Engineering problems involving soft rocks have traditionally been solved by extrapolating from the historically separate technologies of either soil mechanics or rock mechanics. This has generally lead to tenuous, conservative solutions which have contributed little to the overall understanding of soft rock behaviour. It is argued that soft rocks are not vague extensions of these two geotechnical sciences but are in fact central to one continuous science extending from soft soils through to hard rocks. All geotechnical materials behave according to the same engineering principles, with obvious differences a function of degree rather than of fundamental nature. Consequently, it is demonstrated that for soft rock technology to emerge as a rational engineering form, the established methods of soil mechanics and rock mechanics should be applied together and not in isolation.