It has generally been accepted that the volume of cement hydration products is slightly smaller than the original volume of cement and water. However, this does not mean that the hydration reaction causes the hardened cement paste and concrete to contract. In fact, C-S-H shells that grow around anhydrous cement grains push the neighbors apart by crystallization pressure and thus cause the solid framework of cement paste to expand. Proposed here is a new idea-this expansion always dominates over the contraction, i.e., the hydration is, in the bulk, always expansive, while the source of all of the observed shrinkage, whether autogenous or due to external drying, is a compressive elastic strain in the solid caused by a decrease of chemical potential of pore water, with the corresponding changes in pore humidity, surface tension and disjoining pressure. From recent observations of autogenous shrinkage growing logarithmically in time over many years it follows that the growing C-S-H shells surrounding cement grains must act as diffusion barriers for water and ions, which slow down the hydration process and can extend it over many years and even decades. The new idea implies that all of the autogenous shrinkage must be caused by elastic compression (probably with no, or almost no, creep) of crystalline nano-sheets in the solid framework subjected to stresses that arise as a reaction to pore water stresses. Swelling under water immersion is explained by insufficient elastic compression when water is permanently supplied to the pores. The lecture first presents the aforementioned theory and then summarizes some recent advances in related phenomena, particularly a model for oriented damage due to alkali-silica reaction and a method for shrinkage extrapolation.