A significant number of mentally ill defendants pass through the courts, where clinicians advise upon their ability to fairly stand trial. This is codified in the assessment of 'fitness to plead'. In England and Wales, the narrow 'Pritchard' criteria that determine fitness date from 1836. The authors undertook a piece of qualitative research with the aim of determining whether the 'Pritchard' test remains fit for purpose. A semi-structured interview was devised to study the experiences of senior criminal barristers. Framework analysis was adopted in order to identify relevant themes within the data. Five key themes emerged, helping to define the true construct of 'fitness' and illustrating serious procedural problems with the current approach. A range of practical solutions emerged. It was clear from the results that formal findings of unfitness are extremely rare. Many mentally ill defendants are missed or stand trial unfairly. The Pritchard criteria do not protect their best interests. A possible reformulation of the approach to assessing fitness is presented for discussion.