The last line of Jonah’s psalm, and its loudest. Three moves in one verse:

1. A fronted pronoun of contrast. וַאֲנִי (“But I”) sets the speaker over against the מְשַׁמְּרִים הַבְלֵי־שָׁוְא of v. 8 — those who keep worthless vapors and forsake their covenant loyalty. Jonah is the counter-example.
2. Two cohortatives of cultic intent. אֶזְבְּחָה and אֲשַׁלֵּמָה are volitive forms, not plain futures. They are the standard vocabulary of the thanksgiving psalm: a rescued worshipper announces the sacrifice and the vow-payment he owes.
3. A one-word confession. יְשׁוּעָ֖תָה לַיהוָֽה — two words in Hebrew, verbless, with the preposition ל carrying possession: salvation is YHWH’s. This is the theological center of the book, and it is placed in the mouth of the prophet who ran.

Two features require comment: the suffix on לָּךְ, which OSHB parses as feminine, and the unusual ending on יְשׁוּעָ֖תָה. Neither changes the sense, but both are real forms worth naming.

The irony belongs to the narrative, not the translation: the sailors of 1:16 already sacrificed sacrifices and vowed vows to YHWH; Jonah, inside the fish, only promises to. The English should not editorialize — it should let the same vocabulary recur and let the reader hear it.

Built onhalot · waltke-oconnor

Proposed — pending reviewJack M. Sasson, Jonah: A New Translation with Introduction, Commentary, and Interpretation, Anchor Bible 24B, New York: Doubleday, 1990.

The independent pronoun is syntactically unnecessary — אֶזְבְּחָה already encodes a first-person subject. Its presence is emphatic and contrastive, a fronted topic (casus pendens in effect, though here simply a fronted subject with the verb following). Joüon–Muraoka and Waltke–O’Connor both treat the redundant independent pronoun as marking contrast or emphasis on the subject.

The contrast is with v. 8, which has no explicit subject at all (מְשַׁמְּרִים, a bare masculine plural participle, “those keeping…”). Hebrew sets a nameless they against an emphatic I.

The same construction opened Jonah’s recollection at v. 4 (וַאֲנִ֣י אָמַ֔רְתִּי, “But I, I said”). Rendering both the same way — “But I” — keeps the two hinges of the psalm audible in English.

The waw is best taken as adversative here, not merely additive. English “But I —” with a dash preserves the fronting; “As for me” is also defensible and is used in the liturgical line.

Built onjouon-muraoka · waltke-oconnor

תּוֹדָה covers both the verbal act (thanksgiving, confession, praise) and the concrete offering (the tôdâ sacrifice of Lev 7:12–15, a subtype of the שְׁלָמִים). HALOT and BDB register both; DCH keeps them under one entry with the ranges distinguished contextually.

The construct בְּקוֹל תּוֹדָה — literally “with a voice of thanksgiving” — makes the audible dimension explicit. The ב is instrumental/accompanying: the sacrifice is performed with sung or spoken thanks. This is exactly the pairing Lev 7:12 and Ps 50:14, 23 assume, where the tôdâ offering and the spoken confession go together.

I keep קוֹל as “voice” rather than “sound” or “loud.” The word can mean either, but with a human speaker and a verbal noun in construct, “voice” is the precise sense, and it preserves the physicality: something is said aloud. Flattening it to “thankfully” or “with thanksgiving” loses a word the Hebrew spent.

Built onhalot · bdb · dch

Proposed — pending reviewJacob Milgrom, Leviticus 1–16, Anchor Bible 3, New York: Doubleday, 1991 (on the tôdâ offering, Lev 7:12–15).

Both verbs carry the cohortative ending ־ָה: אֶזְבְּחָה (qal 1cs) and אֲשַׁלֵּמָה (piel 1cs). OSHB codes both h (cohortative), and the forms bear it out.

The cohortative expresses the speaker’s will — resolve, self-exhortation, or (toward a superior) a request for permission. It is not a neutral future. Waltke–O’Connor classify this use as the cohortative of resolve: the speaker commits himself to an action.

This creates a genuine translation problem. Hebrew has one form; English has to choose:

I render the literal line with “I would” to mark the modality, the modern line with “I will” for the resolve, and the liturgical line with “let me,” which is closest to the form. The study reader should know that all three are reaching at one Hebrew form.

