Psalm 22: "Pierced" or "Like a Lion"?

Summary

Psalm 22 is the passage Christians reach for first when arguing that the Old Testament predicted the crucifixion.  Its centrepiece is one clause, in most English Bibles at verse 16:  "they pierced my hands and my feet."  Jewish anti-missionary writers reply that the Hebrew says nothing of the kind — it says "like a lion, my hands and my feet" — and that "pierced" is a Christian forgery.  Both sides have spent two thousand years asserting more than the evidence supports.

• The Hebrew of the Masoretic Text is כארי (ka'ari), and it is a real word meaning "like a lion." It appears four other times in the Hebrew Bible, and one of those — Isaiah 38:13, Hezekiah's psalm — uses it in almost the same image: "like a lion, so he breaks all my bones." Rashi cited exactly that verse to explain Psalm 22, and the 1985 Jewish Publication Society translation cites Rashi in its own footnote.^{1 2} Anyone who tells you "like a lion" is nonsense Hebrew has not checked.
• But the clause as it stands has no verb, and everyone in antiquity noticed. "Like a lion my hands and my feet" is not a sentence. Every ancient version supplies something: the Targum adds "biting," the Septuagint reads a verb "they dug," the Peshitta "they burst through," Jerome's Hebrew Psalter "they bound," Aquila "they disfigured."³ The Masoretic reading is the hardest one and the one everybody tried to fix.
• No manuscript in any language says "pierced." The Greek verb is ὤρυξαν (ōryxan), from ὀρύσσω, "to dig" — the word used in the Gospels for digging a winepress and digging a hole to bury money.⁴ The Latin foderunt means the same. "Pierced" is an English interpretation of "dug," which is itself a Greek interpretation of contested Hebrew consonants. It is three steps from the text, and the standard Hebrew verbs for piercing — דקר and נקב — appear nowhere near this psalm.
• The Dead Sea scroll everyone cites is not from before Christ. 5/6ḤevPs, the Psalms scroll from the Cave of Letters at Nahal Hever, is dated late Herodian, AD 50–68 — a generation after the crucifixion.³ Mike Winger tells his audience it "goes back to the time before Christ" in both of his talks on the psalm, and draws a timeline graphic on it; he takes the claim from Michael Brown, who dates it "to the century before Yeshua."^{5 6 6} It does not, and he does not. The one Qumran fragment that might have covered the verse, 4Q88, is too damaged to read; a 2025 study says flatly that no letters other than ידי ו can be recovered with confidence.³
• What that scroll does show is that the reading is genuinely old and genuinely divided. 5/6ḤevPs reads כארו (ka'aru), with a waw where the Masoretic Text has a yod — and the scholars who report this most confidently are the ones arguing against it being original. They note that this particular scroll distinguishes waw from yod unusually clearly.³ Tovia Singer's claim that the letter is simply illegible is contradicted by the specialists least motivated to concede it.⁷ Others, including the Jewish-studies scholar Rivka Ulmer, still read it as an elongated yod.⁸
• The tampering charge fails in both directions. Justin Martyr, around AD 160, published a list of the passages he accused Jewish scribes of deleting — Esdras, Jeremiah 11:19, and "from the wood" in Psalm 96. Psalm 22:16 is not on it, and his Jewish interlocutor disputes who the psalm is about, not what it says.^{9 9} Conversely, the Septuagint's "they dug" is pre-Christian Jewish work, and the Greek Psalter's existence before Christianity is attested by the grandson of Ben Sira around 132 BC.¹⁰ Neither side rewrote the other's Bible here.
• The decisive fact is one no apologist uses: the New Testament never quotes this verse. It quotes Psalm 22:1 from the cross, alludes to 22:7–8 in the mockery, quotes 22:18 verbatim for the garments, and quotes 22:22 in Hebrews. Verse 16 is absent — and ὀρύσσω, the Greek verb it would have needed, occurs in the New Testament only three times, all in vineyard-and-buried-money parables.⁴ If the early church had forged or exploited this verse, that is the last pattern you would expect.
• Crucifixion in the modern sense had not been invented, but hanging men on stakes had. Herodotus, writing in the fifth century BC, reports Darius impaling three thousand Babylonians and the Athenians nailing Artayctes to planks and hanging him up.^{11 11} So the imagery was not unthinkable in the ancient Near East. What was distinctively Roman was the combination — public, prolonged, with nails.
• And the only skeleton of a crucifixion victim ever found had nailed feet and, apparently, tied hands. Zias and Sekeles's 1985 re-examination of the Giv'at ha-Mivtar remains found no trauma to the forearms or hands, and concluded "the arms of the condemned were tied rather than nailed to the cross."¹² The Gospels themselves never say nails were driven through Jesus's feet; John mentions nail marks in the hands, and Luke has Jesus show "my hands and my feet" without mentioning nails at all.⁴

Verdict:  the verse will not carry the weight put on it, and it never was the strongest part of the argument.  The textual evidence is real but genuinely split, and the strongest recent study — by two evangelical scholars — concludes that the Masoretic consonants are original and that the poet deliberately wrote an ambiguous word meaning both "like a lion" and "gouging."³  That is a more interesting answer than either popular side offers, and it is not the crucifixion prediction the apologetic needs.  As for whether Psalm 22 as a whole is prophecy that could not have been faked:  the psalm's pre-Christian date is secure, so forgery is ruled out.  What is not ruled out — and what the argument actually turns on — is whether the Gospel writers described the crucifixion in the psalm's language because it happened that way, or told it that way because they had the psalm open.  On that question the honest answer is mixed, and the details that survive scrutiny best are the ones nobody could have staged.

The text, and the claim being made

Here is Psalm 22 in the NET Bible, which follows the Masoretic Text at the disputed verse.  Note the versification:  the Hebrew Bible counts the superscription as verse 1, so Hebrew 22:17 is English 22:16.  This article uses English numbering unless it says otherwise.

Yes, wild dogs surround me — a gang of evil men crowd around me;  like a lion they pin my hands and feet.  I can count all my bones;  my enemies are gloating over me in triumph.  They are dividing up my clothes among themselves;  they are rolling dice for my garments. — Psalm 22:16–18, NET¹³

And the same verses in the New King James Version, which most viewers of the popular apologetics videos will be holding:

For dogs have surrounded Me;  the congregation of the wicked has enclosed Me.  They pierced My hands and My feet;  I can count all My bones.  They look and stare at Me.  They divide My garments among them, and for My clothing they cast lots.¹⁴

The argument built on this is straightforward.  A psalm written centuries before Rome existed describes a public execution by piercing of the hands and feet, together with the victim's bones out of joint, his tongue stuck to his palate, mockers quoting his own trust in God back at him, and soldiers gambling for his clothing.  Since the psalm cannot have been written after the event, and since Christians could not have altered a text that Jews were guarding, the fulfilment must be genuine prediction.

The counter-argument is equally straightforward.  The Hebrew does not say "pierced";  it says "like a lion."  The word "pierced" was introduced by Christian translators who knew better.  As Tovia Singer puts it:

The King James version deliberately mistranslated the Hebrew word kaari (כָּאֲרִי) as "pierced," rather than "like a lion," thereby drawing the reader to a false conclusion that this Psalm is describing the Crucifixion. … Needless to say, the phrase "they pierced my hands and my feet" is a not-too-ingenious Christian contrivance that appears nowhere in Tanach.⁷

Both of these are testable claims about manuscripts.  Let us test them.

