The presence of weak phyllosilicates may explain the low shear strengths of fault zones if they define well-developed fabrics. The growth of phyllosilicates is favored in meteoric water-dominated granitic fault systems, where mineral-aqueous fluid equilibria predict that modal phyllosilicate will increase via feldspar replacement reactions. In deeper, more alkaline, rock-dominated regions, the reactions reverse, and feldspars tend to replace phyllosilicates. In Mg-rich mafic rocks, however, phyllosilicates (chlorite, biotite) can replace stronger framework and chain silicates in both shallower (<~10 km) meteoric H2O-dominated and in deeper, alkaline, rock-dominated regimes. Where these phyllosilicates precipitate in active fault zones, they contribute directly to reaction softening. Because low-temperature deformation of phyllosilicates is not governed by frictional processes alone but can occur by pressure-independent dislocation glide, the strength of phyllosilicate-rich fault rocks can be low at all depths. Low strain rate creep during interseismic periods can align phyllosilicate grains in foliated gouge and phyllonites. Where preferred orientations are strong and contiguity of phyllosilicates is large, strengths of rocks within fault zones may approach minimum strengths defined by single phyllosilicate crystals. Fault zones containing localized high concentrations of phyllosilicate with strong preferred orientations in well-defined folia can exhibit aseismic slip, especially where mafic Mg-rich rocks occur along the fault (like parts of the San Andreas Fault). ¿ American Geophysical Union 1995