Dogmatics must be designed in this way. Above all, every science must vigorously lay hold of its own beginning and not live in complicated relations with other sciences. If dogmatics begins by wanting to explain sinfulness or by wanting to prove its actuality, no dogmatics will come out of it, but the entire existence of dogmatics will become problematic and vague.

Note, p. 58 - The Concept of Anxiety (1844)