Built onwaltke-oconnor · jouon-muraoka

OSHB parses the suffix on לָּךְ as second feminine singular (Sp2fs). That is the normal value of the pointed form לָךְ. But the addressee throughout the psalm is YHWH, addressed with masculine forms: שָׁמַעְתָּ (2ms perfect, v. 2), מִשְׁבָּרֶיךָ וְגַלֶּיךָ (2ms suffixes, v. 3), קָדְשֶׁךָ (2ms, vv. 4, 7).

The explanation is prosodic, not grammatical. The word stands under zaqef qaton — at a major disjunctive pause — and the 2ms suffix ־ְךָ regularly loses its final vowel and appears as ־ָךְ in pause. This is a well-documented pausal phenomenon in the grammars.

So: the form is pausal 2ms, and the OSHB code should be read as describing the surface form rather than the referent. I note the disagreement rather than silently overriding it, per the working rule. Nothing in the sense is at stake — no serious reader takes Jonah to be addressing a goddess — but the form is instructive about how the Masoretic pointing works.

Built onjouon-muraoka · waltke-oconnor · bhs

Proposed — pending reviewWilhelm Gesenius, Gesenius’ Hebrew Grammar, ed. E. Kautzsch, rev. A. E. Cowley, 2nd English ed., Oxford: Clarendon Press, 1910 (on pausal suffix forms and the cohortative).

Draft — witness readings here are an AI’s recollection and have not been checked against the apparatus (BHQ, the Göttingen Septuagint, DJD). Verify before relying on them.

נָדַרְתִּי (qal perfect 1cs) is the cognate verb of נֶדֶר, the vow. The glossary already binds both to “vow,” set at 1:16 where the sailors וַיִּדְּרוּ נְדָרִים (“vowed vows”). Jonah now uses the same root.

The Israelite נֶדֶר is a specific legal-cultic instrument: a conditional self-obligation, typically “if God does X, I will do Y” (Gen 28:20–22; Num 30; Judg 11:30–31). It is voluntary to make and binding once made — Deut 23:21–23 and Eccl 5:4–5 both press the obligation to discharge it promptly. The perfect נָדַרְתִּי therefore points to a completed act of vowing, the terms of which are now owed.

The Hebrew does not say what Jonah vowed, or when. The headless relative אֲשֶׁר נָדַרְתִּי leaves it unspecified. Do not fill it in.

Built onhalot · bdb

אֲשַׁלֵּמָה is piel of שׁלם. The qal means “be complete, be at peace”; the piel is factitive — “make complete,” hence “repay, requite, discharge.” With נֶדֶר as object it is the fixed technical idiom for discharging a vow (Deut 23:21; Ps 22:25; 50:14; 61:8; 66:13; 116:14, 18).

English has two serviceable options:

• “pay” — the standard commercial-legal metaphor, and the one the idiom’s financial coloring supports. Used in the modern line.
• “fulfill” — captures the completion sense of the root more directly, without the transactional overtone. Used in the literal line.

The root is also the root of שָׁלוֹם, but that observation is of limited exegetical value here and must not be pressed: the piel’s established cultic meaning is set by usage, not by reasoning from the noun. (Root fallacy, per the working rules.)

Translation loss recorded: English cannot keep “vow… vow” as a figura etymologica the way Hebrew does elsewhere; here the Hebrew itself does not use the cognate noun, so nothing is lost at this verse. But the link back to 1:16 (נֶדֶר/נָדַר) must survive, and “vow” at both places secures it.

Built onhalot · bdb · dch

Two things here.

The form. יְשׁוּעָ֖תָה is an odd spelling of the feminine noun יְשׁוּעָה. OSHB analyzes it as Ncfsa/Sh — absolute noun plus paragogic/directional ה. The grammars generally treat this ending as an archaic or emphatic lengthening rather than a true directive (“to salvation” makes no sense here), and it is paralleled by similar lengthened forms in poetry (e.g. Ps 3:8 [Heb 3:9], יְשׁוּעָה with the same construction). It is a poetic form, not a different word. Nothing hangs on it semantically.

The syntax. The clause is verbless: יְשׁוּעָ֖תָה לַיהוָֽה. The ל is the lamed of possession — “salvation [belongs] to YHWH” — which is the construction’s normal force in a verbless predication of this shape. Three renderings are current:

• “Salvation belongs to YHWH” (possessive ל; the majority reading)
• “Salvation is from YHWH” (source; stretches ל toward מִן)
• “Salvation is YHWH’s” (same as the first, differently ordered)

I take the possessive. The accent is silluq on לַיהוָֽה, closing the verse with the divine name — a deliberate final placement.

