The theophanic vision report in Habakkuk 3* contains variations among verb forms that seem to defy explanation. For some scholars the solution is simply to ignore the variation. For example, note the treatment of the verb tense-aspect-mood (TAM) in some major English versions/ translations of the vision report of 3:3–15 (comprising 32 verb forms in all): NRSV, NIV, and NKJV translate all the verbs with past forms; while the REB, NAB and JB use mostly present verbs; and the NASB and NLT both show a split between present verbs in verses 3–7 as past verbs in verses 8–15. Recently de Regt (2008: 92) argued for a future temporal reference for the verbs vision based on treating the Perfects as “prophetic perfects.”
I have had several occasions to visit and revisit this passage recently: Calvin Park kindly called my attention to the verb issue some months back; at the same time I was teaching an advanced exegesis class on several minor prophets, including Habakkuk, though sadly we did not reach this chapter; and most recently, in preparing the entry on “Hebrew language” for the IVP Dictionary of the Old Testament: Prophetic Books I found in this poem a good illustration of the challenges of interpreting temporal reference of the verb forms in the prophetic books. So I took the later occasion to finally sit down and draft my own analysis of the poem.
My basic approach is as follows. First, I assume a certain static quality to poetic visionary reports, which immediately makes some sense out of the variety of verb forms inasmuch as the temporal reference of the forms intersect in and around the sphere of present temporal reference (e.g., think of their appearance in proverbial sayings or present temporal reference use of the Perfect such as in performative expressions and combined with stative predicates). Second, I admit up front that it is unlikely we can be equally, independently certain regarding temporal interpretation of each verb in the poem. Therefore, I allow those forms about which I am more certain (i.e., those with less temporal ambiguity) to provide guidance for my temporal interpretation of other verbal temporal reference within the poem. Finally, the central question in my mind is whether an explanation for the variation of verbs is possible. That is, the goal is not simply to explain what the temporal reference is (I’m already assuming a dominant present time reference because of it being a vision report and the promiscuity of Perfects and Imperfect forms), but to seek a plausible explanation for why this particular form in a given instance rather than another. Although I would never advocate replacing a semantic analysis of the verb forms for a strictly discourse-pragmatic one, given a semantic theory as a starting point (and I am presuming the basic correctness of the semantic theory developed in Cook 2001, 2002, etc.), we might profitably ask what discourse-pragmatic functions these verb forms might fulfill in a poem in which temporal reference appears to be mostly moot.
While most commentators devote their attention to the switch of persons between vv. 1–7 (third-person) and vv. 8–15 (second-person address to God), I’m inclined to organize the poem based on which verb forms are dominant: stative Perfects (which default for present temporal reference) and Imperfects on the one hand, and dynamic Perfects and Past Narratives (wayyiqtol) on the other. This approach results in five verse paragraphs (or strophes if you like):
vv. 3–5 – framed by Imperfect forms with two intervening Perfect forms, the second which is a stative
vv. 6–7 – dynamic Perfect and Past Narrative forms with the exception of the Imperfect in v. 7b
vv. 8–9 – Imperfect forms and a stative Perfect in v. 8a
vv. 10–11 – dynamic Perfect forms with subordinated Imperfect clause (unmarked relative) in v. 11
(I’m tempted to emend the Imperfect in v. 10a to a Past Narrative on the basis of the pattern of Perfect-Past Narrative in v. 6)
v. 12 – two Imperfect forms in v. 12
vv. 13–15 – dynamic Perfect forms and a subordinate Imperfect (unmarked relative) in v. 14
The dominance of a particular verb form I take as indicative of how to understand the other, more ambiguous or incongruous forms in each paragraph. Thus, כִּסָּה in v. 3 has a (present) perfect sense ‘has covered’, in as close as possible keeping with the present temporal reference of the Imperfect forms and the stative Perfect מָלְאָה is full’.** Similarly, in other instances where a Imperfect and Perfect are juxtaposed, I’m inclined to interpret them as temporally similar as well: ‘the mountains have seen you (רָאוּךָ), they are writhing (יָחִילוּ)’ (v. 10a); ‘the moon has stood (עָמַד) in its lofty abode—at the light of your arrows (which) are coming (יְהַלֵּכוּ)’ (v. 11).
Verse 6 is perhaps the most awkward inasmuch as it will not admit a continuance of the present temporal reference. It begins with two narrative sequences of Perfect-Past Narrative: ‘He stood (עָמַד) and the earth quaked (וַיְמֹדֶד), he looked (רָאָה) and the nations jumped/shook (וַיַּתֵּר) and the ancient mountains shuddered (וַיִּתְפֹּצְצוּ)’. My explanation for these and other uses of the Past Narrative in poetry (made in an conference paper several years ago) is rather nuanced (read, apt to be misunderstood). First, the Past Narrative, as its name implies, is “native” to prose narrative, where it denotes the (relatively) most salient events in a narrative rather than (as often argued) temporal succession; the latter is a property of narrative discourse and not a semantic or discourse-pragmatic component of the Past Narrative verb. That is, narrative events are by definition iconically ordered, understood as occurring in the order they are reported unless otherwise indicated (see Cook 2004 for the complete argument). Second, Hebrew poetry, with its constitutive element of parallel lines, represents the inverse of this iconic narrative ordering: events in parallel are not understood as iconically ordered separate events, but as referring to the same event. Therefore, and third, I would argue that Past Narrative in poetry is employed for its discourse-pragmatic implicature of temporal succession that it brings from its narrative use, in order to override the non-narrative parallel structure of poetry and denote events as in iconically ordered. This function of the Past Narrative in poetry is most extensively seen in the narrative poems (Ps 18, 78, 105–107), which account for the majority of Past Narrative forms in Psalms (57%; 96 psalms lack any Past Narrative form). However, the form also constitutes briefer narrative sequences (only two iconically ordered events are required to constitute a “narrative”): e.g., Ps 7:13b קַשְׁתּוֹ דָרַךְ וַיְכוֹנְנֶהָ׃ his bow he bent and aimed it’. Thus, in Hab 3:6a there is a break in the static description of the vision by the the introduction of two brief narrative sequences that describe the reaction of the earth and nations to God’s actions. (The two Past Narratives following רָאָה should be understood analogously to a “compound” semichut/construct consisting of a single nismach/construct item in relation to two somech/absolute items: the nations shook and the everlasting mountains shuddered concurrently, both in succession (reaction) to Yhwh’s look.)
