69 THE GOSPEL ACCORDING TO JOHN generative and traditional settings and redeployed as parts of a new literary whole; henceforth, they are known and read in terms of this collection. In this way their historically secondary context becomes their hermeneutically primary context 1. To be sure, Gamble’s final assertion is controversial: James D. G. Dunn, for one, would argue that the “final author/composition level” must retain hermeneutical priority over the “canonical level” of reading the text 2. Still in all, it suffices for present purposes to observe that, insofar as it makes any sense at all to speak of a “canonical context” (and I believe that Gamble’s argument makes that case), there is value in considering what difference it makes that we do, in fact, have a collection of books in a certain order, and not merely a random roster 3. A brief illustration of such “canonical effect” may be in order, as we look at the case of the NT books of Luke and Acts. There is little question that the author 4 intended the work as a two-part set, describing as a single narrative “the events that have been fulfilled among us” (Luke 1,1), specifically, “all that Jesus began to do and teach, until the day he was taken up” (Acts 1,1-2) and the spread of the Gospel through the work of the apostles empowered by the Holy Spirit. However, the separation of the two books by John and the inclusion of Luke with the “four-fold gospel” have served to detach Acts from this original purpose and have transformed it into the history of the early church that now links the gospels-section with the epistles-section (despite some early variation in its 1 H.Y. GAMBLE, The New Testament as Canon. Its Making and Meaning (Philadelphia, PA 1985) 73, 75. 2 J.D.G. DUNN, “Levels of Canonical Authority”, Horizons in Biblical Theology 4 (1982) 13-60. 3 The point is made particularly clear as one compares the canons of the (Jewish) Hebrew Bible and the (Protestant Christian) Old Testament. The books included are the same. But as numerous scholars have observed, surely it makes a difference in how one reads the collection as a whole whether one ends that collection with Chronicles (and the Edict of Cyrus), as in the HB, or with Malachi (and the prophesied return of Elijah), as in the OT. 4 With the preponderance of scholarship, I speak here of the one “author” of Luke and Acts. However, see now P. WALTERS, The Assumed Authorial Unity of Luke and Acts. A Reassessment of the Evidence (SNTSMS; Cam- bridge 2008), for a dissenting view based on statistical analysis of prose com- positional patterns.