Well, why not? If I break it down, there is still force being moved along a curve.

Carefully and deliberately, she sketched in three more semicircular arches of ascending size within the right hand spandrel, did a very quick set of calculations in her head, and called to the three Masters, who were discussing something else. “I think I have a flood solution,” she said.

They hurried over to her, and once they saw her sketch, all four of them went to work.

She didn’t go to bed until late that night, and not for several nights thereafter, but when the marathon was over and some small adjustments made, they were all satisfied. The arches within the spandrel would work, and work magnificently. Not only that, but they reduced the weight of the bridge by more than fifteen percent!

Plank by plank, the wooden version of the bridge took shape, fitted together like a puzzle, and every bit of it pegged and glued until it formed a single, solid structure. This was because it was going to be taken apart—cut apart, in fact, and each separate piece used as the template for a corresponding stone block or slab, creating a giant puzzle in stone that would be reassembled on the banks of the river. The main arch would eventually be made of twenty-eight “relatively” thin slabs of limestone, each long enough for two big wains and a balustrade. On top of that would come the three spandrel-arches on each side of the bridge, and on top of those would come the rest of the spandrel and the roadbed. The roadbed itself would have an incline of no more than a few degrees.

Meanwhile, Rudi and Master Ketnar created the plans for the scaffolding to support the work. Because the scaffolding would only have to last as long as the construction went on, they could use the existing piers to support it, and during a dry and clement spot of weather, work actually began laying stone on the old piers while the water was low. This work to support the wooden scaffolding got laid in long before the model itself was more than a quarter built.