Experimental results for attenuation reported in connection with the Seabed Characterization Experiment 2017 are discussed and analyzed. Much of the results suggest attenuation depends on depth within a mud layer and that, over a wide frequency range, attenuation is nearly proportional to frequency. The present paper suggests that attenuation in mud is due almost entirely to localized processes, including (1) viscous flow (Stokes flow) around suspended sand particles, (2) sliding friction between particles, and (3) relaxation effects associated with breaking and re-association of bonds between clay particles. Present theories due to Buckingham and Chotiros and others are reviewed, and a fresh case is made for relaxation processes as a dominant contributor. Theoretical results show that a continuous smear of relaxation processes can account for a linear or power-law dependence of attenuation on frequency over a wide range, but excluding frequencies close to zero. Similar considerations apply to propagation of ultrasound in blood. All physically viable mechanisms will result in a frequency-squared dependence at low frequencies, but the dependence can change to linear dependence at frequencies slightly above zero. Theories based on Stokes flow around suspended sand particles can explain the dependence on porosity and on the number density of particle grains. Relaxation processes appear to be ubiquitous, and can be associated with diverse types of solid particles nominally touching each other, sliding, and separating.