Part 1 — What the Hebrew actually says

The Masoretic reading

The consonantal text of the Leningrad and Aleppo codices reads:

כִּי סְבָבוּנִי כְּלָבִים  עֲדַת מְרֵעִים הִקִּיפוּנִי  כָּאֲרִי יָדַי וְרַגְלָי

Word by word:  ki sevavuni kelavim — "for dogs have surrounded me";  adat mere'im hiqqifuni — "a company of evildoers has encircled me";  ka'ari yaday weraglay — "like-a-lion my-hands and-my-feet."¹⁵

The first two clauses each have a finite verb.  The third has none.  That is the whole problem, and it has been the whole problem for two thousand years.  As one Hebrew study of the verse puts it, "the essence of the difficulty is the absence of a verb."³

"Like a lion" is real Hebrew — a concordance check

The claim that ka'ari cannot mean "like a lion" here is sometimes supported by an argument from Michael Brown:  that the normal Hebrew for lion is the longer form אריה (aryeh), which Psalm 22 itself uses three verses earlier at 22:13, so the shorter ארי here is suspicious.¹⁶

That argument can be checked mechanically.  Running the Westminster Leningrad Codex and stripping the vowel points gives these counts for "like a lion" with the prefixed כְּ:

So both forms are attested with the preposition, and the longer one is more than twice as common.¹⁷  Within the Psalter the longer form wins 3 to 1, which is a modest point in Brown's favour.  But the short form is not an anomaly — it occurs four other times, and the counter-claim sometimes made online, that כארי is the only possible form after כְּ, is also wrong.

The decisive datum is not the count but the parallel.  Isaiah 38:13 — Hezekiah's psalm of thanksgiving after his illness — reads:

שִׁוִּיתִי עַד־בֹּקֶר כָּאֲרִי כֵּן יְשַׁבֵּר כָּל־עַצְמוֹתָי "Then it was as though a lion were breaking all my bones;  I cried out until morning."¹⁸

Same word, same short form, same image:  a sufferer describing his body being broken like a lion would break it.  And the next verse of Psalm 22 is "I can count all my bones."  Rashi drew exactly this connection in the eleventh century:

like a lion, my hands and feet — As though they are crushed in a lion's mouth, and so did Hezekiah say (in Isa. 38:13):  "like a lion, so it would break all my bones."¹

The JPS Tanakh translates Psalm 22:17 "like lions [they maul] my hands and feet" and footnotes it "With Rashi;  cf. Isa. 38.13."²  This is not a desperate rabbinic dodge.  It is the reading a competent Hebraist reaches for, and the NET Bible — a conservative evangelical translation — adopts it:  "like a lion they pin my hands and feet."^{13  19}

There is one further detail here that neither camp mentions, and it is worth the thirty seconds it takes to check.  The five occurrences of these consonants split by the vowel under the kaph.  Numbers 23:24, Numbers 24:9 and Ezekiel 22:25 have patach.  Isaiah 38:13 and Psalm 22:17 have qamats — the pointing that marks the preposition plus a swallowed definite article, "like the lion" — and no other verse in the Bible does.²⁰  The Masoretes pointed our verse identically to Isaiah 38:13 and to nothing else.  That is strong evidence that they read it as a lion.  It is also precisely the pair the Okhlah we-Okhlah flags as one word occurring twice with two different meanings, discussed below — which means the same tradition that pointed the two verses alike was on record that they do not mean the same thing.

The other reading, and why it is not simple either

The competing reading is כארו (ka'aru), identical except that the final letter is a waw rather than a yod.  In the Hebrew square script of the turn of the era those two letters differ mainly in the length of a downstroke, and confusing them is the single most common scribal error in the Dead Sea Scrolls.

But ka'aru is not a Hebrew word either.  Singer is right about this:

Transforming kaari (כָּאֲרִי) into kaaru (כָּאֲרוּ) … does not create the Hebrew word for "pierced." … In fact, kaaru doesn't mean anything.  In other words, this word kaaru does not exist in the Hebrew language.⁷

To get a verb you have to do something with the aleph.  The obvious candidate is כרה (karah), "to dig," whose third-person plural perfect is כָּרוּ (karu) — but that has no aleph at all.  So the choices are:

1. Treat the aleph as a vowel letter, a spelling quirk, giving a form of כרה, "to dig." This is what the Septuagint translator evidently did, and it is not an ad hoc rescue: Friedrich Delitzsch catalogued cases of aleph used irregularly for e, i, o and u vowels, and one of them is the exact morphological parallel — Jeremiah 30:16's שאסיך, "your plunderers," a qal active participle from שסה with aleph marking the o-vowel. Hosea 10:14 writes וְקָאם for a form of קום, with the aleph standing where a waw belongs.²¹
2. Treat the word as a plural participle with the mem dropped — כָּאֲרִים written defectively — from the root כאר/כער, which in Rabbinic Hebrew means "to be repulsive," in the intensive "to disfigure." This is what Aquila evidently did, and what Barthélemy argued was the ancient understanding preserved in the Masorah.²²
3. Posit a lost root. J. J. M. Roberts proposed a כרה-V cognate with Akkadian and Syriac karû, "to be short," yielding "my hands and feet are shrivelled"; the NRSV adopted it. John Kaltner proposed an Arabic cognate kwr meaning "to bind." Both were criticised by Brent Strawn, who noted that the aleph is unexpected on either proposal and that the Classical Arabic evidence largely postdates the Hebrew Bible.³

Notice what none of these produce.  Not one of them means "pierce."  The nearest is "dig," and the semantic distance from digging a cistern to driving a nail through a wrist is the whole apologetic argument, unstated.

"Pierced" is not a translation of anything

This deserves to be said plainly, because it is the single most misunderstood point in the whole dispute, and it is misunderstood in the same direction by both sides.

The Septuagint reads ὤρυξαν χεῖράς μου καὶ πόδας — "they dug my hands and feet."²³  ὀρύσσω is a common word for excavation.  It appears three times in the New Testament, and in all three it means to dig in the ground:  a man digs a winepress (Matthew 21:33), a man digs a winepress (Mark 12:1), a servant digs a hole and buries his master's money (Matthew 25:18).⁴  The Old Latin and Vulgate foderunt is a straight rendering of it, and the Douay-Rheims, translating the Vulgate, gives the honest English:  "They have dug my hands and feet."²⁴

The chain, then, runs:  contested Hebrew consonants → a Greek verb meaning "dug" → a Latin verb meaning "dug" → an English verb meaning "pierced."  The last step is an interpretation, and it happened in the sixteenth century.  The 2025 study of the verse in TC: A Journal of Biblical Textual Criticism is careful to say "they dug my hands and feet" and then adds, in a phrase both camps should sit with, that this is "a reading that seems to portray the piercing of the hands and feet."³

Singer's polemical version of this point — that no Hebrew word here means "pierce," and that if David had wanted "pierce" he would have written daqar or ratza — is correct as far as it goes.  What it does not establish is his conclusion, that "pierced" is a knowing fraud by King James's translators.  They were following the Greek and the Latin, as Christian translators had for fifteen centuries, and the Geneva Bible of 1587 ("they perced mine hands and my feete") and the Bishops' Bible of 1568 had already done the same.²⁵  Being wrong about a verb is not the same as lying.

