A Few Remarks on the Problem of Psalm 22:16

In the comments section of my page on mistranslation in the NIV, frequent contributor John Kesler recently suggested that the NIV’s translation of Psalm 22:16 (Hebrew 17) and its associated footnote were incorrect. Dr. David Instone-Brewer, a member of the NIV translation committee, replied with a comment of his own defending the NIV against some of John’s remarks.

A copious amount of literature has been written on this very verse, not only because it is apparently corrupt and uncertain in meaning, but also because of its importance in Christian interpretation. I have by no means read all this literature, but after perusing most (if not all) of the articles published in the last decade, I believe I can summarize the problem and provide some suggestions on what might be an acceptable English translation approach.

I am not a Hebrew scholar, so any mistakes in what follows are purely my own.

Psalm 22: An Overview

Psalm 22 is a lament psalm in three sections; the first two describe the miserable situation the psalmist finds himself in, and the last section praises Yahweh for rescuing him.

It seems to have taken on particular significance among Christians in the first century. Although it is not a messianic psalm, the Gospels (and especially Mark) make numerous allusions to it in the story of the crucifixion. Many Christians have therefore understood the psalm to be a sort of prophecy about Jesus.

Verse 16: The Main Textual Problem

Verses 16–17 (Hebrew 17–18) are translated fairly literally by the Complete Jewish Bible (CJB) as follows:

Dogs are all around me,
a pack of villains closes in on me
like a lion [at] my hands and feet.

I can count every one of my bones,
while they gaze at me and gloat.

The main difficulty is the italicized phrase like a lion. This is the literal translation of the Hebrew word כארי (ka’ari) that appears here in almost all Masoretic Hebrew manuscripts. However, this text is generally regarded by scholars as unsatisfactory for several reasons:

1. Contextually and syntactically, the phrase ought to have a verb there instead of a noun.
2. The relationship of “hands and feet” with the rest of the sentence is ambiguous and hard to explain. How are the author’s hands and feet “like a lion”? The CJB attempts to improve the sentence by adding the preposition “at”, but that is not in the Hebrew.
3. Our Masoretic manuscripts are all from the Medieval period or later, and early textual traditions in other languages (Greek, Latin, and Syriac) vary widely in their translations; only the Aramaic Targum has the word “lion” there. This confusion suggests that the Hebrew text became corrupt and ambiguous in meaning at an early stage.

Verse 16 in the Greek Septuagint

The Septuagint (LXX) has an odd translation that is quite different: “They have dug my hands and feet.” It is generally believed that the translator must have had a Hebrew original (Vorlage) that read כארו (ka’aru), which is very similar to the Masoretic כארי (the letters waw and yod are easily confused). The problem is that כארו doesn’t actually mean anything in Hebrew; but it does look a bit like a misspelling of כרו (karu), which would be the perfect tense of כרה (karah), meaning “to dig” (as in “to dig a well”). The translator thus assumed that the word meant “they have dug”. Early Christians, who used the LXX almost exclusively, were soon interpreting this passage as a prophecy about the crucifixion of Christ, whose hands and feet, according to John, were pierced with nails. (Never mind that digging and piercing aren’t quite the same thing.) Mainly for that reason, most Christian translations up until recently translated this phrase as “they pierced my hands and feet”, even though Hebrew manuscript support is flimsy, as we shall see.

Masoretic Hebrew Variants

A small minority of Masoretic (medieval) manuscripts read ka’aru instead of kaari—hence the assumption that this is what the LXX translator had before him. However, there are at least three problems with this variant.

1. As noted, its meaning is unclear. If it is supposed to be a conjugation of the root כרה (karah, “to dig”), then it should not have the aleph (א).
2. It is unlikely that a word used for digging pits would be used to describe the piercing of someone’s hands and feet with nails. Hebrew has several words that would be more appropriate if “pierce” were intended.
3. Crucifixion is not appropriate to the context of Psalm 22 or its historical setting.

There is also a very rare variant כרו (karu, the correct way of writing “they have dug”), but by the textual-critical principle of lex difficilior (“the more difficult variant is to be preferred”), this is almost certainly a late Medieval correction to the text.

Other Hebrew Variants

Just one fragmentary copy of Psalm 22 has been identified among the Dead Sea Scrolls at Qumran. Unfortunately, this document, known as 4QPs^f, becomes illegible at precisely this location in the text. The faded remnants of what may be כר (kr) are visible, but nothing definitive can be learned.

A somewhat later fragment called 5/6HevPs from a cave at Nahal Hever appears to read כארו (ka’aru), but several scholars who have examined the only published facsimile say the ink is too faded to be certain. (See citations for Swenson and Strawn below.)

