Abstract
Rapidly rotating stars are readily produced in binary systems. An accreting star in a binary system can be spun up by mass accretion and quickly approach the break-up limit. Mergers between two stars in a binary are expected to result in massive, fast rotating stars. These rapid rotators may appear as Be or Oe stars or at low metallicity they may be progenitors of long gamma-ray bursts. Given the high frequency of massive stars in close binaries it seems likely that a large fraction of rapidly rotating stars result from binary interaction. It is not straightforward to distinguish a a fast rotator that was born as a rapidly rotating single star from a fast rotator that resulted from some kind of binary interaction. Rapidly rotating stars resulting from binary interaction will often appear to be single because the companion tends to be a low mass, low luminosity star in a wide orbit. Alternatively, they became single stars after a merger or disruption of the binary system during the supernova explosion of the primary. The absence of evidence for a companion does not guarantee that the system did not experience binary interaction in the past. If binary interaction is one of the main causes of high stellar rotation rates, the binary fraction is expected to be smaller among fast rotators. How this prediction depend on uncertainties in the physics of the binary interactions requires further investigation.