ABSTRACT
In order to know the past we rely on some form of document, using the term in its widest sense: on a treaty, an account book, a building, painting, photograph, a surviving eye-witness and of course on earlier histories which themselves depended on some documentary evidence. In the case of architecture we have to rely heavily, though not solely, on visual evidence. Manuscripts of Vitruvius have come down to us with no illustrations except for one diagram in the margin though Vitruvius refers to illustrations which should be at the end of several
considerable insight into Roman architecture, much of it however dependent on the survival of built Roman remains.
