In digital control of flexible mechanical systems, aliasing in data sampling can cause instability in those aliased modes whose natural frequencies are higher than the Nyquist frequency. The sampling rate of a digital controller is usually limited by the hardware restrictions, and by the time needed to calculate the control inputs. Although some low-pass filters have been developed, they only make aliased modes uncontrollable, and do not guarantee stabilization of those modes. Motivated by the need for new techniques of handling aliased modes, this paper investigates stabilization and control of aliased modes of a flexible system via a digital controller of a low sampling rate. As a first attempt to develop fundamental theories on control of aliased modes, we will show that stabilization and control of aliased modes can be achieved even if they can not be identified from the sampled sensor outputs. A stability criterion for digital control of aliased modes will be derived. In addition, through a vibration analysis, the physical explanation for control of aliased modes will be provided.