Part 2 — The manuscript evidence, witness by witness

The Hebrew from the Judean Desert

4Q88 (4QPs^f), late Hasmonean, c. 50 BC.  This Qumran fragment covers Psalm 22:14–17 and is sometimes cited — including in The Dead Sea Scrolls Bible — as reading כרו, "they dug."  It should not be.  The 2025 study went back to the infrared plates and concluded:

In our view, no letters other than ידי ו can be reconstructed with confidence from the bottom line of the fragments in question.  This illustrates the need for caution when using the DJD volumes for text-critical decisions.³

They add that 4Q88 is a liturgical scroll, not a scripture scroll, and that within these four verses its scribe also wrote מדבש ("honey") for מדבק ("is stuck") and a sin for a samekh.  A scribe that free is a poor witness to anything.  4Q88 should be dropped from the argument by both sides.

5/6ḤevPs (5/6Hev 1b), late Herodian, AD 50–68.  This is the scroll everyone means when they say "the Dead Sea Scrolls read pierced."  It comes from the Cave of Letters at Nahal Hever, about thirty kilometres south of Qumran, used by refugees of the Bar Kokhba revolt.  Its images are public in the Leon Levy Digital Library.²⁶

Three things about it matter, and the popular presentations get two of them wrong.

First, the date.  It is a first-century AD manuscript, from a generation after the crucifixion.  It is Jewish, not Christian — but it cannot establish what Hebrew Bibles said before Jesus was born.  Mike Winger tells his audience that "the oldest Jewish manuscript that's still in Hebrew is a Dead Sea Scrolls copy of this passage that goes back to the time before Christ."⁵  That is simply an error, and it is the error that does the most work in his presentation, because "before Christ" is what would make the reading unfakeable.

Second, the reading.  The DJD 38 editors read כארו, with a waw.  Singer replies that "it is difficult to discern by studying the faint, ancient text whether the word in question ends in a elongated י (yud) or a shortened ו (vav)."⁷  The most recent specialist treatment disagrees with Singer, and it disagrees while arguing against the waw being the original reading:

This manuscript clearly reads a vav instead of MT's yod in the relevant phrase … This manuscript is known for its clear distinguishing of vav and yod, as seen … in the sequence of a verb that ends with vav followed by a verb that begins with yod (יביטו יראו).³

That is a concession against interest and should be weighted accordingly.  It is not unanimous:  Rivka Ulmer describes the final letter as "a somewhat elongated letter י (yod), almost appearing like the letter (vav)."⁸  So the honest summary is:  the editors of the scroll and the most recent detailed study read waw;  a competent minority still reads yod;  nobody thinks the question is trivial.

Third — and this is the point both camps skate past — the aleph is in the scroll.  Even the manuscript that reads a verb spells it כארו, with the aleph that no form of כרה should have.  Whatever the original was, it had an aleph in it.  That single observation constrains every theory on the table, and it is the hinge of the 2025 solution described below.

Against that, the same study notes that the very next word in the scroll is written ידיה — "her hands" — where the text needs ידי, "my hands."  A scribe who has just miswritten one word is a slightly weaker witness on the next.³

The medieval Hebrew manuscripts and the Masorah

The apologetic literature commonly reports "about a dozen medieval manuscripts" reading כארו.⁵  The apparatus of Biblia Hebraica Stuttgartensia is more modest:  pauci manuscripti (a few manuscripts) read כארו, and two read כרו.³  Kennicott and de Rossi's eighteenth-century collations are the source of the larger figures.  This article has not collated Kennicott directly, and readers should treat "a dozen" as an unverified number;  what is securely reported in the standard critical apparatus is "a few" plus two.

Stranger and more interesting is the Masorah itself.  The Masorah Gedolah of the Venice rabbinic Bible records כארו as ketiv (what is written) with כארי as qere (what is read).  Jacob ben Hayyim, who edited that Bible in the sixteenth century, was puzzled by his own note:

in a few of the grammar books I found written כארו with qere כארי.  However, I looked for other words with ketiv ו at the end of the word and qere י, but I did not find it listed in their calculation.³

And the Okhlah we-Okhlah, an early Masoretic word-list, includes כארי among words that occur twice in the Bible with two different meanings — the two occurrences being Isaiah 38:13 and Psalm 22:17.²²  In Isaiah the meaning is certainly "like a lion."  So the old Masoretes are on record that the Psalm 22 occurrence means something else.  This is a remarkable piece of evidence, and it comes from the very scribal tradition accused of suppressing the alternative.

The Greek

The Septuagint Psalter renders the verse:

ὅτι ἐκύκλωσάν με κύνες πολλοί, συναγωγὴ πονηρευομένων περιέσχον με, ὤρυξαν χεῖράς μου καὶ πόδας. "For many dogs encircled me, an assembly of evildoers surrounded me;  they dug my hands and feet."²³

The translator plainly read, or chose to construe, a finite verb from כרה.  Two features of his rendering are worth noting.  He also adds "many" to the dogs, which is not in the Hebrew and is probably borrowed from verse 13's "many bulls" — so he was not a mechanical copyist.  And the surrounding verses are otherwise close to the Masoretic Text.

Aquila, Symmachus, and the later Greek

The three later Jewish Greek translations matter more than their obscurity suggests, because they were made deliberately to correct the Septuagint toward the Hebrew.  Where they agree with the Masoretic Text, that tells us little.  Where they diverge, it is evidence of a real variant.

• Aquila is cited two different ways. Eusebius, quoting him directly, and manuscript 1175 both give ᾔσχυναν χεῖράς μου καὶ πόδας μου — "they disfigured my hands and my feet."²² The margin of the Syrohexapla instead gives a Syriac phrase that Frederick Field retroverted as ἐπέδησαν, "they bound."³
• Symmachus is likewise doubled. Field, again from the Syrohexapla margin, printed ὡς ζητοῦντες δῆσαι, "as those seeking to bind." But Charles Taylor published a seventh-century Cairo Genizah palimpsest of Psalm 21 (=22) in the versions, and it gives Symmachus as ὡς λέων τὰς χεῖράς μου καὶ τοὺς πόδας μου — "like a lion, my hands and my feet." Taylor argued, and Alison Salvesen finds it persuasive, that the "binding" reading arose from an inner-Greek corruption of ΩϹΛΕΟΝΤΑϹ to ΩϹΔΕΟΝΤΑϹ, which the Syriac translator then paraphrased.³

So the best evidence has Symmachus supporting "like a lion" and Aquila reading a verb of disfigurement.  Barthélemy connected Aquila's choice to his known taste for Homer:  in the Iliad the same verb αἰσχύνω is used of dogs defiling the body of the dead Priam.²²

The Syriac, Latin, and Aramaic

• Peshitta: ܒܙܥܘ (bza'w), "they burst through / tore open / pierced." This Syriac verb normally renders Hebrew בקע, "to split," and once renders קרע, "to tear." It may reflect the translator hearing קרעו in the consonants, or it may reflect Christian interpretation — the Peshitta Psalter is Christian-transmitted.³
• Vulgate, Gallican Psalter (translated from the Greek): foderunt manus meas et pedes meos — "they dug my hands and my feet."²²
• Vulgate, iuxta Hebraeos (Jerome's own translation from the Hebrew): vinxerunt manus meas et pedes meos — "they bound my hands and my feet."²²
• Targum (the Jewish Aramaic version): נכתין היך כאריא אידי ורגלי — "biting like a lion my hands and feet."³

Jerome deserves a moment.  He is the one ancient Christian scholar who worked directly from a Hebrew text with rabbinic help, and he is the last person who would have suppressed a crucifixion prophecy.  Given the chance to render the Hebrew himself, he wrote "they bound."  Whatever his Hebrew manuscript said, it did not tell him "they pierced."