יְשׁוּעָה itself is broader than later soteriological usage: deliverance, rescue, victory, help. In Jonah’s mouth, inside a fish, the concrete sense of rescue from death is primary. I render “salvation” to keep the term’s weight and the lexical link to the rest of the psalm, but a note is owed: this is not yet the full New Testament semantic field, and reading that back in would be anachronistic (per the diachronic-check rule).

Built onhalot · bdb · dch · waltke-oconnor · jouon-muraoka

Verse 9 is three cola:

waʾănî bəqôl tôdâ ʾezbəḥâ-lāk      But I — with a voice of thanksgiving I will sacrifice to you
  ʾăšer nādartî ʾăšallēmâ             what I have vowed I will pay
    yəšûʿātâ la-YHWH                    salvation belongs to YHWH

The first two cola are synthetic parallelism — the second advances the first rather than restating it: sacrifice, then the vow-payment that the sacrifice discharges. The third colon is not parallel to either. It is a detached nominal acclamation, shorter than the lines before it, and that shortness is the point: the psalm narrows from “I” to “YHWH” and stops.

The accents support the division: zaqef qaton after לָּ֔ךְ, athnach after אֲשַׁלֵּ֑מָה, silluq at the end. The Masoretic phrasing and the colometry agree.

The line breaks in the liturgical rendering follow this structure. Per the genre protocol, the parallelism is preserved as line breaks rather than smoothed into a single sentence.

Built onbhs · waltke-oconnor

LXX Jonah 2:10 reads, in substance: ἐγὼ δὲ μετὰ φωνῆς αἰνέσεως καὶ ἐξομολογήσεως θύσω σοι, ὅσα ηὐξάμην ἀποδώσω σοι εἰς σωτηρίαν μου τῷ κυρίῳ.

Three differences worth naming:

1. A doubled noun. The Greek has two nouns (αἰνέσεως καὶ ἐξομολογήσεως, “of praise and confession”) where the Hebrew has one (תּוֹדָה). This most likely reflects the translator unpacking the two sides of תּוֹדָה — praise and acknowledgment — rather than a longer Hebrew Vorlage. ἐξομολόγησις is the Greek translators’ standard equivalent for the hodayah vocabulary.
2. An added pronoun. The Greek supplies σοι (“to you”) with ἀποδώσω, which the Hebrew leaves implicit. Routine translational filling-out.
3. The last clause. The Greek reads something like “for my salvation, to the Lord” — taking the lamed and attaching a first-person possessive to σωτηρία. The Hebrew יְשׁוּעָ֖תָה לַיהוָֽה has no pronominal suffix; the ה is the poetic ending discussed above. It is plausible that the translator read the final ה as a first-person suffix or otherwise construed the form as “my salvation,” but this is an interpretive move within the Greek, not evidence of a different consonantal text. The Hebrew stands.

κύριος renders the Tetragrammaton, as throughout. Per the project rule, the Hebrew is not corrected toward the Greek at any point here.

Built ongoettingen-lxx-xii · muraoka-lxx · bhq

Draft — witness readings here are an AI’s recollection and have not been checked against the apparatus (BHQ, the Göttingen Septuagint, DJD). Verify before relying on them.

Vulgate. Jerome renders the verse with ego autem in voce laudis immolabo tibi; quaecumque vovi reddam pro salute Domino — “but I, with the voice of praise, will sacrifice to you; whatever I have vowed I will pay, for salvation, to the Lord.” Jerome follows the Hebrew closely (iuxta Hebraeos), with reddam for the piel of שׁלם, exactly the expected legal verb. His pro salute takes the last clause as purpose/cause rather than possession, which differs from the reading preferred above.

Targum Jonathan expands this psalm throughout in the manner of the Targumim, typically paraphrasing cultic language and making the addressee explicit. The general character of the expansion is well known; the specific wording at this verse should be checked against a critical Aramaic text before being quoted.

Qumran. The Minor Prophets are attested at Qumran and in the Naḥal Ḥever Greek scroll, but I will not state what any manuscript reads at this verse without the apparatus in front of me. BHS and BHQ should be checked.

Built onbhs · bhq

Draft — witness readings here are an AI’s recollection and have not been checked against the apparatus (BHQ, the Göttingen Septuagint, DJD). Verify before relying on them.

