8Houses which are set level with the ground will no doubt last to a great age, if their foundations are laid in the manner which we have explained in the earlier books, with regard to city walls and theatres. But if underground rooms and vaults are intended, their foundations ought to be thicker than the walls which are to be constructed in the upper part of the house, and the walls, piers, and columns of the latter should be set perpendicularly over the middle of the foundation walls below, so that they may have solid bearing; for if the load of the walls or columns rests on the middle of spans, they can have no permanent durability.
2It will also do no harm to insert posts between lintels and sills where there are piers or antae; for where the lintels and beams have received the load of the walls, they may sag in the middle, and gradually undermine and destroy the walls. But when there are posts set up underneath and wedged in there, they prevent the beams from settling and injuring such walls.
3We must also manage to discharge the load of the walls by means of archings composed of voussoirs with joints radiating to the centre. For when arches with voussoirs are sprung from the ends of beams, or from the bearings of lintels, in the first place they will discharge the load and the wood will not sag; secondly, if in course of time the wood becomes at all defective, it can easily be replaced without the construction of shoring.
4Likewise in houses where piers are used in the construction, when there are arches composed of voussoirs with joints radiating to the centre, the outermost piers at these points must be made broader than the others, so that they may have the strength to resist when the wedges, under the pressure of the load of the walls, begin to press along their joints towards the centre, and thus to thrust out the abutments. Hence, if the piers at the ends are of large dimensions, they will hold the voussoirs together, and make such works durable.
5Having taken heed in these matters to see that proper attention is paid to them, we must also be equally careful that all walls are perfectly vertical, and that they do not lean forward anywhere. Particular pains, too, must be taken with substructures, for here an endless amount of harm is usually done by the earth used as filling. This cannot always remain of the same weight that it usually has in summer, but in winter time it increases in weight and bulk by taking up a great deal of rain water, and then it bursts its enclosing walls and thrusts them out.
6The following means must be taken to provide against such a defect. First, let the walls be given a thickness proportionate to the amount of filling; secondly, build counterforts or buttresses at the same time as the wall, on the outer side, at distances from each other equivalent to what is to be the height of the substructure and with the thickness of the substructure. At the bottom let them run out to a distance corresponding to the thickness that has been determined for the substructure, and then gradually diminish in extent so that at the surface their projection is equal to the thickness of the wall of the building.
7Furthermore, inside, to meet the mass of earth, there should be saw-shaped constructions attached to the wall, the single teeth extending from the wall for a distance equivalent to what is to be the height of the substructure, and the teeth being constructed with the same thickness as the wall. Then at the outermost angles take a distance inwards, from the inside of the angle, equal to the height of the substructure, and mark it off on each side; from these marks build up a diagonal structure and from the middle of it a second, joined on to the angle of the wall. With this arrangement, the teeth and diagonal structures will not allow the filling to thrust with all its force against the wall, but will check and distribute the pressure.
8I have now shown how buildings can be constructed without defects, and the way to take precautions against the occurrence of them. As for replacing tiles, roof timbers, and rafters, we need not be so particular about them as about the parts just mentioned, because they can easily be replaced, however defective they may become. Hence, I have shown by what methods the parts which are not considered solid can be rendered durable, and how they are constructed.
9As for the kind of material to be used, this does not depend upon the architect, for the reason that all kinds of materials are not found in all places alike, as has been shown in the first book. Besides, it depends on the owner whether he desires to build in brick, or rubble work, or dimension stone. Consequently the question of approving any work may be considered under three heads: that is, delicacy of workmanship, sumptuousness, and design. When it appears that a work has been carried out sumptuously, the owner will be the person to be praised for the great outlay which he has authorized; when delicately, the master workman will be approved for his execution; but when proportions and symmetry lend it an imposing effect, then the glory of it will belong to the architect.
10Such results, however, may very well be brought about when he allows himself to take the advice both of workmen and of laymen. In fact, all kinds of men, and not merely architects, can recognize a good piece of work, but between laymen and the latter there is this difference, that the layman cannot tell what it is to be like without seeing it finished, whereas the architect, as soon as he has formed the conception, and before he begins the work, has a definite idea of the beauty, the convenience, and the propriety that will distinguish it.
I have now described as clearly as I could what I thought necessary for private houses, and how to build them. In the following book I shall treat of the kinds of polished finish employed to make them elegant, and durable without defects to a great age.