The witnesses in one table

What the standard critical committee concluded

The United Bible Societies' Hebrew Old Testament Text Project, whose findings Dominique Barthélemy published in Critique textuelle de l'Ancien Testament, examined this verse and gave the Masoretic reading a grade of {B} — meaning "some doubt."  Two of the committee's members dissented down to {C}.²²

Their recommended translation is neither of the popular options:

car des chiens me cernent, une bande de vauriens m'entoure, défigurant mes mains et mes pieds "for dogs surround me, a band of scoundrels encircles me, disfiguring my hands and my feet."²²

Their reasoning was that the versions are "simple exegesis of M's reading" — that is, that every ancient translator was looking at consonants very like the Masoretic ones and guessing at a meaning.³

Part 3 — Testing the tampering charges

Both sides accuse the other of forgery.  Both accusations can be checked.

Did Christians insert "pierced"?

Singer's version of the charge has two parts:  that the King James translators knowingly mistranslated, and — a much larger claim — that the Greek Old Testament outside the Torah is not Jewish work at all:

It is universally conceded, and beyond doubt that the rabbis who created the original Septuagint only translated the Five Books of Moses, and nothing more. … In addition, the rest of the Septuagint is a translation by Christian scholars with a strong motive to twist the messages of the Jewish Bible.⁷

The first half of that is standard scholarship:  the Letter of Aristeas and Josephus do describe only a Torah translation.  The second half is refuted by a Jewish source written before Christianity existed.  The prologue to the Greek Ben Sira, written by the author's grandson after he arrived in Egypt "in the thirty-eighth year, in the reign of Euergetes the king" — that is, 132 BC — says:

what was originally expressed in Hebrew does not have the same force when it is in fact rendered in another language.  And not only in this case, but also in the case of the Law itself and the Prophets and the rest of the books the difference is not small when these are expressed in their own language.¹⁰

A Jewish translator in Egypt in 132 BC is telling us that the Law, the Prophets, and "the rest of the books" already existed in Greek.  He does not name the Psalms;  strictly, the inference that "the rest of the books" includes the Psalter is an inference.  But it is not a strained one — the Psalms are the largest and most liturgically used book in that third division — and it does not stand alone.  The later Jewish revisers, Aquila and Symmachus and Theodotion, were revising an existing Greek Psalter rather than making the first one, and the New Testament authors quote a Greek Psalter that matches ours.  There is no room in that chain for a Christian origin.

That does not vindicate "pierced."  It relocates the question.  The Septuagint is a genuinely pre-Christian Jewish witness, and it read a verb;  but the verb it read was "dug," not "pierced."

Did Jewish scribes remove it?

The mirror-image charge is that the Masoretes changed כארו to כארי to escape the Christian argument.  Christians have been making versions of this charge since the second century, so we can test it directly:  Justin Martyr, writing around AD 160, published a specific list of the passages he believed Jewish scribes had deleted.  He names three:

• a passage about the Passover attributed to Esdras;
• Jeremiah 11:19 in an expanded form ("let us lay on wood on His bread");
• and, most famously, the words "from the wood" in Psalm 96:10 — "Tell among the nations, the Lord has reigned from the wood."^{9 9}

He introduces the list by saying that Jewish teachers "have altogether taken away many Scriptures from the translations effected by those seventy elders."⁹  Trypho's reply, which is worth having because it is the only Jewish answer to the tampering charge Justin lets us hear, is:  "Whether the rulers of the people have erased any portion of the Scriptures, as you affirm, God knows;  but it seems incredible."  Justin's own comeback is a small masterpiece of unintentional candour — "Assuredly, it does seem incredible."⁹

Psalm 22:16 is not on the list.  And Justin was not ignoring the psalm — he quotes it in full at Dialogue 98, comments on it verse by verse through chapter 106, and cites "they pierced my hands and my feet" three separate times.^{9  9  9}  If the Hebrew he or his Jewish contemporaries knew had read "like a lion," Justin had both the occasion and the appetite to say so.  He does not.

More telling still is what Justin reports of the other side of the argument:  "And this very Psalm you maintain does not refer to Christ;  for you are in all respects blind."⁹  The dispute in the second century, as Justin presents it, is about the psalm's referent, not its wording.

That is real evidence, and it cuts against the theory of a late Masoretic alteration.  But it is not decisive, for two reasons.  Justin read Greek, argued from Greek, and probably knew no Hebrew;  his silence tells us the Greek was uncontested, and cannot tell us what stood in the Hebrew scrolls of the synagogues.  And Trypho is Justin's own literary construction, so what "Trypho does not object to" is really what Justin did not think worth staging an objection about — which is weaker evidence, though not worthless, since a controversialist normally stages the objections his opponents actually make.

Where a tampering charge does have some bite

Barthélemy notes one place where the accusation lands, and it lands on the Christian side.  The Complutensian Polyglot — the great Catholic scholarly Bible printed at Alcalá in 1514–17 — prints the Hebrew of this verse as כארו.  Barthélemy's comment is dry:  "one may suspect there a tendentious choice of the type that Joseph Qimhi deplored" when he complained of Christians altering the Hebrew.²²

So the honest finding is symmetrical and unflattering to nobody's tribe.  In the second century, the reading was not a point of dispute at all.  In the sixteenth, a Christian editorial team appears to have put a thumb on the scale.  And the Masoretic tradition, far from suppressing the alternative, preserved a note in the Okhlah we-Okhlah saying the word means something different here than in Isaiah, and a ketiv/qere tradition reading כארו.

Part 4 — Where scholarship now stands

The verse has generated a small literature of its own, with article titles that tell the story:  Vall's "The Old Guess" (1997), Kaltner's "Second Guessing 'The Old Guess'" (1998), Strawn's "More Guessing" (2000), Swenson's "Circling around the Problem Again" (2004), Linville's "A New Guess" (2005).³  Strawn's closing line in 2000 was:

Who knows how many more guesses will come along before the crux at Ps 22:17b is solved once and for all?  Only one thing is certain:  in the present methodological context, recourse to the versions and to scholarly emendations is no longer enough.³

The most recent full-dress treatment is Seth Postell and Joseph Justiss, "The Crux of Psalm 22:17," published in TC: A Journal of Biblical Textual Criticism in November 2025.³  Both authors are evangelicals — Postell teaches at Israel College of the Bible — and their conclusion is not the one an apologist would order from a catalogue.

