Chapter 2 discusses Kant’s account of the ‘logical use of reason,’ which consists in drawing ‘mediate inferences’ (e.g., syllogisms). It is guided by the ‘Logical Maxim,’ which requires us to search for the conditions of or conditioned cognitions until we arrive at unconditioned cognitions. Its aim is to transform our manifold cognitions about nature (both empirical and a priori) into a complete system of scientific knowledge and thus to achieve the ‘unity of reason.’ The chapter also places Kant’s conception of systematic cognition in the context of the Wolffian school and explains what it means for cognitions to be conditioned or unconditioned. Finally, the Logical Maxim is defended as a plausible expression of our rational striving for systematicity and completeness of knowledge.

A central aspect of Kant’s Transcendental Dialectic is his claim that there are concepts – the transcendental ideas – that necessarily arise from rational reflection. According to Kant, these ideas of reason, like the categories of the understanding, form an a priori system. In Chapter 6, we first look at Kant’s conception of transcendental ideas and survey the system they form. Next, it is argued that Kant does not attempt to derive the transcendental ideas in questionable ways from the forms of rational inferences or the possible relations between subject, object, and representation (even though the text suggests this), but rather considers them, much more plausibly, as concepts we arrive at through rational inferences about specific (psychological, cosmological, and theological) subject matters. The central philosophical point here is that concepts can be the result of (what Kant calls) ‘necessary inferences of reason.’ A first instance of this is Kant’s derivation of the concept of the unconditioned; the chapter then turns to the three classes of transcendental ideas (psychological, cosmological, theological).