Reconstruction

I’m struggling with the last phase of the project for my architecture class these days. For the project, we were asked to spatially translate Italo Calvino’s essays which describes some “invisible cities”.  The first step of the project was to produce several drafts of both free-hand drawings and drawings with drafting tools. All of us had read through the assigned text again and again, and had the  desk critiques after each phase to build up our thoughts. After we were settled with our final drawing, we then need to proceed to 3D and produce quick models. Having working on the translation back and forth for three weeks, everyone somehow developed a relatively complete space in their mind, and finally finished the final model and took a satisfying photograph of it. However, when our professor passed the handout of the last phase to us, we found that the next step is totally out of our expectation. According to the instruction she gave us, we are supposed to extract the line work of the photograph we took and see it as, for instance,  the top view of our model regardless of what orientation it should have been. In this step, all the distorted perspectives became merely flat contours, and the depth no long exist.  And then, we have to further construct the bottom, the front and the sides by ourselves through technical drawing. That is to say, the “city” we will define is going to be a completely different one from the one in the photograph.

Since we’ve been repeating the process of building up the mental model of our “city” for three weeks, the internal representations we have are far more than competent to recognize. Salient features like the general orientations, interwoven tunnels, shapes of small pieces and their relationship to the text are sufficiently extracted and linked together. As you know, once something is recognized, it is hard NOT to see it. Through large amount of repetitions, the channel capacity taken to turn on the internal representation is fairly low. Therefore, even when we have flattened our photograph to merely line works, our feature analyzer will recognize the depth that was there spontaneously by capturing only some distorted angles. Although during the last phase, I had got rid off the shades that usually imply the 3D geometry, I can still see the 3D object as a part of human nature to achieve speed in perception when dealing with incomplete information. Meanwhile, the fact that it is an outcome of our repeated exposure to the model we are working on otherwise proves the immediate difference I felt when I tried to re-define others’ model. Since both the original text and our translation are highly abstract, classmates are unlikely to know others’ model so well without collecting enough salient features. In this case, we will not have the image of the model popped up in their mind upon seeing the flat line work, which allows us to convert the perception with ease.

I found that building new internal representations of the same object over the old one is such a hard and frustrating task, but I think this kind of practice can help us view the world from a new perspective before the quick guess determines our sight.