Based on an Atoms-in-Molecules (AIM) analysis, Matta et al. recently claimed evidence for the existence of hydrogen-hydrogen bonding between ortho-hydrogen atoms, pointing towards each other from adjacent phenyl groups in planar biphenyl. This AIM result is opposed to the classical view that nonbonded steric repulsion between the ortho-hydrogen atoms is responsible for the higher energy of the planar as compared to the twisted geometry of biphenyl. In the present work, we address the question if hydrogen-hydrogen bonding in biphenyl exists, as suggested by AIM, or not. To this end, we have analyzed the potential energy surface for internal rotation of biphenyl in terms of two interacting phenyl radicals using density functional theory (DFT) at BP86/TZ2P. A detailed analysis of the bonding mechanism and a quantitative bond energy decomposition in the framework of Kohn-Sham DFT show that Pauli (or overlap) repulsion, mainly between C(ortho)--H(ortho) phenyl MOs, prevents biphenyl from being planar and forces it to adopt a twisted equilibrium geometry. Furthermore, a derivative of biphenyl in which all four ortho-hydrogen atoms have been removed does adopt a planar equilibrium geometry. Thus, our results confirm the classical view of steric repulsion between ortho-hydrogen atoms in biphenyl and they falsify the hypothesis of hydrogen-hydrogen bonding.