The verse is the low point of Jonah’s psalm and its most physical line. Three clauses, each tighter than the last, close the water over the speaker: waters engulfed me, the deep swirled around me, reeds bound about my head. The movement is from the mass of water, to the cosmic deep beneath it, to a single tangible detail — weed wrapped round a drowning man’s skull.

Three decisions carry the verse.

1. עַד־נֶפֶשׁ (ʿad-nepeš) is genuinely two-sided: “up to the throat” (water line at the neck) or “up to the life” (threatening the self). Both are lexically defensible and the Hebrew does not choose. The project glossary binds נֶפֶשׁ to “life,” so the main rendering says “up to the life / up to my very life,” with the throat image noted rather than suppressed.
2. תְּהוֹם is not generic deep water but the tehôm of Gen 1:2 and Gen 7:11 — the primeval deep. Jonah is describing drowning in creation’s own vocabulary, which is why the psalm can move from here to Sheol in v. 6.
3. סוּף חָבוּשׁ לְרֹאשִׁי uses the noun of the yam-sûp, the Reed Sea of Exod 15. There the sea covered Israel’s enemies and Israel went through on dry land; here the reed binds the Israelite prophet. The allusion is lexical and almost certainly deliberate.

The verb forms are mixed — perfect (אֲפָפוּנִי), imperfect/polel (יְסֹבְבֵנִי), passive participle (חָבוּשׁ) — which is ordinary in Hebrew poetry describing past distress and should not be flattened into three uniform English pasts without noting the shift.

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

אֲפָפוּנִי is qal perfect 3cp with a 1cs suffix: “they engulfed me.” The root אָפַף is rare — five occurrences in the Hebrew Bible, and every one is in poetry of mortal peril. Twice it takes water as subject (here and Ps 18:5 ‖ 2 Sam 22:5 in the parallel wording of the cords/torrents of death), and twice the subject is the cords or snares of death or Sheol (Ps 18:5–6; Ps 116:3).

The lexicons agree on the core sense — encompass, surround, enclose — and differ mainly in emphasis: BDB “encompass,” HALOT “surround, engulf,” DCH listing both. What the distribution shows is not an etymology but a register: this is death-language. The subject is plural מַיִם with a 3cp verb, perfectly regular.

“Engulfed” is chosen over the flatter “surrounded” because the second clause already carries “encircle” (סָבַב), and the two Hebrew verbs are not synonyms: אָפַף closes in, סָבַב goes round.

Built onbdb · halot · dch

The phrase is two words and no suffix: literally “up to nepeš.” Both readings are live.

Throat / neck. נֶפֶשׁ has a well-attested concrete sense of throat or gullet (so HALOT, with Isa 5:14, where Sheol opens its נֶפֶשׁ, and Ps 69:2, “the waters have come up to the נֶפֶשׁ,” the closest parallel to this verse). On this reading the clause is a water-line: the flood reaches his neck.

Life / self. נֶפֶשׁ as the living self under threat of death is the dominant biblical sense and the one the project glossary binds for Jonah (1:14; 2:5, 7). On this reading עַד־נֶפֶשׁ means “to the point of death,” the water reaching not a body part but the life itself.

Note the choice not to supply “my.” The Hebrew is anarthrous and unsuffixed; the literal line keeps that bareness (“up to [the] nepeš”), and only the modern line supplies “my very life” for English sense.

Built onhalot · bdb · dch

תְּהוֹם is the primeval deep. It is the word of Gen 1:2 (darkness over the face of the tehôm), of Gen 7:11 (the fountains of the great tehôm burst open), and of Exod 15:5, 8 (the tehōmōt cover Pharaoh’s army; the deeps congeal). Hebrew regularly leaves it without the article, treating it almost as a name, and it does so here.

The term is cognate with Akkadian tiāmtu/tâmtu, “sea,” and the Babylonian creation myth’s Tiamat belongs to the same lexical family. That is cognate evidence for the word, not evidence that Jonah’s psalm is reciting a combat myth: in biblical usage תְּהוֹם is a created thing under YHWH’s hand, never a rival deity, and no mythological personification is grammatically marked here.

The choice of תְּהוֹם rather than יָם (“sea”) is what lets the poem move downward in the next verse to מִשַּׁחַת (“from the pit”) and connects this line to שְׁאוֹל in v. 2. Jonah is not merely in deep water; he is in the deep.

Built onhalot · bdb · dch

The form is identical to the one in v. 3 (וְנָהָר יְסֹבְבֵנִי, “and a river kept encircling me”), and the repetition is structural: the same verb binds the two halves of the descent. OSHB codes it Vmi3ms — the m marks the polel (a doubling stem of hollow roots, functionally equivalent to piel for סָבַב). Third masculine singular, with תְּהוֹם as subject; תְּהוֹם is of common gender and takes masculine agreement here.