Nahal Hever, where a cave containing a small cache of early-second-century Jewish documents was discovered

Other Greek, Latin, and Aramaic Translations

Other ancient translations add to the confusion. The first Greek translation by Aquila (2nd century CE) read “they have disfigured”, while his second translation read “they have bound.” The Greek translation by Symmachus read “like those who seek to bind”. Jerome translated it initially into Latin as foderunt, “they have pierced”, in his Gallican Psalter (c. 387), but later in his Psalterium iuxta Hebræos (c. 391), he changed it to vinxerunt, “they have bound”—presumably having found a different or better Hebrew source.

Some Modern Proposals

“To bind” — Gregory Vall (Ave Maria University) proposes that the text originally said אסרו (’asaru), which means “they have bound”. He theorizes that an inadvertent swapping of two letters resulted in the meaningless סארו, which was then interpreted as כארו, and eventually “corrected” to כארי, as most extant manuscripts now read.

The advantage of this view is that it would explain why Aquila, Symmachus, and Jerome all translated it as a verb meaning “to bind”.

Source: Gregory Vall, Psalm 22:17B: “The Old Guess”, JBL 116/1 (1997)

“To become short/shriveled” — Michael L. Barré (St. Mary’s Seminary and University) recently revived an earlier proposal by J.J.M. Roberts, who proposed an otherwise unattested Hebrew verb כרי (kari) meaning “to become short/shriveled” on the basis of similar words found in Akkadian and Syriac. Barré concurs that not only did such a verb exist, but that כארו (ka’aru) would be a valid alternative spelling in late Hebrew.

Barré marshals additional evidence from Babylonian medical texts as well a Dead Sea Scroll (1QH) that contains similar language to Psalm 22. He concludes that the psalmist means to say his arms and legs are paralyzed and unable to move.

He proposes that the problem is compounded by a misunderstanding of the following line, “I can count every one of my bones”—a statement that is difficult to explain in context. The Syriac translation reads “all my bones have wailed”, and Barré concludes that the Hebrew verb in this line was corrupted at some point. He interprets this line, properly corrected, to express a lament on behalf of the psalmist’s impending death: “All my bones have intoned my funeral dirge.” He cites further parallels with Babylonian texts and shows how this makes greater sense of the overall psalm: the psalmist is surrounded by attackers, unable to defend himself (with his hands) or run away (with his feet), so he is bemoaning his impending death.

The advantage of this view is that is based on a known variant (ka’aru) and makes more sense of the passage than any other interpretation I have seen. The disadvantages are its need to propose an unattested verb and to emend the following line.

Sources:
• J. J. M. Roberts, “A New Root for an Old Crux, Ps. XXII 17c,” Vetus Testamentum 23/2 (1973)
• Michael L. Barré, “The Crux of Psalm 22:17c: Solved at Long Last?” Bernard F. Batto et al. (eds.), David and Zion; Eisenbrauns, Winona Lake, Ind. (2004)

“To pick/pluck” — Two earlier scholars, R. Tournay and M. Dahood, argued that the word was derived from ארה (’arah, “to pick, pluck”) with a preposition attached. In other words, the psalmist’s body has been “picked clean” by the “dogs” who encircle him, exposing his bones. More recently, James Linville has taken up the case, arguing not only that this is the correct reading, but also that the author has chosen a word that sounds like “lion” in Hebrew as a deliberate pun, since lions and other wild animals are mentioned elsewhere.

Source: James R. Linville, “Psalm 22:17b: A New Guess”, JBL 124/4 (2005)

The “Lion” Reading Reconsidered

A few scholars have also recently lent support for the MT’s reading of “like a lion”. Brent A. Strawn has analyzed the chiastic structure of Psalm 22 and believes that a reference to a lion in v. 16 fits well. He proposes the addition of a missing verb, טרף (taraf), meaning “to tear”.

Kristin M. Swenson believes v. 16, which is currently divided into three phrases, should be a bicolon (two-part verse). Putting the divider in its correct location produces the following text:

Dogs surround me, a pack of wicked ones. Like a lion, they circumscribe my hands and feet.

Sources:
• Brent A. Strawn, “Psalm 22:17b: More Guessing”, JBL 119/3 (2000)
• Kristin M. Swenson, “Psalm 22:17: Circling around the Problem Again”, JBL 123/4 (2004)

Consensus amidst Discord

These examples are just a few of the ways in which scholars reconstruct Psalm 22:17 in order to make sense of it. While there is no consensus as to what the original reading was—and barring a new manuscript find, there may never be one—a few general observations can be drawn.

