May be the addition of Directive ה◌̞
I do not believe the passage needs to be viewed as two distinct lexemes.
The vowel pointing with the ה◌̞ suffix matches the form of what ה takes for its known directive uses, or the ה locale. Ronald J. Williams in Williams' Hebrew Syntax, 3rd ed. (Toronto: University of Toronto Press, 2007) notes of this suffix:
The directive suffix ה◌̞ was once thought (e.g., GKC §90a, c) to be the old accusative case ending that had dropped off elsewhere (§31 [Williams' section]). This view has been discarded now that materials in Ugaritic have been discovered and deciphered; Ugaritic is very close to Hebrew and has both the case endings (which Hebrew dropped) and a separate directive suffix (spelled 'h' in Ugaritic transcription). (p.25 §61)
The most likely candidate from Williams' subcategories appears to be:
Separative ה◌̞. The directive suffix ה◌̞ on a noun that is prefixed with the preposition מִן 'from' can indicate the direction away from the noun (p.25 §64a).
Here, the preposition is prefixed to the preceding construct word "from the belly" (מִמְּעֵ֖י) and the suffix attached to the absolute word "fish." Williams does not note if the ה locale ever happens in a construct chain with a preposition affixed, but Gerhards' quote (see below) notes the ה locale does appear "rarely as the nomen rectum of a construct relationship," giving as examples Josh 15:5 (found twice: וּגְב֥וּל קֵ֙דְמָה֙ [and the border toward the east], לִפְאַ֤ת צָפֹ֙ונָה֙ [on the side toward the north]; note the 2nd use is an example of a preposition [לִ] affixed to the construct of the chain, with the ה locale affixed to the absolute, exactly as I am proposing the reading here in Jonah to be) and also Josh 18:15 (found twice: וּפְאַת־נֶ֕גְבָּה [and the side toward the south], הַגְּבוּל֙ יָ֔מָּה [the border toward the west]).
The combination of the מִן preposition (prefixed or not) with the directive ה is what shifts the location of the he as being the source point rather than the towards point (which is more typical of a straight use of the directive ה). Some examples with מִן are Josh 10:36, 15:10; Isa 45:6; Jer 1:13, 27:16.
Certainly 2:1 [English] is referring to the belly as the "place" where Jonah is while making his prayer, so it is not only logically possible, it seems to me highly probable the idea is simply emphasizing Jonah prays "out from the belly of the fish." This is opposed to praying a prayer from his own belly, i.e. a heart-felt prayer as ambiguity could have arose in interpreting מְעֵי if דָּג were omitted—but that distinction probably could have been made without the directive ה added to דָּג. Why then add it? At least two reasons seem plausible to me (and both may be valid at the same time):
- To create a "container" for the construct chain so that "the belly of the fish" is viewed as a single unit being modified by the מִן, the directive ה making sure דָּג is included within the "from" statement, so that it is not just "from the belly," but specifically "from the fish's belly."
- Stylistic reasons of rhythm (more along the lines of what Joseph's answer argues).
A Couple Others Have Noted or Proposed Likewise
My argument above was originally made purely from the grammatical indicators, as I had not found others who posed the possibility, but have since found some support, and incorporated some insights from that into the above argument. The very few sources found so far that mention this as possible are noted here.
Non-scholarly
This discussion board post, though there the directive is taken as being swallowed into the fish, i.e. praying as he is being swallowed, with the idea "toward the belly of the fish" (that does not seem as logical to me as the solution I give above).
Scholarly
The entry for Jonah on academia-bible.com (The Scholarly Bible Portal of the German Bible Society) has a note about this possibility. I quote the German and then a translation:
Eine weitere mögliche, aber sicher auch nicht unproblematische
Erklärung bietet die Annahme, dass in Jon 2,2 gar kein Femininum
vorliegt, sondern eine um ה locale erweiterte Form (vgl. Gerhards,
2006, 52). [accessed 6-1-2015]
A further possibility, although certainly not without problems of its own, is the assumption that Jonah 2:2 does not have a feminine form, but rather a form augmented by ה locale (cf. Gerhards 2006, 52).
The work being referenced is M. Gerhards, Studien zum Jonabuch (BThSt 78; Neukirchen-Vluyn, 2006). The proposal appears on pages 51-52 (italics in original; bold emphasis added by me1):
In 2:2 there is another striking linguistic variation: the fish, which in 2:1, 11 is called דָּג, is here called דָּגָה. Instead of the masculine, an apparently feminine form is used. What in Jewish reception gave occasion for explanatory elaboration,137 was analysed in literary-critical terms by many of those holding to the unoriginality of the psalm of Jonah, in order to demonstrate that the introduction of the prayer is also literarily secondary.138 But if one nevertheless has substantial grounds for considering the psalm as an original component of the narrative, its introduction cannot be eliminated. There is another possible explanation of this difficult form; it is also a preferable explanation, since literary criticism cannot resolve the problems with the form דָּגָה, should the feminine actually be intended. In the Old Testament, the feminine form is always used as a collective.139 Since the feminine in its ordinary usage has no place in this text, it is likely that [bietet es sich an] the ה- in דָּגָה should not be construed as a feminine ending, but rather as a ה locale. 2:2 would thus be translated: ‘Jonah prayed to Yahweh, his God, from the innards in the fish’.140 Then there is no problem for the form דָּגָה to be written by the same author as that of 2:1, 11. The local form can thus have its motivation in reinforcing the inhospitable setting of the prayer.
137 Cf. on this Steffen, Jona-Geschichte, 28f.
138 Cf. Böhme, ZAW 7 (1887), 233f.; L. Schmidt, De Deo, 56; Weimar, Jonapsalm, 48, n. 28.
139 The feminine is found apart from Jonah 2:2 a further 14 times in the OT [listed].
140 The ה locale is found only rarely as the nomen rectum of a construct relationship, but cf. Josh 15:5; 18:15.
Conclusion
I cannot explain why this solution has not been mentioned more often in literature, nor explored in more detail at the scholarly level (or perhaps it has been explored, I the reference has not yet surfaced). For this reason, I leave the phrasing "may be" in my opening title to the answer, though I would replace it with "likely" from my perspective.
In my approach to the text, where the consonantal text is the inspired text (the pointing being useful commentary on pronunciation, but not without potential flaw),2 the directive ה is (1) possible, (2) makes the most sense of the text, and (3) has far fewer difficulties than seeing it as the feminine form of the word. This is because the original text, which would have been without pointing, would be הדגה (with the article as the text has) for either the feminine form of the word fish (דגה) or the masculine form with the directive he added (דגה).
NOTES
1 I am indebted to Davïd for the English translations of the German texts here, as I do not read German (though I was able to find the academia-bible.com lead through searching on "ה locale" and then use Google translate initially to determine the value of the quote).
2 Even if one does not believe in the text being inspired, the earliest textform as we know it would have had no pointing, so the form of the word argument I make stands either way for the earliest version of the text.
