What is a photomontage?

Combining photographs is older than the word for it: photographers were making composite prints in the nineteenth century. What changed in the years around the First World War was the purpose. Tate dates the technique's first use by the dadaists to 1915, in their protests against the war, and the National Galleries of Scotland credits the Berlin Dada group with popularising it as an art form. Dada itself began in Zurich; it is photomontage specifically that belongs to the Berlin circle, whose members included Hannah Hoch, Raoul Hausmann, George Grosz and John Heartfield. Britannica records that earlier artists had incorporated photographic elements into collage, and that the Dadaists were the first to make the photograph the dominant medium of the montage. Soviet artists took the technique up from the early 1920s, and Heartfield turned it into sustained political argument against the rise of fascism.

This is where the authorities do not agree with each other, and a page that states one answer flatly is picking a side without saying so. Tate defines a photomontage as a collage constructed from photographs, which makes it a subset. Britannica's definition is wider, covering images made by exposing several negatives onto one sheet, where nothing is pasted at all. The Getty vocabulary, which museums use for cataloguing, cuts it differently again: it reserves photomontage for combinations arranged into a somewhat unified image, sends work that prints two negatives as one to composite photographs, and sends less unified assemblies of pasted material to collage. Read that way the test is unity of the finished image rather than the materials used, which is also the most useful distinction for anyone looking at a building dropped into a street.

When a project team says photomontage it means a photograph of the site with the proposed building placed into it at the right position and scale. This use is not loose talk: the Landscape Institute, whose technical guidance on visual representation is the most widely cited in the field, defines photomontages as visualisations where site photography forms the basis of the image and a three dimensional model is overlaid on it, and says they are intended to represent design, form and context to a reasonable degree of objectivity, such that competent authorities and others can rely on them. The Getty vocabulary happens to anticipate this use too, naming architectural drawings among the materials a photomontage may combine. So the art term and the planning term are the same term, not two words that happen to sound alike.

An image made to sell is allowed to be flattering. An image somebody will rely on is not, and the difference is entirely in method. A montage made from a photograph whose position, height and lens were never recorded cannot be checked by anyone, however careful the work was; a slightly wrong focal length or eye height makes a building read as smaller or further away than it will be. That is why the same guidance separates montages that need no verification data from a higher tier that requires survey and recorded method, and why the honest minimum on an unverified image is a label saying it is illustrative. Which tier a given project needs is a question for the team preparing the application and the body assessing it.

• An illustrative montage: site photographs, the model, and an honest label
• A verified one: a surveyed viewpoint, a known lens and recorded method
• The difference is checkability, not picture quality
• Which is required is decided by the assessing body, not by the studio