In August 2014, a special issue of Safety Science contested the foundations of safety science as a scientific domain on methodological, theoretical and philosophical grounds. Safety specialists in social, behavioural and organisational sciences discussed what seems to be an identity crisis in occupational safety and social sciences. A tension between scientific and societal relevance of the notion of safety was noted, raising doubts about the scientific validity of safety science (Safety Science, 2014). As three of the founding fathers of the Safety Science Group (SSG) at Delft University of Technology (DUT) in 1978, we were challenged by this debate to reflect on our ambitions and mission that were expressed during the inception of safety science at DUT. In addition to what has already been described by Hale and De Kroes (1997) and Hale (2014), we elaborate also on personal experiences and insights, gained over a period of some 40 years of work. These additional experiences were gained in particular by the first two authors in the transport domain from a technological and engineering design methodological perspective. They provide a more encompassing scope on the development of safety science in general as a scientific activity at DUT. Parallel developments in other domains have been described in other papers (Hale, 1985, 2006, 2014; Hale and de Kroes, 1997). First, we have to correct a long term omission by translating the founding documents for safety science at DUT into English to make them accessible for non-Dutch experts. Second, we highlight the development of three basic notions that were identified in 1978, at the inception of the SSG, as the cornerstones for safety science as a scientific discipline, no matter what domain it is related to: interdisciplinarity, problem-solving orientation and systems approach. We document in this paper their development and use in the transport domain in the DUT as a whole, i.e. more broadly than the SSG. Third, we discuss more in general the observations, particularly of the first two authors on the value and unique role of safety investigation methodology and systems engineering as powerful feedback loops with learning potential and drivers of change through their potential to address safety in complex, dynamic and open transport systems. Fourth, we elaborate on three additional basic notions that are needed as extra building blocks for a paradigm shift in safety thinking, irrespective of disciplines and domains – a full information supply, engineering design methodology and multiple intervention strategies. Finally, we advocate a mutual recognition of the value and validity of scientific paradigms as developed in the various disciplines, that in conjunction constitute safety science as a distinct, multidisciplinary activity in the academic community. @en