Impact assessment (IA) is carried out as an ex ante process to inform decision-making. It includes requirements for engagement with stakeholders (including the public) regarding actions proposed by a proponent. A key issue with the various stakeholders involved is the perceived legitimacy of the IA, which can have implications both for the reputation of the proponent, and the likelihood of conflict over the decision. But the understanding of legitimacy in the IA literature has changed over time in line with an ontological shift from positivism (that scientifically generated information leads to better informed decisions) to the post-positivist acknowledgement of the limitations of scientific method whereby assumptions must be subject to transparency, deliberation and openness. This has led to an epistemological shift towards greater subjectivism which, we suggest, has created new opportunities (which have been realised in political decision-making) to subvert knowledge through the increased use of the Internet and social media. To address the potential for such subversion of legitimacy, we seek to conceptualise legitimacy in the IA context through framing IA around a critical realist ontology and a reliabilist virtue epistemology. This allows us to identify ‘knowledge legitimacy’ as an equally important component of IA legitimacy along with organisational legitimacy. We conceptualise knowledge legitimacy through literature review drawing on rich understandings of knowledge from IA and other fields of research in order to develop a four-dimensional typology. This includes the dimensions of: knowledge accuracy; knowledge restriction; knowledge diffusion; and knowledge spectrum. This is the first theoretically grounded attempt to understand legitimacy in IA. It is hoped that it will provoke discussion in the IA community to further advance theoretical understandings of IA and legitimacy of practice.

中文翻译：

---

影响评估中的合法性：基于认识论的概念化

影响评估 (IA) 是作为为决策提供信息的事前过程进行的。它包括与利益相关者（包括公众）就提议者提议的行动进行接触的要求。涉及的各个利益相关者的一个关键问题是 IA 的合法性，这可能对支持者的声誉和决策发生冲突的可能性都有影响。但是随着时间的推移，对 IA 文献中合法性的理解随着本体论的转变而发生了变化，从实证主义（科学产生的信息导致更明智的决定）到后实证主义承认科学方法的局限性，其中假设必须服从透明、审议和公开。这导致认识论转向更大的主观主义，我们认为，这创造了新的机会（在政治决策中已经实现），通过增加使用互联网和社交媒体来颠覆知识。为了解决这种合法性颠覆的可能性，我们试图通过围绕批判现实主义本体论和可靠性美德认识论来构建 IA 来概念化 IA 上下文中的合法性。这使我们能够将“知识合法性”确定为 IA 合法性与组织合法性同等重要的组成部分。我们利用对 IA 和其他研究领域知识的丰富理解，通过文献综述来概念化知识的合法性，以开发四维类型学。这包括以下维度： 知识准确性；知识限制；知识传播；和知识谱。这是理解 IA 合法性的第一次有理论依据的尝试。希望它能在 IA 社区引发讨论，以进一步推进对 IA 的理论理解和实践的合法性。