Force spectroscopy has emerged as a new tool to study protein folding, in which force replaces the chemical denaturant used in traditional folding experiments. This new technique complements older methods and allows a range of new questions to be investigated. What sort of protein is able to resist mechanical unfolding, and to what extent is mechanical stability dictated by fold or function? What is the effect of force on the unfolding energy surface? Do proteins unfold by the same pathway in mechanical and chemical denaturation experiments? Answers to these are starting to emerge based on a combination of experimental and computational approaches. We present some of the forced unfolding experiments to date and simple methods for characterizing the unfolding potential from the results. Several studies have also begun a more fine-grained description of mechanical unfolding, for example by invoking intermediates to explain features seen in unfolding traces and by using mutagenesis to try to localize the origin of mechanical stability. We propose further experimental approaches to this goal using the protein engineering method to characterize transition states, similar to those used in conventional folding experiments. However, it is likely that a high-resolution picture of mechanical unfolding will only emerge through a combined interpretation of careful experimental work and computer simulation.