1. Most, though not all, critical scholars believe the majority Masoretic text is defective on syntactic grounds and probably not original.
2. There is widespread agreement among critical scholars (including all those surveyed here) that Psalm 22 is not a depiction of crucifixion, and that the Septuagint’s translation of “dig” is incorrect. It is best explained as a misunderstanding of the word כארו, which may or not have been the original Hebrew reading.
3. Most solutions require emending the passage, and this must be done in a way that is sensitive to the passage’s poetic structure and the meaning of v. 16 in its larger context.

English Translations of Psalm 22

Historically, most Christian, church-oriented translations of Psalm 22 have deviated from the Masoretic text in this verse, preferring instead a christological interpretation inspired by the Septuagint.

KJV: For dogs have compassed me: the assembly of the wicked have inclosed me: they pierced my hands and my feet.

RSV: Yea, dogs are round about me; a company of evildoers encircle me; they have pierced my hands and feet—

NASB: For dogs have surrounded me; A band of evildoers has encompassed me; They pierced my hands and my feet.

Recently, more ecumenical and academically-oriented translations have approached passages like Ps 22:16 with a clean slate. The NRSV, for example, adopts Roberts’ interpretation as the best one while stating in a footnote that the meaning of the Hebrew is uncertain:

For dogs are all around me; a company of evildoers encircles me. My hands and feet have shriveled;

The Jerusalem Bible gives more weight to the witness of Jerome, Aquila, and Symmachus:

A pack of dogs surrounds me, a gang of villains closes me in; they tie me hand and foot…

The recently completed Common English Bible (CEB) adheres to the majority Hebrew:

Dogs surround me; a pack of evil people circle me like a lion—oh, my poor hands and feet!

What should we consider to be a correct translation versus an incorrect one, if we cannot confidently reconstruct and interpret the original Hebrew? In part, it comes down to the stated intent of the translation. If a translation claims to be based on the Hebrew Masoretic text, then “like a lion” is probably the best translation. If a translation’s goal is to provide a “best guess” at what the original author intended, then there are many scholarly reconstructions that would suffice, and explanatory footnotes are in order as well. I regard the latter three examples (the NRSV, JB, and CEB) all to be adequate translations, though they differ completely from each other.

Psalm 22:16 in the NIV

The NIV purports to be based on the Hebrew Masoretic Text, though its translators also consult other ancient versions in places where the MT “seems doubtful”:

For the Old Testament the standard Hebrew text, the Masoretic Text as published in the latest editions of Biblia Hebraica, has been used throughout. …The translators also consulted the more important early versions—the Greek Septuagint, Aquila, Symmachus and Theodotion, the Latin Vulgate, the Syriac Peshitta, the Aramaic Targums, and for the Psalms the Juxta Hebraica of Jerome. …these versions were occasionally followed where the Masoretic Text seemed doubtful and where accepted principles of textual criticism showed that one or more of these textual witnesses appeared to provide the correct reading. …These departures from the Masoretic Text are also indicated in the textual footnotes. (Quotation from the preface of the NIV, 2011 edition.)

In Psalm 22:16, the NIV has not followed the standard Masoretic Text, even though Jewish translations and some English translations find it acceptable.

Dogs surround me, a pack of villains encircles me; they pierce my hands and my feet.

If the translators felt the MT was dubious, then fair enough. But did they choose in its place a textual witness that “appeared to provide the correct reading”, or did they simply acquiesce to Christian tradition and theological bias of what the verse should say?

Three of the sources they mention—Aquila, Symmachus, and Jerome—interpret the psalmists hands and feet as being “bound”, without any reference to crucifixion or piercing. The Aramaic Targum supports the Masoretic reading of “lion”. Even the Septuagint can only be vaguely construed as meaning “pierced”. I would say that the NIV, having abandoned the Hebrew text, has chosen the least viable of the alternative interpretations available.

The NIV redeems itself somewhat by including a footnote:

Dead Sea Scrolls and some manuscripts of the Masoretic Text, Septuagint and Syriac; most manuscripts of the Masoretic Text me, / like a lion.

This footnote is helpful but perhaps also a bit misleading.

1. The “Dead Sea Scrolls” are generally taken to mean the scrolls found at Qumran, all dating from the centuries before the Jewish War. No such manuscript supports the NIV’s reading. Presumably, the Nahal Hever fragment is meant, but this dates to a later period and its reading is still inconclusive.
2. It is a bit misleading to say that “some” MT manuscripts support this reading. In fact, it is only two manuscripts that say “dug” (not “pierced”). The reader is likely to infer better manuscript support than there actually is.
3. The Septuagint says “dug”, not exactly “pierced”.

To its credit, the NIV footnote at least mentions the majority Hebrew reading.

I have been reluctant to add this verse to my list of bad translations, since the MT is problematic and there is genuine room for different interpretations. However, I think it falls short of being accurate and unbiased for all the reasons given above.