Urgent Need to Correct Geotechnical Theory

The theoretical arguments I put forward in the six-part series the Water in the Soil deal with the most vulnerable of soils types: water-saturated non-cohesive deposits such as silts, sands, and gravels below the water table. Some of these arguments were first introduced in the monograph Liquefiable Materials and their Treatment by Vibro-Drainage which was published in 1998.

My approach is simple: To combine the principles of Soil Mechanics with those of Fluid Mechanics, disciplines which both belong to Civil Engineering. This reunion of those disciplines then allowed the void water entrapped within the soil-structure the benefit of being appraisable under the long-established rules of hydrodynamics.

With the aid of hydrodynamics we are lead to the uncomfortable realization that current geotechnical theory has gone astray and is in urgent need of revision. The stark fact is that at the very heart of our geotechnical design philosophy we have confused cause and effect. The prevailing belief is that increases in pore water pressure lead to destabilization of the soil-structure, whereas the truth is the other way around: It is in fact soil-structure deformation that cause pressure responses in the pore water.

Some of the more important implications of this reversal are:


Reliance on vertical drains as a means of ground protection against earthquake waves is wrongheaded simply because pore water pressures do not cause soil-structure collapse, they are a consequence of prior failure. Vertical drains, by facilitating drainage, will only allow faster ground settlements after liquefaction has occurred.

Only very loose sands can liquefy, well graded or coarse granular masses cannot.

Stability Analyses

The way in which excess (above hydrostatic) pore water pressures are incorporated in current methods of stability analyses is somewhat confused. Far better to apply the water pressures in the form of Seepage Forces and to use the phreatic surface only to determine buoyant forces. Transient hydraulic gradients due to shear straining within the failure zone should also be taken into account.

Observational Method

Monitoring of piezometric elevations, as is common in earthdams and similar earthworks, does not warn of approaching distress. Changing pore water pressures only tell you that shear straining is under way. This fact was dramatically illustrated by the 265 piezometers at Tarbela dam.

One of the potential design benefits attending this otherwise bleak picture of where geotechnical engineering now stands is that it is clearly possible, after some further work by the next generation of young people, to develop a real-world relationship between soil shear strain and pore pressure levels temporarily prevailing throughout the soil strata, such that one can be derived from the other. Then the triaxial machine which lead us into this embarrassing situation in its undrained mode of operation, may be the tool, in it drained mode, to now redeem itself by paving the way for our honourable extraction.