Rejoinder to Hillier and Penn

After having detailed some concerns about space syntax in this issue of Environment and Planning B (Ratti, 2004), I did not anticipate the kind of response offered by Hillier and Penn (2004). I had hoped for a more academic response, stating the field of applicability of the technique, defining its limits, and suggesting future work: how it might be improved and integrated with new computer techniques for urban analysis. Something like: touchë, let's move on. After all, as I have stated, the questions under-lying space syntax are very interesting and deal with a long-lasting dilemma amongst architects and urban planners: what is the impact of built form on social life? To the contrary, Hillier and Penn's rejoinder is a strenuous and defensive rebuttal of each point highlighted in the paper. Instead of addressing the broad meaning of the criticism (dealing with the limitations arising from the use of the topological measures), they have chosen to restate most of the assumptions of space syntax in an effort to prove the technique flawless. However, this defence at all costs, if accepted, leads to paradoxical results, as explained below. 2 Absurdum primum (first section in Hillier and Penn, 2004) It is said that ` ̀ a picture is worth a thousand words.' ' I would like to illustrate the paradox of Hillier and Penn's logic with a simple image. Consider the following quiz: given the two urban configurations shown in figure 1, which of the statements below is true? (A) The ` ̀ two grids are simply not the same, and the kind of discontinuities arising from minor geometric changes... [do] have significant effects.' ' ` ̀ Behavioural evidence is also compelling.' ' Therefore, we should assume that pedestrian movement in two hypothetical cities based on the two patterns of figure 1 is likely to be radically different