What the finished version hides

Text One: An assistant's account of reconstructing a broken sign for a local exhibition.

Fourteen painted fragments of an old hall sign were brought to our workroom. I began as if I were completing a jigsaw, looking for anything that made a continuous word or stripe. Three pieces formed a secure group: their broken edges interlocked, and a narrow red line crossed both joins without changing direction. Other matches were less certain. Two pieces carried the same blue paint but did not touch, while a triangular piece could occupy either of two gaps.

An inventory described a 'refreshment board', so it was tempting to arrange the surviving letters as part of the word 'refreshments'. An angled photograph also showed a long word in the correct area. However, a reflection concealed its final letters. We recorded the suggested arrangement but did not treat the expected word as proof of every letter's position.

For the trial display, we mounted the fragments on a blue backing. From several steps away, the backing blended with the original paint and six of the eight visitors thought that they were seeing an almost complete sign. We replaced it with a grey backing that made every missing section visible. Beside the firmly joined group, we marked the two possible positions for the triangular piece. The display looked less finished, but it showed where the evidence ended and our choices began.

Text Two: A report on the reconstruction of a painted theatre backdrop.

A project group used surviving pieces of cloth, two photographs and a rehearsal note to recreate a lost theatre backdrop. The photographs agreed that a stone arch occupied the centre, but performers concealed the lower part of the scene. The note placed a silver moon 'beside the tower' without identifying which side.

The group prepared three reconstructions. Each included four placements supported by both the cloth and the written or photographic records. Other placements depended on a single clue or were added to produce a complete-looking scene.

Visitors often described Reconstruction A as 'finished', while several said that the gaps in Reconstruction B made the surviving evidence easier to recognise. Reconstruction C tested alternative positions for details such as the tower and moon. The project group did not use the visitors' choices as proof that A was historically accurate. Instead, the final exhibition presented B at full size and displayed A and C as smaller alternatives.

A programme discovered later stated that the moon rose on the audience's left. This allowed the group to reject some positions, although it did not establish the moon's exact height. The exhibition was amended so that visitors could see how new evidence had narrowed, rather than completely settled, the reconstruction.

Synthesise what you learn from both texts about how incomplete objects can be reconstructed and how those reconstructions may influence an audience.