Two Genuinely Different Ways a Wall Can Fail
A retaining wall's stability against the lateral thrust of retained soil is checked through (at minimum) two independent failure modes: overturning, where the wall rotates forward about its toe, and sliding, where the entire wall/footing assembly translates horizontally along its base. These are physically distinct failure mechanisms, checked with different force and moment balances, and a wall that comfortably passes one check can still fail the other — which is exactly why both are required, not treated as redundant checks of the same underlying risk.
The Overturning Check — A Moment Balance
Overturning stability compares two moments taken about the toe (the front edge of the footing, the natural pivot point for a forward-rotating failure): the resisting moment (Mr), contributed by the wall's own weight — stem, footing, and any backfill soil sitting on the heel — each acting through its own centroid at some horizontal distance from the toe; and the overturning moment (Mo), contributed by the active earth pressure thrust acting at its point of application (H/3 above the base for the triangular soil-pressure component, H/2 for a uniform surcharge component). The factor of safety against overturning is FS = Mr / Mo, with a standard target of at least 2.0 for allowable-stress design.
The Sliding Check — A Force Balance
Sliding stability, by contrast, compares horizontal forces directly rather than moments: the resisting force, typically calculated as the total vertical load (ΣV) multiplied by an assumed friction coefficient (μ) between the footing base and the underlying soil, against the driving horizontal force from active earth pressure (Pa). The factor of safety against sliding is FS = (μ·ΣV) / Pa, with a standard target of at least 1.5.
Why a Wall Can Pass One Check and Fail the Other
Because overturning depends on moment arms (how far the wall's weight components sit from the toe) while sliding depends only on total vertical load and the friction coefficient (not on where that load is positioned), design changes that help one check don't automatically help the other equally. A wall with a very wide heel (lots of soil weight far from the toe) can have excellent overturning resistance from that far-out weight, but if the total vertical load isn't large enough relative to the horizontal thrust, sliding resistance (which only cares about total ΣV, not its distribution) could still be marginal. Conversely, a compact wall with most of its weight concentrated near the toe might slide-check adequately (enough total weight) while overturning-checking poorly (that weight isn't positioned far enough from the toe to generate a large resisting moment).
Design Responses Target Each Check Somewhat Differently
Improving overturning resistance most directly comes from widening the heel (moving weight — both wall concrete and backfill soil — further from the toe, increasing the resisting moment arm) or increasing the base width generally. Improving sliding resistance most directly comes from increasing total vertical load (a wider or thicker footing, more backfill soil engaged), improving the base friction coefficient (a rougher footing-soil interface, sometimes achieved with a shear key or a roughened footing underside), or adding a shear key that engages passive soil resistance in front of (or below) the footing. A design that fails sliding specifically often benefits more from a shear key than from simply widening the heel, since the shear key targets the sliding force balance directly rather than working through vertical load and friction alone.
Why Both Checks Use Different Standard Safety Factors
The different target factors of safety (2.0 for overturning, 1.5 for sliding) reflect differences in how each failure mode typically progresses and how much warning or ductility exists before failure, along with differing historical convention — a wall approaching overturning failure often shows visible tilting or cracking as a warning sign, while sliding failure can be more sudden. These conventional target values are widely used in allowable-stress design practice, though the governing project-specific code or geotechnical report should always be the final source for required factors of safety on a real project.