Besides being interesting obects if study in their own right, galaxies are important tracers of structure formation, clustering properties, chemical enrichment, reionization, etc. In general, studying galaxies at high redshift is hampered by the technical challenges of getting enough photons frome the objects. The objects we can study are usually unresolved, and the samples biased by their brightness, either in radio, UV or strong, nebular emission lines. However, in some cases, distant galaxies line up behind massive galaxies or clusters whose gravity acts as a lens, greatly magnifying the galaxy and allowing for study in a level of detail which is otherwise only possible for local, low-redshift galaxies. Since the lense not only magnifies the background object spatially, but also enhances its brightness, it allows for spectroscopic studies not otherwise possible for galaxies at these distances. In this colloquium, I will show how this can help us compare these galaxies to local ones in an apples-to-apples way, wich allows to draw more direct, less model-dependent conclusions about galaxy evolution over time spans of ~ 85% of the age of the Universe.