This verse is built out of the standard vocabulary of the individual thanksgiving psalm. The clearest overlaps:

• Ps 50:14 — “Sacrifice to God a tôdâ, and pay (שַׁלֵּם) your vows (נְדָרֶיךָ) to the Most High.” The same three terms (זבח / תודה / שׁלם + נדר) stand together. This is the closest parallel in the Psalter.
• Ps 116:17–18 — “To you I will sacrifice a sacrifice of tôdâ… my vows to YHWH I will pay.” Again the same cluster, in the same order.
• Ps 66:13–14 — entering the house with offerings, paying vows the lips uttered in distress.
• Ps 3:8 (Heb 3:9) — “Salvation belongs to YHWH” (לַיהוָה הַיְשׁוּעָה), the nearest formal parallel to the closing colon, with the elements inverted.

This is not quotation. It is a prophet’s prayer composed in the inherited idiom of temple thanksgiving — which is itself a datum about the book: Jonah, in the fish, prays in the liturgical language of the sanctuary he cannot reach, and twice in this psalm looks toward הֵיכַל קָדְשֶׁךָ (vv. 4, 7).

Built onbdb · halot

Proposed — pending reviewHans Walter Wolff, Obadiah and Jonah: A Commentary, trans. Margaret Kohl, Minneapolis: Augsburg, 1986.

Three observations, kept to what the Hebrew supports.

Vow-payment is not bargaining. The נֶדֶר is conditional in form but the obligation runs one direction: having received, the vower owes. Jonah is not negotiating with YHWH here; he is stating a debt. The perfect נָדַרְתִּי places the vow in the past and the cohortative אֲשַׁלֵּמָה places the discharge ahead.

The confession is impersonal, and that matters. The final colon has no first-person element at all. Not “YHWH saved me” but “salvation belongs to YHWH” — a general predication. The psalm’s movement is from the drowning I of vv. 3–7 to a statement about God that is true whether or not the speaker is rescued.

The book will test this line. Ch. 4 shows the man who says יְשׁוּעָ֖תָה לַיהוָֽה angry that YHWH exercised exactly that prerogative over Nineveh. The narrator does not comment. Neither should the translation — the irony is constructed by juxtaposition, and rendering 2:9 with any added edge would be the translator doing work the text deliberately leaves to the reader.

A note on the sailors. At 1:16 the Gentile crew וַיִּזְבְּחוּ־זֶבַח לַיהוָה וַיִּדְּרוּ נְדָרִים — sacrificed a sacrifice to YHWH and vowed vows. Jonah here promises to sacrifice and to pay a vow. The same two roots, the same order; the sailors in the perfect, Jonah in the cohortative. The Hebrew vocabulary makes the comparison; English preserves it only if “sacrifice” and “vow” are used at both places, which the glossary requires.

Built onhalot · bdb

How later readers used the verse — not evidence for what the Hebrew means.

Rabbinic. The tractate on vows (Nedarim) and the discussions of prompt vow-payment grounded in Deut 23:22 and Eccl 5:4 form the natural setting in which a verse like this was read, though I am not able to cite a specific passage treating Jonah 2:9 without checking. Jonah as a whole is read in the synagogue as the haftarah at the Minḥah service of Yom Kippur — a liturgical placement that puts this verse’s language of vow and thanksgiving directly into the day’s penitential frame. That liturgical use is well established; the reasoning behind its selection is a matter of commentary rather than record.

Early Church. Patristic reading of Jonah 2 is dominated by the typological frame set by Matt 12:40, where the three days in the fish prefigure the burial and resurrection. Within that frame, the thanksgiving of 2:9 was read as the prayer of one already as good as raised, and Jerome, in his commentary on Jonah, treats the psalm in these terms. The move from Jonah’s I will sacrifice to a Christological reading is a reception-level development; nothing in the Hebrew supports it, and nothing in the Hebrew excludes it from being read that way afterwards.

Liturgical afterlife. יְשׁוּעָ֖תָה לַיהוָֽה and its near-twin at Ps 3:8 became a stock acclamation in Jewish and Christian prayer alike. Its brevity made it portable. That portability is reception, not meaning.

Proposed — pending reviewJerome, Commentarii in Ionam prophetam (Commentary on Jonah), ed. Y.-M. Duval, Sources Chrétiennes 323, Paris: Éditions du Cerf, 1985.

Draft — the later uses cited here (New Testament, rabbinic, patristic) are an AI’s recollection and have not been checked against the sources. Verify before relying on them.