Nothing is true,
everything is an index

The true friction angle does not exist. The true moisture content doesn't either. In soil mechanics, nothing is true. Everything is an index.

In the previous episodes, we met the grains. Gravel, sand, silt and clay. A thousand times the difference between the finest gravel and the coarsest clay. Ten thousand times between the extremes.

And the grain size distribution curve. The soil's identity card. As we saw in episode three, the grains are never alone: between them there is water and air.

Now imagine you could separate a piece of soil into its three phases. All the grains together, at the bottom, with no void between them, all the water together in the middle and all the air together, on top. Solid, liquid and gas: the three physical states of matter, together in the same piece of soil. This is the three-phase model. It does not exist in nature. It is a way of looking at soil.

On the left side, the volumes. The total volume, the volume of solids and the volume of voids, which splits into volume of water and volume of air.

On the right side, the weights. The total weight, the weight of solids, the weight of water. And the weight of the air? Air takes up space, but it barely weighs anything. Next to the others, we neglect it.

Everything from here on is a ratio. Weight divided by volume. Volume divided by volume. Weight divided by weight. Each of these ratios gets a name and becomes a physical index.

And here comes the line that gives this episode its name. In soil mechanics, every parameter is only an index. No value in this episode, in all the others, in the whole geotechnical profession, is real.

However real a number may look, it is an index.

This looks like a flaw. It is a choice. An index does not aim to be reality. It aims to be comparable. If every laboratory measures moisture the same way, in the same oven, at the same agreed temperature, then I can compare the soil from a site in Brazil with the soil from a site in Indonesia, in Ghana or in Japan. Nothing means anything on its own. But relatively, it does.

The indices are organized into three groups. The group of unit weights: the natural one, the water's, the grains', and a certain dry unit weight, which will come up a lot in this series.

The group of voids: porosity and void ratio. And the group of water: moisture content and degree of saturation.

That is a lot of indices, but here comes the surprise. Of all of them, only three can be determined directly in the laboratory. Only three have a test of their own.

The first is the natural unit weight of the soil. It is the most demanding of all. It requires an undisturbed sample, with its moisture preserved.

The sample has to leave the field exactly as it was in nature. It cannot lose water by evaporation. It cannot gain water. It cannot be squeezed or shrink in volume. In the laboratory, it goes to the hydrostatic balance.

The second is the moisture content. This one is less demanding. It only asks that the moisture be preserved.

You can dig with a shovel, a pickaxe or an auger, close it in an airtight container, a well-sealed bag or a tightly capped aluminium can. And take it. Disturbed? It doesn't matter. Here it is all about weights.

The third is the unit weight of the solid grains, measured in a flask called a pycnometer. This one accepts a disturbed sample and does not even ask for protection against moisture loss. Only the solids matter.

And all the other indices come from mathematical relations.

And to derive those relations, there is an elegant trick. The move is to go back to the three-phase model and call the volume of solids 1. One unit. One cubic metre, one cubic centimetre. It doesn't matter.

The moment the volume of solids equals 1, the volume of voids becomes the void ratio itself. The weight of solids becomes the unit weight of the grains. The weight of water becomes that value multiplied by the moisture content.

The whole model switches language. Where there were weights and volumes, now there are indices. And from it come all the relations between all the physical indices.

I measure three in the laboratory. The natural unit weight, the moisture content and the unit weight of the grains. I calculate the rest on paper.

In practice, it is always like this. These three I measure, the others I calculate.

The four indices of this model, one at a time. The void ratio: the volume of voids divided by the volume of solids.

Porosity: the same voids, now divided by the total volume. The degree of saturation: the part of the voids filled with water.

And the moisture content: the weight of water divided by the weight of solids. The only one of the four made with weights.

So here is the question that holds all of this up. The most demanding index asks for an undisturbed sample. A piece of soil taken from nature without disturbing anything, not the voids, not the water. How do you take a piece of the ground without spoiling exactly what you want to measure? That is the story of the next episode.