Theoretical reasoning (inference) is a conscious personal-level activity and a causal process. It is the process of revising one's beliefs for a reason whereby some of our beliefs cause or result in other beliefs. But inference is more than mere causation. This raises the question of what exactly distinguishes theoretical reasoning from mere causal processes. Paul Boghossian has located the distinguishing feature of inference in, what he calls, the “taking condition” requirement (TC). It turns out, however, that all attempts to explain this notion fail. In this paper, I shall offer an indirect argument for the taking condition grounded in a specific account of the structure of epistemic reasons that distinguishes two levels of epistemic normativity, namely, the possession and justification levels. I show how such an account can legitimize the role of (TC) in reasoning and deflect the standard objections raised against it. Finally, I explain how this proposal naturally accommodates the defeasibility of reasoning.

中文翻译：

---

原因、推理和采取条件

理论推理（推理）是一种有意识的个人活动和因果过程。这是由于我们的某些信念导致或导致其他信念的原因而修正一个人的信念的过程。但推论不仅仅是因果关系。这就提出了一个问题：理论推理与纯粹的因果过程究竟有何区别。Paul Boghossian 将推理的显着特征定位在他所说的“采取条件”要求（TC）中。然而，事实证明，所有解释这一概念的尝试都失败了。在本文中，我将为采取条件提供一个间接论证，该论证基于对认识论理由结构的具体说明，该结构区分了认识论规范性的两个层次，即占有和理由层次。我将展示这样的解释如何使（TC）在推理中的作用合法化，并转移针对它提出的标准反对意见。最后，我解释了这个提议如何自然地适应推理的可废止性。