They argue that the Masoretic consonants כארי are the earliest recoverable form, on the grounds that a scribe under pressure from the two preceding finite verbs would naturally "fix" a puzzling word into a third verb, whereas nobody would create the harder reading on purpose.³  But they also argue that the word is not simply "like a lion."  Their case rests on the psalm's own poetics.

Psalm 22 plays this game elsewhere.  Its opening אלי אלי can be read "my God, my God" or "to me!  my God" — and the Septuagint translator, confronted with it, rendered both, adding "pay attention to me."  In verse 11 the string אלי אתה can be "you are my God" (looking back) or "come to me!" (looking forward).  In each case an aleph is what makes the double reading possible, and the psalm's first three lines all begin with aleph.³

On their reading, כארי in verse 17 is a deliberate third instance:  a string of consonants meant to be heard as "like a lion" looking back to the animal imagery, and as a participle of כרה — "gouging at," "digging into" — looking forward to the hands and feet.  This explains the otherwise inexplicable aleph that even the Nahal Hever scroll preserves.  Their proposed English, which imitates the pun with a different animal, is:

For dogs have surrounded me.  An assembly of evildoers has encircled me, bearing down on "my hands" and "my feet."³

They support this with a structural observation worth repeating.  Psalm 22 pairs each complaint with a matching request:  bulls (v. 13) with wild oxen (v. 22), the roaring lion (v. 14) with "save me from the lion's mouth" (v. 22), dogs (v. 17) with "from the power of the dog" (v. 21).  If verse 17 also mentions a lion, that lion has no matching request — but verse 21's request is to be delivered from the sword, an instrument of cutting, which has no matching complaint.  Put the two gaps together and verse 17 should be describing a cutting, gouging injury.³

They then raise, in a footnote, a possibility neither popular side has considered.  Both יד ("hand") and רגל ("foot") are attested Hebrew euphemisms for genitals.  Aquila's "they disfigured," read alongside the Homeric scene of dogs defiling Priam's body, and the Greek verb ὀρύσσω's use in Aristophanes for violent assault on body parts, may mean the ancient translators understood the verse as describing the exposure and abuse of a stripped victim.  And if so:

this could explain why there is no direct allusion to this verse in the gospel crucifixion accounts even though several verses from this psalm are alluded to quite obviously.  In other words, the evangelists may not have understood the verse to be describing the piercing of literal hands and feet but rather the cruel exposure of nakedness.³

That is a striking suggestion from a conservative source, and it points directly at the fact that decides this whole question.

Part 5 — Could this have been faked? Six tests

The question that actually matters is not "what does the Hebrew say."  It is whether Psalm 22 is evidence that could not have been manufactured.  That question has six distinct components, and they have different answers.

Test 1 — Is the pre-Christian date secure?

Yes, decisively.  This is the strongest part of the whole case and it is rarely stated properly.

Psalm 22 in Hebrew is attested at Qumran, and the Greek Psalter's existence by the second century BC is established by the prologue to Ben Sira, quoted above.  The Greek reading "they dug" is therefore pre-Christian, whatever it means.  No Christian invented it, and no Christian could have inserted it into a text already circulating in Jewish communities from Alexandria to Judea.

The Hebrew side is not equally secure, because the only Hebrew manuscript that preserves the disputed word — Nahal Hever — is from AD 50–68.  But this does not matter much, because the Hebrew is not in dispute in the way that would help.  The Masoretic consonants are old;  the versions all presuppose consonants close to them.

So forgery is ruled out.  What is not ruled out is that a pre-Christian Greek translator, faced with an unreadable Hebrew clause, guessed "dug," and that Christians two centuries later found the guess congenial.  Nothing about that requires bad faith from anybody, and it is what most textual critics actually think happened.

Test 2 — Could the Gospel writers have built the scene from the psalm?

This is the serious objection, and it is not answered by manuscript evidence at all.

John Dominic Crossan named the theory "prophecy historicized":  the details of the passion are not memories but midrash, generated by early Christians reading Psalm 22 and Isaiah 53 and turning them into narrative.  In The Birth of Christianity he put it this way:

The individual units, general sequences, and overall frames of the passion-resurrection stories are so linked to prophetic fulfillment that the removal of such fulfillment leaves nothing but the barest facts, almost as in Josephus, Tacitus or the Apostles' Creed.²⁷

There is a real observation underneath this.  Mark's crucifixion narrative is saturated with Scripture, and the allusions are not decorative:  "It is not as if one could take away the Scriptural elements and still have a coherent narrative."²⁷

Mark Goodacre's reply, which is the best-argued version of the alternative, is that Crossan has offered a false choice between "history remembered" and "prophecy historicized."  A third process is available and is well attested elsewhere:  scripturalization, in which a remembered event gets retold in scriptural language.  Goodacre's illustration is Mark 15:40, where the women watch "from afar" — a phrase from Psalm 38:11.  Nobody thinks Mark invented the named women out of that psalm;  the tradition was told in the psalm's words.  "In other words, in this verse we see the exact opposite of the process of 'prophecy historicized.'"²⁷

He also lands a logical point on Crossan's own premise.  Crossan argues that the disciples had all fled, so nobody could have reported the crucifixion.  But the source for "everyone deserted him and fled" is Mark 14:50 — which follows immediately from "let the scriptures be fulfilled" and Jesus's quotation of Zechariah 13:7.  On Crossan's own criteria that detail is prophecy historicized too, and if it is, the premise that there were no witnesses collapses.²⁷

So the honest position is:  the passion narratives were certainly shaped by Psalm 22, and some details may have been generated by it.  Others show marks of independent tradition.  Sorting them one by one is the only way to get anywhere, and it is a different job from textual criticism.

Test 3 — The silence that decides it

Here is the strongest single fact in this entire article, and it belongs to neither side's talking points.

The New Testament never quotes or alludes to Psalm 22:16.

Consider what it does use from this psalm:

John quotes verse 18 with an explicit fulfilment formula — "that the scripture might be fulfilled" — and reproduces the Septuagint word for word:  Διεμερίσαντο τὰ ἱμάτιά μου ἑαυτοῖς καὶ ἐπὶ τὸν ἱματισμόν μου ἔβαλον κλῆρον.⁴  He is reading the Greek Psalter, closely, at the crucifixion, in the very sentence before the one that supposedly predicts the nails.  And he says nothing about it.

The Greek verb tells the same story.  If any New Testament author had wanted to make this connection, the word to hand was ὤρυξαν.  It appears in the New Testament three times — Matthew 21:33, Matthew 25:18, Mark 12:1 — and every one is a parable about digging in dirt.⁴

And when the New Testament does want a piercing prophecy, it goes somewhere else entirely.  John 19:37 quotes Zechariah 12:10 — Ὄψονται εἰς ὃν ἐξεκέντησαν, "they will look on the one whom they pierced" — of the spear thrust into Jesus's side.  Revelation 1:7 uses the same verb.⁴  The vocabulary for piercing was available, was used, and was applied to a different wound.

What does this prove?  Two things, in opposite directions.

