The first decade of the 21st Century promises to become the Golden Age of extragalactic astronomy. The 2dF and Sloan Digital Sky Surveys have already made significant contributions to our knowledge of the extragalactic universe. On-going and planned wide-field (pencil-beam) imaging and spectroscopic surveys with 4m (8m)-class telescopes from the ground and in space (e.g., SIRTF, SOFIA, Herschel, NGST) will nicely complement these large-scale surveys and should go a long way to answer some of the most fundamental questions in extragalactic astronomy: How do galaxies form? How do they evolve? How do supermassive black holes fit in this picture of galaxy formation? Which objects are the main contributors to the overall energy budget of the universe? To properly answer these questions, one will need to differentiate objects powered by nuclear fusion in stars (i.e. normal and starburst galaxies) from objects powered by mass accretion onto supermassive black holes (quasars and AGNs). A wide variety of diagnostic tools have been used in the past for this purpose with different degree of success.
Due to space limitations, the present discussion focusses on emission-line diagnostics. The fundamental principles behind these diagnostics are reviewed in Section 2. Next, the main diagnostic tools available in the ultraviolet, optical, and infrared domains are described in Section 3, Section 4, and Section 5, respectively. A table listing the main diagnostic lines is given in each of these sections. Additional factors which may complicate the use of these tools are discussed in Section 6. A summary is given in Section 7 along with an outlook on the future. Note that this review is not meant to be exhaustive; it is meant to emphasize the practical aspects of starburst/AGN spectral classification. Readers who are looking for a more detailed discussion of the physics behind these diagnostic tools should refer to the original papers listed in the text.