In his 2006 Jesus and the Eyewitnesses, Richard Bauckham suggests:
that we need to recover the sense in which the Gospels are testimony. This does not mean that they are testimony rather than history. It means that the kind of historiography they are is testimony. An irreducible feature of testimony as a form of human utterance is that it asks to be trusted. This does not mean that it asks to be trusted uncritically, but it does mean that testimony should not be treated as credible only to the extent that it can be independently verified. There can be good reasons for trusting or distrusting a witness, but these are precisely reasons for trusting or distrusting. Trusting testimony is not an irrational act of faith that leaves critical rationality aside; it is, on the contrary, the rationally appropriate way of responding to authentic testimony. . . . It is true that a powerful trend in the modern development of critical historical philosophy and method finds trusting testimony a stumbling-block in the way of the historian’s autonomous access to truth that she or he can verify independently. But it is also a rather neglected fact that all history, like all knowledge, relies on testimony. (5; italics original)
Thus, it is perhaps not without irony that we find ourselves still under the sway of a certain kind(s) of testimony even when we seek most to avoid or to exercise our independence from testimony of some other kind(s) (cf. Gadamer, Truth and Method, 354; Lewis, “Meditation in a Toolshed,” 215).