Against the apologetic:  the first Christians, who had the Greek Psalter in front of them and were combing it for crucifixion prophecies, did not read Psalm 22:16 as one.  The reading that is now the centrepiece of the argument was invisible to the people best placed to see it.  The first Christian to cite it is Justin Martyr, in the First Apology around AD 155, and Judith Lieu argues Justin was probably also the first person to read the whole psalm of Jesus.^{28  8}

Against the forgery charge:  a church that had manufactured or exploited this verse would have used it.  Instead we have the extraordinary situation of a text supposedly forged for apologetic purposes going wholly unused in the apologetic documents.  Whatever produced the "dug" reading, it was not the passion narrative working backwards, because the passion narrative does not know about it.

Test 4 — Was crucifixion even conceivable when the psalm was written?

The standard claim, repeated by Winger, is that "crucifixion had not been invented until the 4th Century by the Persians."⁵  The date is wrong, and it is wrong in a way that weakens the argument slightly.

Herodotus, writing in the 430s BC about the previous century, gives three relevant reports:

• Oroetes killed Polycrates of Samos, around 522 BC, and ἀνεσταύρωσε him — hung him on a stake. Herodotus adds the grim detail that the hanging body was "washed by Zeus whenever it rained, and anointed by the sun."¹¹
• Darius, after retaking Babylon around 521 BC, ἀνεσκολόπισε about three thousand of its leading men.¹¹
• The Athenians executed the Persian governor Artayctes at Sestos in 479 BC: πρὸς σανίδας προσπασσαλεύσαντες ἀνεκρέμασαν — "having nailed him to planks, they hung him up."¹¹

That last one is the closest pre-Roman parallel there is:  nails, wood, public suspension.  So execution by fastening a living or dead body to timber was known in the ancient Near East by the sixth century BC, and impalement is depicted on Assyrian reliefs earlier still.  What the Romans developed was the refinement — the prolonged public death, standardised, with a reusable upright.

This matters for the argument in a specific way.  Under the traditional tenth-century date for a Davidic psalm, the image would still predate any known crucifixion by four hundred years.  Under a post-exilic date the gap shrinks to nothing much.  And the date of Psalm 22 is genuinely open.  Michael Lyons's survey of the compositional models notes that "there are no unambiguously early or late linguistic features that would allow us to date Psalm 22 with certainty" — Avi Hurvitz's lexicon of Late Biblical Hebrew flags none of its vocabulary, but a psalm this short cannot be dated on absence.²⁹  Proposals run from pre-exilic (with Psalm 22 influencing Isaiah's Servant) to a post-exilic composition dependent on Isaiah 40–66, with most commentators treating verses 28–32 as a later expansion.²⁹

An argument that requires the tenth century is resting on a date that critical scholarship does not grant and that conservative scholarship cannot demonstrate.

Test 5 — Does the fulfilment even match?

Suppose "they dug my hands and my feet" is the original reading and does mean piercing.  Does the crucifixion of Jesus fit it?

Less exactly than the sermons suggest.

The Gospels never say nails were driven through Jesus's feet.  John 20:25 has Thomas demand to see "the mark of the nails" in Jesus's hands and to put his hand into Jesus's side.  Luke 24:39 has Jesus say "See my hands and my feet, that it is I myself" — and the point of that sentence is flesh and bones, not wounds;  it mentions no nails, and the following verse is textually uncertain.⁴  There is no New Testament sentence stating that Jesus's feet were nailed.

The archaeology complicates it further.  Only two skeletons anywhere in the world carry osteological evidence of crucifixion, and both were nailed through the heel and neither shows a nailed hand.

The first is Yehohanan, found in an ossuary at Giv'at ha-Mivtar north of Jerusalem in 1968.  Nicu Haas's original 1970 study reconstructed a single nail through both heels together and a second through the forearms.  Joseph Zias and Eliezer Sekeles re-examined the remains in 1985 and corrected almost all of it:  the nail was 11.5 cm, not 17–18, so it could not have pierced two feet;  the supposed nail-scratch on the radius was one of many non-traumatic marks found on ancient bone, with two identical ones on the same skeleton's right fibula.  Their conclusion:

the lack of traumatic injury to the forearm and metacarpale of the hand seems to suggest that the arms of the condemned were tied rather than nailed to the cross.  There is ample literary and artistic evidence for the use of ropes rather than nails to secure the condemned to the cross.¹²

They reconstruct the victim straddling the upright with each foot nailed sideways.¹²

The second was found in 2017 at Fenstanton in Cambridgeshire and announced in 2021 — a man aged 25 to 35, radiocarbon-dated AD 130–360, with a 5 cm iron nail driven horizontally through his right heel bone, and a second indentation beside it where a first attempt had missed.  Corinne Duhig, the osteoarchaeologist who examined him, explains why such finds are almost nonexistent:

Although crucifixion was common in the Roman world, osteological evidence for the practice is extremely rare … as nails were not always used — the victim was normally just tied to a crossbar — and bodies were not typically given formal burials.³⁰

None of this means Jesus's hands were not nailed — John 20:25 says they were, and Roman practice varied.  It means the picture in the standard apologetic — one nail per hand, one nail per foot, exactly as Psalm 22 "predicts" — is a devotional image rather than an evidenced one.  The two skeletons we have do not match it, and the specialist's summary of Roman practice is that the hands were normally tied.

Test 6 — What would faking it have required?

Set aside the reading question and ask the counterfactual directly.  For the Psalm 22 argument to be a fraud, one of the following has to be true:

1. Christians altered the Hebrew. Ruled out: the Hebrew reads "like a lion" in the Masoretic tradition, which is the unhelpful reading. If Christians had that power they used it badly.
2. Christians altered the Greek. Ruled out by date: the Greek Psalter is second-century BC, and Jewish communities used it for two centuries before Christianity.
3. Jews altered the Hebrew after the fact. Not supportable: Justin's tampering list omits the verse, the Masorah preserves both a ketiv כארו tradition and a note that the word means something other than "lion" here, and the Nahal Hever scroll's aleph shows the difficulty is older than the controversy.
4. The Gospel writers built the crucifixion scene out of the psalm. Partly true for language, unproven for events, and specifically false for verse 16, which they never use.
5. The whole crucifixion is fiction. Not a serious position. Jesus's execution under Pilate is reported by Tacitus, by Josephus in the passage's uncontested core, and by Paul writing within twenty-five years of the event.

None of these gets you a forgery.  What you can get, without accusing anybody of anything, is a much more modest and much more likely story:  a genuinely obscure Hebrew clause;  a Greek translator's guess;  a Christian reader two centuries later who found the guess suggestive;  and fifteen hundred years of English Bibles inheriting the reading without re-examining it.

Part 6 — The rest of Psalm 22, which is the better argument

Strip verse 16 out entirely and see what is left.  A great deal is left, and it is stronger, because it is what the New Testament actually uses.

The cry from the cross.  Mark 15:34 has Jesus quote Psalm 22:1 in Aramaic:  Ἐλωῒ ἐλωῒ λεμὰ σαβαχθάνι.  Matthew 27:46 gives it as Ἠλὶ ἠλὶ λεμὰ σαβαχθάνι — a Hebrew "Eli" with an Aramaic remainder, which is also what makes the bystanders' confusion with Elijah work.⁴  If this were a scribal insertion from the Greek Psalter, we would expect the Greek Psalter's words.  Instead we have a transliterated Semitic sentence that has to be translated for the reader, and Mark's translation (εἰς τί) does not even match the Septuagint's (ἵνα τί).  The line looks like tradition, not citation.  Joel Marcus's judgement is that the church "would not have created such problems for itself."²⁷  Goodacre disagrees, arguing Mark is not embarrassed at all but is using the psalm to make an abject death theologically legible.²⁷  The disagreement is worth knowing about;  the datum is worth having.

