The standard critical lexicons converge on the gloss "crown of the head, pate, vertex" and diverge only on etymology. HALOT lists Ugaritic qdqd, Akkadian qaqqadu(m) (from older qadqad-), and Jewish Aramaic qodqoda as cognates and treats the Hebrew as a reduplicated primary noun, cross-referenced to but not derived from the verb qdd "bow down". BDB (869) already in 1906 assigned the noun to a separate root II ("meaning dubious") precisely to detach it from qdd I. DCH lists the eleven attestations and the word-pairs with se'ar "hair" (Ps 68:21), ro'sh "head", and pe'ah "side, temple" (Jer 48:45) without deriving it. The picturesque derivation from qdd, current from the Tregelles Gesenius (1857) through Strong, is abandoned in Gesenius18 and in modern comparative work: Butts (JAOS 131) documents the productive Semitic reduplicated nominal pattern, and Militarev and Kogan's Semitic Etymological Dictionary reconstructs Proto-Semitic *qadqad- "top of the head" from Akkadian, Hebrew, Ugaritic, and Aramaic. TDOT/ThWAT gives the noun no standalone article, treating it under ro'sh (ThWAT VII cols. 271ff. = TDOT XIII 248ff.).
The comparative evidence is unusually exact. Ugaritic qdqd preserves the unassimilated base and appears in the same word-pair with r'is that Hebrew keeps in poetry; in the Baal Cycle (KTU 1.2 IV 21-25) Baal's club strikes the qdqd of Prince Yamm, in Kirta (KTU 1.16 VI 54-57) a curse formula pairs r'isk with qdqdk, and in Aqhat (KTU 1.18 IV 22-23, 33-34) Yatpanu strikes Aqhat on the qdqd. Watson and Sivan treat the ro'sh // qodqod pair as inherited Northwest Semitic diction rather than a Hebrew innovation. The Akkadian merism istu qaqqadi adi sepe "from head to foot" (CAD Q 106-108) is the direct ancestor of Hebrew mikkaf regel we'ad qodqod, so the totality idiom of Deut 28:35, 2 Sam 14:25, and Job 2:7 is a shared Semitic literary convention; Clines and Habel read the Job occurrence as a deliberate echo of the Deuteronomic curse.
The Septuagint chose koryphe ("summit, crown of head"), the semantically exact equivalent, in seven of eleven cases. Job 2:7 flattens to kephale in the free style of Old Greek Job; Deut 33:20 renders archonta ("ruler"), a metaphorical reinterpretation that Symmachus corrected back to koryphen; Isa 3:17 paraphrases the clause away entirely, with 1QIsa-a confirming qdqd in the Vorlage; Jer 2:16 loses the noun through a graphic dalet/resh confusion analysed by Tov; and Jer 48:45 falls inside one of the well known LXX-Jeremiah minuses. Ps 68:21 [LXX 67:22] calques the Hebrew as koryphen trichos.
Three exegetical debates touch the word without unsettling its sense. On the Joseph blessings (Gen 49:26 // Deut 33:16) the disputed term is nazir: Speiser, Wenham, Westermann, Christensen, and Tigay read "prince, elect, distinguished one", Waltke and Nelson keep "consecrated, set apart", and Sasson argues both senses stay active, since the nazir's status is marked precisely at the crown of the head by diadem or unshorn hair; the parallelism intensifies rather than chooses. On Ps 68:21, Dahood (following Albright) read qodqod se'ar as a near quotation of Baal Cycle diction; Barr's methodological critique and Craigie's cautions, and the mediating position of Tate, Kraus, Hossfeld and Zenger, Goldingay, and Mark S. Smith, retain the inherited word-pair and the Divine Warrior type-scene while rejecting detailed mythological reconstruction. On Isa 3:17 the hapax verb sippah divides between "smite with a scab" (Wildberger, Blenkinsopp, Oswalt) and "make bald", with Williamson concluding it probably fuses skin disease and hair loss; on Jer 2:16 the verb yir'uk is read as "graze upon" (Holladay) or "shave" (Lundbom, McKane). BDB, following Ewald and Dillmann, proposed emending weqarqar in Num 24:17 to weqodqod on the strength of the Jer 48:45 parallel, but the reading is conjectural and witnessed in no manuscript, so the count of attestations stands at eleven.