Permanent earth-retaining structures are designed for soil pressure, not water pressure, and the difference between those two loading cases is not a refinement. It is a different problem. A wall backfilled with free-draining material behind a working drain carries the lateral push of the soil skeleton alone. Let that backfill saturate and the wall carries the soil plus a full column of water, with the resultant sitting lower on the stem where it does more damage. That is why drainage sits behind so many wall failures.
Two loading cases, not one degree of difference
The design of a drained wall rests on an effective-stress picture: the retained soil pushes, water passes through, and the pressure reaching the wall is a fraction of the overburden set by the soil's friction angle. Saturate the same backfill and the buoyant soil pushes somewhat less, but water pushes with the full weight of the fluid, and water has no friction angle to reduce it. The combined resultant is substantially larger.
Where that load sits matters as much as its size. Hydrostatic pressure grows linearly with depth below the water surface, so it concentrates near the base, and the overturning moment about the toe rises faster than the total force does. A wall with a comfortable margin in the drained condition can have none at all in the undrained one.
What the codes and manuals require
The lateral soil load provisions of the International Building Code give design pressures for drained backfill and are explicit that walls retaining undrained material must account for the added hydrostatic component. The AASHTO LRFD bridge design specifications treat earth pressure the same way. FHWA's guidance on mechanically stabilized earth walls and the NCMA design manual for segmental retaining walls both treat internal drainage as a design element rather than an accessory.
A drained design therefore carries an implied condition: it is valid only while the drainage it assumed keeps functioning.
The drain is a system, not a pipe
A wall drain has several parts, and any one can be the failure point: the free-draining aggregate or geocomposite panel behind the facing, the separation geotextile that keeps fines out of it, the collector pipe at the base, the outlets that discharge, and the surface grading that keeps water out of the backfill entirely.
Failures cluster at the ends of that chain. Outlets get buried by landscaping, paved over, or crushed. Filter fabric is omitted, or specified with the wrong opening size, and blinds off as fines migrate into it. The aggregate itself rarely fails; what feeds it and what drains it usually does.
What an obstructed drain leaves behind
A drainage system that stopped working writes it down in the soil. Excavation typically finds fines infiltrated into what should be clean open-graded aggregate, precipitate cementing the voids, root intrusion at pipe joints, and a moisture boundary marking where water stood. Gradation testing of the recovered aggregate quantifies how far it drifted.
Standing water also marks the structure: efflorescence tracking out of joints, freeze-thaw damage concentrated where drainage was worst, corrosion at exposed reinforcement, and in segmental walls the loss of fines through block joints, seen as settlement immediately behind the facing.
Omission, obstruction and overwhelmed
These three look alike from outside the wall and point at different parties. A drain never installed leaves no aggregate, no pipe and no fabric where the drawings show them, which is a construction question. An installed drain that clogged raises a maintenance question, and depending on how the filter was specified, sometimes a design question too. A drain built as specified that could not handle the water actually arriving raises a design question about the assumed groundwater regime.
Those distinctions are only available while the evidence is in the ground. Once the wall is demolished and the backfill hauled away, the argument becomes a paper argument between drawings and recollections.
Water that was never supposed to be there
Some walls drain adequately as built and then acquire a water source. Irrigation installed above the crest, a downspout redirected into the backfill, a leaking service line, new paving that changed the infiltration pattern, or regrading that ponds runoff behind the wall all change the loading case as surely as a design error does. Aerial imagery, permit records and utility documentation date those changes.
Backfill material decides whether drainage can work
Drainage design assumes a particular backfill. Free-draining granular material, classified under ASTM D2487 as a clean sand or gravel, lets water reach the drain quickly. On-site clay or silt placed instead holds water and can be near enough to impermeable that a perfectly built drain never sees the water at all.
This is why classifying the recovered backfill matters as much as inspecting the drain hardware. A wall with a textbook drain and the wrong soil behind it still fails for a drainage reason.
Back-analysis and the pressure the wall saw
Stability calculations are run in reverse. Using as-found geometry, measured soil properties and the water level the physical evidence supports, factors of safety against sliding, overturning and bearing are recomputed for both the drained and the saturated condition. Where the drained case passes comfortably and the saturated case does not, the water is doing the explaining. How sensitive that result is to the assumed water level should be stated openly.
Preserving what answers the question
The instinct after a wall fails is to clear the debris, cut the slope back and rebuild. Every step of that removes the record. Photographing drainage components in place before excavation, sampling aggregate and backfill at known depths, probing the outlets, and documenting moisture conditions before the site dries are what let the drainage question be answered rather than argued.
This article is general technical orientation, not a failure analysis, an engineering opinion, or advice on any specific matter. Determining the cause of a particular incident requires hands-on examination by a credentialed expert.