The mockery.  Matthew 27:43 puts into the mouths of the chief priests:  "He trusts in God;  let God deliver him now, if he wants him."  Psalm 22:8 reads:  "Commit yourself to the LORD!  Let the LORD rescue him!"  Matthew is the only evangelist with this line, and it is very close to the psalm.  This is a plausible candidate for scripturalization — the mockery being told in the psalm's words.

The garments.  John 19:23–24 is the strongest and the most interesting case, and it cuts both ways.  John reports two distinct actions:  the soldiers divide the outer garments four ways, then cast lots for the seamless tunic rather than tear it.  He then quotes Psalm 22:18 exactly and says the scripture was fulfilled.⁴

The problem a critic will raise immediately is that Hebrew poetry works by parallelism:  "they divide my clothes among themselves;  they cast lots for my garments" is one action described twice, not two actions.  John has read the parallelism as a sequence and produced a narrative with two stages to match.  That is precisely the fingerprint of a story shaped by a text — and the other three Gospels do not do it.  Matthew 27:35 and Mark 15:24 both use the psalm's vocabulary while keeping it a single act:  "they divided his garments, casting lots."  John is the outlier, and he is the outlier in exactly the direction that a writer working from the psalm would go.⁴

Against that:  the underlying event needs no prophecy to explain it at all, because Roman law recognised it.  The pannicularia were the effects of a condemned man, and the Digest preserves a rescript of Hadrian defining them:  "the clothing he was wearing, or small coins in his money-belt kept for his food, or light rings — that is, not exceeding five aurei in value."  Ulpian's gloss makes the setting explicit:  pannicularia are "the spoils in which he is clothed when a man is led to execution, as the very name shows," and he goes on to restrain the executioner-soldiers — the speculatores — from claiming them.³¹  A rescript restricting a perquisite presupposes the perquisite, and the soldiers Ulpian names are the soldiers John names.

So the event is ordinary Roman routine and the shaping is scriptural — which is exactly Goodacre's scripturalization model, and it is the model that fits the evidence best.  It is worth noting that Winger reaches the same conclusion on this point by a different route, quoting A. N. Sherwin-White on the executioner squad's customary right to a victim's minor possessions.⁶

The ending.  The last third of the psalm turns from lament to worldwide praise:  "Let all the people of the earth acknowledge the LORD and turn to him. … A whole generation will serve him;  they will tell the next generation about the LORD.  They will come and tell about his saving deeds;  they will tell a future generation what he has accomplished."¹³

The Hebrew of that last clause is כִּי עָשָׂה — ki asah, "that he has done it."  Preachers regularly line it up with Jesus's τετέλεσται, "It is finished," in John 19:30.  It is a genuinely arresting parallel;  it is also two common words, and John gives no citation formula.  Take it as resonance, not proof.

Hebrews 2:12 quotes Psalm 22:22 straight, putting "I will declare your name to my brothers" in the mouth of the risen Christ — evidence that the earliest Christians read the psalm's vindication half messianically too, not only its suffering half.⁴

What Jewish tradition did with the psalm

It is worth knowing what the psalm meant to readers who were not arguing about Jesus.

Rabbinic tradition overwhelmingly read Psalm 22 of Esther.  The Babylonian Talmud has Esther cry "My God, my God, why have you forsaken me" as she enters Ahasuerus's court, and identifies Ahasuerus with the "dog" of verse 21 and the "lion" of verse 22.³²  Midrash Tehillim runs the whole psalm through the Esther story:  the bulls of Bashan are Ahasuerus's hosts, the dogs are Haman's sons, and the dividing of garments is a squabble over Esther's purple robe.³³  On our clause it preserves R. Judah's reading — Esther's hands were made repulsive in Ahasuerus's eyes — which is the same כער derivation Aquila used.²²

But rabbinic tradition also produced a suffering-Messiah reading.  Pesiqta Rabbati 36–37 applies Psalm 22 to Messiah Ephraim, who accepts affliction for the sins of others:

In the future the sins of those that have been hidden with you will bring you under an iron yoke … and because of their sins your tongue will stick to the roof of your mouth (Ps. 22:16).  Are you willing to endure this?⁸

Rivka Ulmer's assessment of the whole tradition is balanced and worth quoting on both halves:

Prior to the attestation in the New Testament, there is no evidence of Psalm 22 being used in a Jewish messianic context. … Psalm 22 is rarely cited in tannaitic literature, and this avoidance may suggest a reaction against the Christian use of this Psalm.⁸

So:  no pre-Christian Jewish messianic reading of Psalm 22 exists.  The Christian reading is a Christian innovation, first fully articulated by Justin.  And the subsequent Jewish avoidance of the psalm looks like a reaction to that innovation.  Rashi, for his part, does place verse 27 "at the time of the redemption, in the days of the Messiah" — but that is a claim about when, not about who.³⁴

Part 7 — Grading the popular presentations

Mike Winger

Winger has two talks on this psalm:  "Jesus in Psalm 22: Evidence for the Bible pt6" (2019, 53 minutes) and a longer live stream, "Really Specific Prophecy Jesus Fulfilled on the Cross!  Amazing!" (11 September 2019, 85 minutes), which answers named critics — the atheist YouTuber AronRa and Tovia Singer.^{5  6}  Both were transcribed for this article.

His instincts are good.  He states the strongest objection he knows and plays Singer's own audio rather than paraphrasing.  He grants that even on the "like a lion" reading, mauled hands and feet would still suit a crucifixion.  And in the longer talk he makes a concession most apologists do not:  he lists sixteen points of correspondence, removes the pierced-hands one outright, and says the case stands on the remaining fifteen — "even if you lose verse 16, Psalm 22 is still pretty powerful."⁶  That is the right way to argue, and this article agrees with him about it.

He also answers the direction-of-dependence objection rather than ignoring it, which is more than most popular treatments do — though his closing move there is weak.  "The skeptic who says the New Testament authors made this stuff up so that it would look like Psalm 22 … they're admitting it looks like Psalm 22."⁶  They are, and that is common ground:  both the invention model and the scripturalization model predict the resemblance.  The resemblance is the thing to be explained, not the explanation.

He gets four things wrong, in ascending order of importance.

• "Crucifixion had not been invented until the 4th Century by the Persians."⁵ Herodotus reports Persian impalement in the 520s BC and Athenian nailing-to-planks in 479 BC. The longer talk revises this to "300 years later, the Assyrians, the Persians started using it," which is closer to right and should be the version he keeps.⁶
• "[The Septuagint] translates this into a Greek word [ōryxan] which means pierced."⁵ ὀρύσσω means "to dig." Winger describes the Hebrew כרו correctly a moment later — "to dig out or to bore through" — but the Greek gets the gloss it needs rather than the one it has. The longer talk repeats it: the Septuagint "favors the pierced reading. It gives us pierced."⁶
• "There's about a dozen medieval manuscripts that also agree with this."⁵ BHS reports "a few," plus two reading כרו. The longer talk makes the number load-bearing in an accusation — Singer "was just wrong … either knowingly, willfully lying about it, or he just said stuff he didn't know" — while sourcing it to another video rather than to the manuscripts.⁶ A charge of lying should not rest on a number the speaker has not checked.
• "The oldest Jewish manuscript that's still in Hebrew is a Dead Sea Scrolls copy of this passage that goes back to the time before Christ."⁵ This is the load-bearing error, and the longer talk repeats it three times and builds a timeline graphic on it: the Dead Sea Scrolls reading is "a thousand years earlier than the Masoretic … we're talking before Jesus," and again, "the oldest Hebrew copy, which is the Dead Sea Scrolls before Christ."⁶ 5/6ḤevPs is dated AD 50–68.

