Ground engineers routinely use simple relationships – rules of thumb – to obtain soil parameters and to design ground works. Some of these have a sound theoretical basis and can be applied generally while some are purely empirical and so should be applied only within the limits of the observations used to derive them in the first place. A classification for rules of thumb was suggested by Wroth (1984) and this has been used to examine the theoretical basis – or lack of it – for some of the more common empirical rules in geotechnical engineering.