For being such a short work, the book of Joel contains a host of problematic issues that have troubled its interpreters. One of the most significant is the historical/literary relationship between the first two chapters. The question is, is the locust plague of chapter one a literal infestation of locust portending the invasion of chapter two? Or, does the imagery portray the devastation left in the wake of an enemy army throughout both chapters? Indeed, answers do not come easy.
The latest issue of the Journal of Biblical Literature contains an article by Ronald L. Troxel that addresses the question once again. I have not yet had an opportunity to read the article, but the abstract does sound intriguing:
Scholarship on the book of Joel has long been vexed by the wayyiqtol verb forms in 2:18-19a. Ibn Ezra suggested that they are analogous to the prophetic perfect, expressing certainty about the outcome, while Adalbert Merx suggested that they should be read as simple wāw + jussive, and Julius Bewer argued that the imperative forms in 2:15-16 should be read as simple qatal forms, enabling vv. 15-17 to be read as a report of the people’s response to the exhortation of vv. 12-14. More recent studies of Joel 2 have found it difficult to explain the interchange of qatal and yiqtolverbs in vv. 2-11. Some have explained these as signaling the intrusion of redactional materials, while others have sought to accommodate them under a tense or aspectual understanding of the verbal system. Still others have despaired of finding a solution and have adopted readings of the verbs based solely on the context. Both of these problems are, however, amenable to rather straightforward solutions. On the one hand, the wayyiqtol verbs of 2:18-19a come into focus once we recognize the narrative structure of the book. The wayyiqtol verbs are embedded in speech by the narrator, whose voice was last heard in 1:4. On the other hand, the qatal and yiqtol verbs in 2:3-11 follow typical morphosyntactic and morphosemantic patterns, once we take into account their discourse settings, particularly the pragmatics of their clauses.
For those with access to JBL, you can find the article here.