That last error is not originally Winger's.  He reads it out of Michael Brown, who writes that "the oldest Hebrew copy of the Psalms we possess, from the Dead Sea Scrolls, dat[es] to the century before Yeshua."⁶  It is worth being precise about what is lost.  A first-century-AD Jewish scroll from a Bar Kokhba refuge cave is still meaningful evidence — it shows the reading was current among Jews with no motive to help Christians — but "before Christ" is what would make it unanswerable, and the manuscript is not that.

Neither talk mentions that the New Testament never quotes the verse.  That is both the strongest argument against his position on this clause and, as shown above, one of the better arguments against the forgery charge.  A viewer who came to these videos for the objections would not learn it.

One more refinement rather than a correction.  In the longer talk he answers the objection that "Jesus was probably tied to the tree" by producing the Giv'at ha-Mivtar heel bone, concluding that "crucifixion did involve nails, at least in certain cases."⁶  That is right, and the qualification he adds is the important part — because on Zias and Sekeles's re-examination the same skeleton shows nailed feet and, on the absence of any forearm or hand trauma, arms that were probably tied.  The one skeleton he cites answers half the objection and concedes the other half.

Tovia Singer

Singer's article is the best-known statement of the other side.⁷  He is right about several things that Christian popularisers get wrong:  ka'ari is a real word;  it does mean "like a lion" in Isaiah 38:13 and everywhere else;  ka'aru is not a Hebrew word;  כרו means "dig," not "pierce";  the New Testament never quotes the verse.  On the lexicography he has the better of the exchange.

He is wrong, or overreaching, on three points.

• The forgery accusation. "Christian translators were well aware of the correct meaning of this simple Hebrew word … and deliberately twisted their translations."⁷ The KJV translators were following the Greek and Latin, as everyone before them had. Attributing to malice what is explained by a fifteen-hundred-year-old translation chain is not an argument.
• The Septuagint claim. That the Greek Prophets and Writings are "a Christian work, doctored and edited exclusively by Christian hands" is contradicted by the prologue to Ben Sira, a Jewish document of 132 BC.¹⁰
• The Nahal Hever reading. His claim that the letter is illegible, and that if it is a waw it is "the result of the scribe's poor handwriting or spelling mistake,"⁷ is contradicted by the specialists who read a waw while arguing that it is nevertheless secondary. His supporting observation — that the same scribe misspelled ידי as ידיה — is real and is made by those specialists too, but it is an argument about this scribe's reliability, not about legibility.

His strongest single point is the one both Christian and Jewish popularisers underplay:  Psalm 22:16 is never quoted in the New Testament, and "missionaries, who insist that the Christian translation of this verse reflects the original words of King David, must wonder why there was not one New Testament author who deemed this supposed allusion to the crucifixion worthy of being mentioned."⁷  That is a fair question and it deserves a straight answer rather than a change of subject.

Michael Brown

Brown's treatment in Answering Jewish Objections to Jesus, volume 3, is the source most Christian presentations draw on, including Winger's.¹⁶  Two of his moves are sound and one is not.

Sound:  his observation that even on the lion reading, Rashi's "crushed in a lion's mouth" is not a gentle image and does not sit awkwardly beside a crucifixion.  Also sound:  his point that the Septuagint is the oldest surviving Jewish translation and reads a verb, which relocates the burden.

Not sound:  the argument that because Psalm 22:13 uses the longer אריה, the shorter ארי a few verses later is suspicious.¹⁶  Three things are wrong with it.  The two words are not in the same construction — 22:13's אריה stands alone in apposition, while 22:16's has the prefixed כְּ, and with that prefix both כארי and כאריה are standard Hebrew.  כארי occurs four other times.  And one of those four is Isaiah 38:13, with the identical bone-breaking image, which is the very verse Rashi cites.  The counter-observation takes ten minutes with a concordance and it is decisive against this particular argument.

Verdict

On the text:  the Masoretic Text probably preserves the original consonants, "pierced" is not a translation of anything in any manuscript, and neither side forged anything.

The strongest current study — by two evangelicals working from the plates and the versions — concludes that כארי is the earliest recoverable reading, that the aleph nobody can explain is deliberate, and that the poet intended a word heard simultaneously as "like a lion" and as "gouging at."  The United Bible Societies committee, working forty years earlier, graded the Masoretic reading {B} and recommended "disfiguring my hands and my feet."  Neither of those is "they pierced."

On the tampering charges:  both fail.  Justin Martyr's second-century list of Jewish deletions does not include this verse, and the dispute he records is about the psalm's referent, not its wording.  The Septuagint is pre-Christian Jewish work.  The Masoretic tradition preserved, rather than suppressed, notes pointing away from the lion reading.  The one place where a thumb appears on a scale is a Christian one, in the Complutensian Polyglot of 1514–17.

On the headline question — is this prophecy that could not have been faked?  Forgery is ruled out, but forgery was never the interesting hypothesis.  The interesting hypothesis is that a difficult Hebrew clause was guessed at in Greek two centuries before Christ, and that Christians later found the guess evocative.  That requires no dishonesty from anyone and it fits every piece of evidence, including the piece that nobody quotes:  the New Testament never uses this verse, and the Greek verb it would have needed appears in the New Testament only in parables about digging holes.

That silence is the most important fact in the dossier, and it points in a direction neither camp finds comfortable.  It tells against the apologetic, because the first Christians — holding the Greek Psalter, actively hunting for crucifixion prophecies, quoting the verse immediately after this one word for word — did not see a crucifixion here.  And it tells against the forgery charge, because a manufactured proof-text that goes entirely unused for a century is not a manufactured proof-text.

What survives, and what the argument should have been resting on all along, is the rest of the psalm:  a lament whose speaker is mocked in words a Gospel puts in the mouths of priests, whose clothing is gambled for in a line John quotes with a fulfilment formula, whose first verse Jesus cries in Aramaic from the cross, and which ends with the nations of the earth turning to God because "he has done it."  That case has to be argued on the direction of literary dependence, not on a waw and a yod — and it is a genuinely open argument, in which the strongest Christian move is Goodacre's:  that scriptural language over remembered events explains the data better than invention does.

Anyone who tells you this verse settles the matter — in either direction — is quoting one witness and suppressing the rest.

Sources

Scripture quotations are from the NET Bible unless otherwise noted.  Hebrew is quoted from the Masoretic Text (Miqra according to the Masorah / Westminster Leningrad Codex);  Greek New Testament from the SBL Greek New Testament;  Septuagint from the Rahlfs-based text.