While these comments explicate the variation of verbs, they do not explain the significance of the alternation between Imperfects/stative Perfects and dynamic Perfects/Past Narratives in this poem. What strikes me about the alternation is its correlation with the subject matter:
vv. 3–5 describe Yhwh’s appearance
vv. 6–7 describe creation’s reaction
vv. 8–9 return to a description of Yhwh’s approach
vv. 10–11 return to the reaction of creation to Yhwh’s appearance
v. 12 strikingly combines the focii of the preceding alternating sections by describing Yhwh’s actions on earth: marching and threshing
vv. 13–15 breaks off the alternation, providing a reflective comment on why Yhwh has come—for the salvation of his people and to crush the wicked
A plausible explanation for this correlation relates to the immediacy of the theophany from the prophet’s perspective: those portions dominated by Imperfects/stative Perfects portray Yhwh’s approach in vibrant immediacy:
‘Eloah from Teman is coming (יָבוֹא) and the Holy one from Mount Paran . . .
Before him is going (יֵלֶךְ) pestilence, and plague is coming (וְיֵצֵא) forth at his feet’ (vv. 3, 5)
By contrast, the prophet distances himself from the reaction of the creation and nations in those sections dominated by stative Perfect/Past Narrative forms. This is especially evident in vv. 6–7 where the Past Narratives clearly cast the description with a past temporal reference, and the first-person verb רָאִיתִי ‘I saw’ in v. 7a creates a self-conscious distance between the vision and the prophet seeing it. (The use of past temporal reference to denote more metaphoric “distancing” of events from the speaker’s “present” is widespread in the world’s languages: e.g., English counterfactual expressions such as If he were here I would tell him.)
It seems plausible that the interpretative effect of this alternation is to align the prophet with Yhwh in his appearance while distancing him from the havoc that Yhwh’s appearance creates in creation and the nations/wicked. That is, it places the prophet apart from those entities that bear the brunt of Yhwh’s theophanic fury and therefore aligns him with the people for whom Yhwh’s arrival signals salvation (v. 13a יָצָאתָ לְיֵשַׁע עַמֶּךָ ‘you have come forth for the salvation of your people’). This is more than merely aesthetic or even rhetorical flourish given the central importance (and difficulty; see Sweeney 1991) in identifying who precisely are the righteous and who are the wicked in the book—and what is their ultimate destiny.
*My scansion (I follow Andersen  with regard to the scansion of v. 11) and gloss translation of the poem I developed simply to facilitate my analysis and discussion.
**Qal מלא is transitive only about a third of the time (29/101x), making it likely that הָאָרֶץ is subject here with a fronted object and inverted VS word order as opposed to וּתְהִלָּתוֹ. This intransitive sense ‘be full of something’ accounts for about half of the verb’s occurrences [49/101x] and is in keeping with the other 13 instances where the feminine singular Perfect appears with אֶרֶץ, which is best construed as subject in each instance, though some cases are more certain than others (Gen 6:13; Lev 19:29; Is 11:9; Jer 23:10; 46:12; 51:5; Ezek 7:23; 9:9; Hab 3:3; Psa 33:5; 48:11; 104:24; 119:64).
Works cited (also see here)
Andersen, Francis I.
2001 Habakkuk: A New Translation with Introduction and Commentary. The Anchor Bible 25. New York: Doubleday.
Cook, John A.
2001 The Hebrew Verb: A Grammaticalization Approach. Zeitschrift für Althebraistik 14/2: 117–43. (Download PDF)
2002 The Biblical Hebrew Verbal System: A Grammaticalization Approach. Ph.D. diss., University of Wisconsin, Madison. (Download PDF)
2004 The Semantics of Verbal Pragmatics: Clarifying the Roles of Wayyiqtol and Weqatal in Biblical Hebrew Prose. Journal of Semitic Studies 49/2: 247–73. (Download PDF)
Regt, Lénart J. de
2008 Hebrew Verb Forms in Prose and in Some Poetic and Prophetic Passages: Aspect, Sequentiality, Mood and Cognitive Proximity. Journal of Northwest Semitic Languages 34/1: 75–103.
Sweeney, Marvin A.
1991 Structure, Genre, and Intent in the Book of Habakkuk. Vetus Testamentum 41/1: 63–83.