The aspect matters. אֲפָפוּנִי is perfect — a completed event. יְסֹבְבֵנִי is imperfect in a past-time narration, which in Hebrew poetry commonly carries durative or iterative force: the deep went on encircling. Hebrew verbs encode aspect and sequence, not tense, and the shift from perfect to imperfect within one verse is a deliberate texturing — a single closing act, then an ongoing swirl.

English cannot easily carry that in three words. The literal line uses “was encircling”; the modern line uses “swirled,” with the duration carried by the verb’s own sense.

A minority reading takes the imperfects in vv. 3 and 5 as true presents — Jonah still in the water as he prays. The psalm’s framing in v. 1 (he prayed from the fish’s innards) and the resolution in v. 6 (“you brought up my life from the pit”) tell against it, but the forms themselves do not exclude it.

Built onwaltke-oconnor · jouon-muraoka

סוּף is reed or water-plant — collective singular, hence “reeds” in English. It is the noun of יַם־סוּף, the sea of the exodus (Exod 15:4, 22), and of the reeds among which the infant Moses was placed (Exod 2:3, 5).

חָבוּשׁ is a qal passive participle of חָבַשׁ, “bind.” The root is used of binding on a turban (Exod 29:9), saddling an animal (Gen 22:3), and bandaging a wound (Isa 1:6; Ezek 34:4). The participle is masculine singular, agreeing with the collective סוּף, and there is no verb “to be” — the clause is verbless: “reed bound to my head.”

The image is exact and grim: weed wrapped round the head like a wrapping or a bandage. Some have heard in חָבוּשׁ a grotesque echo of the priestly headdress being bound on, since the same verb governs the turban in Exod 29:9; the lexical overlap is real, but nothing else in the clause supports a cultic reading, and it should be left as an observation rather than pressed.

Built onhalot · bdb · dch

Three of this verse’s four content words appear together in the Song of the Sea.

• תְּהוֹם — Exod 15:5, “the deeps covered them”; 15:8, “the deeps congealed.”
• סוּף — Exod 15:4, יַם־סוּף, the sea into which Pharaoh’s chariots were hurled.
• מַיִם covering a man — Exod 15:10, “the waters covered them.”

In Exodus 15 the water closes over Egypt and Israel walks through on dry ground. In Jonah 2 the water closes over the Israelite prophet, and he reaches dry ground (יַבָּשָׁה, 2:10) only by being vomited onto it. The vocabulary is the vocabulary of salvation at the sea, applied to a fleeing prophet who is in the position of the enemy.

The nearer parallels are in the Psalter, which uses this same cluster for distress:

• Ps 69:1–2 (Heb 69:2–3) — waters come up to the נֶפֶשׁ; the deep engulfs; no foothold. The closest verbal relative of this verse.
• Ps 18:4–5 (Heb 18:5–6) ‖ 2 Sam 22:5 — אָפַף with torrents and the cords of Sheol.
• Ps 42:7 (Heb 42:8) — deep calling to deep; breakers and waves passing over, as in Jonah 2:3.

Whether Jonah’s psalm quotes these or draws with them on shared liturgical stock cannot be settled from the wording alone; the overlap is in standard phrases, not in distinctive full lines.

Built onbdb · halot

LXX (Jonah 2:6 in Greek numbering) reads: περιεχύθη μοι ὕδωρ ἕως ψυχῆς, ἄβυσσος ἐκύκλωσέν με ἐσχάτη, ἔδυ ἡ κεφαλή μου εἰς σχισμὰς ὀρέων.

Three differences bear on meaning.

ἕως ψυχῆς — “as far as the soul/life” — takes נֶפֶשׁ in the vital rather than the anatomical sense, matching the decision made above and showing it is ancient. The translator did not read “throat.”

ἄβυσσος … ἐσχάτη, “the uttermost deep,” adds an adjective with no counterpart in the Hebrew; it reads as intensification by the translator rather than a different Hebrew text.

ἔδυ ἡ κεφαλή μου εἰς σχισμὰς ὀρέων, “my head sank into the clefts of the mountains,” replaces the reed clause entirely. The Greek has no equivalent of סוּף and instead imports the mountains that in the Hebrew belong to the next verse (לְקִצְבֵי הָרִים, 2:6). Two explanations are current: the translator did not recognize סוּף in this context and reconstructed from the following line, or he read a slightly different Hebrew. Since Greek σχισμὰς ὀרέων plainly corresponds to the Hebrew of v. 6, the simpler account is a translator reaching forward to make sense of a clause he found obscure, with a consequent redistribution of material across the verse boundary. The precise Göttingen text and apparatus should be checked before this is pressed further.

No variant here gives grounds for altering the Hebrew. The MT clause is harder and coherent; the Greek is smoother and secondary.

The Vulgate keeps the reed: circumdedit me abyssus, pelagus operuit caput meum in Jerome’s rendering departs from the reed image as well, while his Iuxta Hebraeos work elsewhere shows awareness of the Hebrew word; the exact Vulgate wording for this verse should be verified against a critical edition before being cited as evidence.

Built ongoettingen-lxx-xii · muraoka-lxx · bhs

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.

