we review progress inapplying fusion-derived techniques to one of the central questions ofastrophysics: the origin of the cosmic ray population that is magneticallyconfined within our Galaxy. While it is widely believed that supernova remnantshocks provide the main acceleration sites for cosmic ray electrons andprotons, the fundamental `injection' problem remains. Namely, how particlesare initially accelerated from ambient thermal to mildly relativistic energies,beyond which Fermi-type processes take over. The cosmic ray injectionenvironment is magnetized and has many other physical resemblances tobeam-heated and deuterium-tritium tokamak plasmas; in consequence, many of the same physical processes come into play. These include, for example, collective beam-plasma instability, resonant wave-particle coupling, and thestochasticization of particle orbits. A broad range of analytical and numericaltechniques familiar in the fusion context has been successfully applied to theinjection problem (see, for example, Dieckmann M.E. et al 2000Astron. Astrophys. 356 377). Ideas from magnetic fusionhave also been used to help design and interpret recent magnetized plasmaexperiments (Woolsey N.C. et al 2001 Phys. Plasmas8 2439) using the high-power VULCAN laser, which address the cosmic rayinjection problem from a new perspective.