A recurrent theme in software engineering is that it isn't soft and there is no engineering in it. This may be a little cruel, but it makes a perfect T-shirt. In truth, one of the principle defining characteristics of successful 'engineering' is that by analysing why it screws up, it learns from its mistakes. It is abundantly clear that apart from isolated pockets of resistance, software 'engineering' does not. We are used to the mantra 'Lessons must be learned' coming from political descriptions of failed systems and procedures, only to be ignored later, but it is harder to understand why something nominally technical such as software development demonstrably does not learn from its mistakes either. Using a number of case studies, I will try to understand whether this is pervasive throughout the software development hierarchy or is confined to suits with spreadsheets. These may be strong words but will resonate with anybody who has read the background behind the Boeing 737 MAX debacle.