The consonantal text of this verse is stable and the WLC reading raises no serious doubt. BHS records no emendation that has won assent; the apparatus concerns itself chiefly with the LXX’s divergence in the third clause (above), which is a translational rather than a textual problem.

Two points where the Masoretic layer is worth naming:

Accents. The ʾatnaḥ falls on יְסֹבְבֵ֑נִי, dividing the verse after the second clause. This gives a 2 + 1 shape: two clauses of water, then the reed. If one prefers a three-part colon the division is still audible, but the Masoretes heard the reed clause as the verse’s weighted close. The accents record a reading tradition and are evidence, not binding.

Pointing. עַד־נֶ֔פֶשׁ with zaqef qaton keeps נֶפֶשׁ absolute, without suffix. Any rendering that supplies “my” is interpreting, not translating, and the literal line declines to.

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.

The verse is poetry, and the lines should not be run together.

א  אֲפָפוּנִי מַיִם עַד־נֶפֶשׁ      waters engulfed me, to the nepeš
ב  תְּהוֹם יְסֹבְבֵנִי              the deep encircled me
ג  סוּף חָבוּשׁ לְרֹאשִׁי          reed bound to my head

Lines א and ב are synonymous parallelism with intensification: מַיִם → תְּהוֹם moves from water in general to the primeval deep, and אָפַף → סָבַב from closing in to encircling. Line ג is not a third synonym but a narrowing — the only clause with no verb, the only one naming a body part other than by implication, and the shortest. The effect is a zoom: ocean, abyss, skull.

The word order is also worth preserving. In א the verb stands first; in ב and ג the subject is fronted (תְּהוֹם, סוּף). That fronting is what makes the second and third clauses feel like successive facts closing in rather than a continuing narrative, and the renderings keep the subject-first order.

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

The verse makes no theological claim directly. It has no divine name, no verb of prayer, no petition. It is pure description of being overwhelmed — and that is the point of its placement.

Two things nevertheless follow from the vocabulary.

First, the agent is unstated but not unknown. In v. 3 the psalm said plainly וַתַּשְׁלִיכֵנִי — “you hurled me into the deep.” The waters and the deep of v. 5 therefore are not impersonal nature closing over an unlucky man; they are the instrument already attributed to God two verses earlier, as the wind and the storm of 1:4 were. The passive and intransitive clauses here let the reader feel the water without God’s hand being visible, which is exactly the experience of the sufferer.

Second, the psalm uses the standard language of Israel’s worship for a man who had been running from Israel’s God. The one who went down (יָרַד, 1:3, 1:5) into the ship’s hold now finds the deep going round him, and he describes it with the words Israel used of the Egyptians. The structure of the book asks the reader to notice the irony, not to resolve it.

Nothing in the Hebrew of this verse speaks of death and resurrection as such; the psalm says the speaker is as good as dead and that YHWH brought his life up from the pit (v. 6). The later Christological reading of Jonah’s three days belongs to the Reception section, not here.

Second Temple and rabbinic. The Targum to the Minor Prophets renders Jonah’s psalm with its characteristic expansion, making the sea imagery explicit; the exact wording of this verse should be checked against a critical edition of Targum Jonathan before being quoted. Rabbinic midrash on Jonah (notably the aggadic expansions behind the medieval Pirqe de-Rabbi Eliezer / Midrash Jonah tradition) elaborated the prophet’s tour of the deep in great detail, including the reeds and the foundations of the world — reading this verse as itinerary rather than metaphor. Attribution of specific passages here requires verification.

New Testament. Jonah’s three days in the fish are cited at Matt 12:40 and his preaching at Luke 11:29–32, but neither passage quotes this verse. The water-and-deep vocabulary of Jonah 2 was not taken up directly in the New Testament; the Psalter’s parallel language was (for example Ps 69 is quoted repeatedly in the Passion narratives).

Early Church. Jerome’s commentary on Jonah treats the psalm as genuine prayer from inside the fish and reads the waters and the deep as the overwhelming of the whole man, a figure of death; he also discusses the Greek’s divergence in the third clause. Cyril of Alexandria’s commentary on the Twelve Prophets handles Jonah 2 similarly, as description of death’s grip rather than allegory of the sea. Patristic interest in the chapter as a whole is dominated by the typology of Christ’s burial — drawn from Matt 12:40 rather than from this verse’s wording.

All of this is how the verse was read. None of it is evidence for what the Hebrew says.

Proposed — pending reviewJerome, Commentarius in Ionam prophetam (Commentary on Jonah), ed. Y.-M. Duval, Sources chrétiennes 323, Paris: Éditions du Cerf, 1985. · Cyril of Alexandria, Commentary on the Twelve Prophets, trans. Robert C. Hill, Fathers of the Church, Washington, DC: Catholic University of America Press. · Kevin J. Cathcart and Robert P. Gordon, The Targum of the Minor Prophets, The Aramaic Bible 14, Wilmington, DE: Michael Glazier, 1989.

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.