skip to content

The Oldest Fragment of Origen's Hexapla: T-S 12.182

Benjamin Kantor

In the middle of the third century CE in Caesarea, the church father and biblical scholar Origen compiled the Hexapla (ἑξαπλᾶ ‘six-fold’), so named for its format of six parallel columns. It may in fact be the world’s first parallel Bible. The first column contained Hebrew in Hebrew letters, the second column contained Hebrew in Greek letters (i.e., transcription), the third column contained the Greek translation of Aquila, the fourth column contained the Greek translation of Symmachus, the fifth column contained a version of the Septuagint (LXX), and the sixth column contained the Greek translation of Theodotion. The original probably looked something like this (see further below for the relation of this format to T-S 12.182):




There has been scholarly debate about whether or not there really was a first column (Hebrew in Hebrew letters) as part of the Hexapla, since there are no remains of the first column in any of the extant witnesses. It is only attested in ancient authors’ descriptions of the Hexapla. On the basis of the precise measurements and proportions of this Genizah fragment (T-S 12.182), however, it has been persuasively argued that this palimpsest originally did contain a Hebrew column (see further below).


Various suggestions have been made about the original impetus and intended purpose of the Hexapla. Some suggest that its purpose was ultimately text-critical, that is, to establish the most reliable version of the LXX. Others suggest that it was motivated by apologetics. As Christians debated with Jews about messianic prophecies in the Old Testament, they found that the Greek text of the Bible did not always precisely correspond to the Hebrew text of the Bible. The Hexapla, then, would serve to alert Christian scholars of the differences between the Greek and Hebrew versions, so that they would not be ill-informed or caught off guard for such debates. Still others suggest that the Hexapla was actually intended to help Christian scholars learn Hebrew. By comparing the first and second column with the other Greek columns, the claim is that one could gradually familiarize themselves with the Hebrew language. Supporters of this view cite the bilingual Graeco-Latin columnar texts used in schools around the Roman empire to teach Greek-speaking students Latin or Latin-speaking students Greek. A closer parallel may actually be the Graeco-Babyloniaca school texts of Mesopotamia from roughly the same period, in which Akkadian is written in cuneiform on one side of the tablet and then in Greek transcription on the other. There is, however, reason to be sceptical of seeing the Hexapla as a language-learning tool.


All in all, there were probably a number of factors that led to the compilation of the Hexapla by Origen in the third century CE. Whatever his motivations were, however, his intended use has become clearer to scholars in recent years. It is probably the case, as Gentry argues, that Origen assembled the Hexapla (the synopsis) to serve as a sort of scholar’s manual to help him compose the Tetrapla, a highly annotated recension (or ἔκδοσις) of the Septuagint text, which noted interpolations, omissions, and discrepancies between the LXX, the Hebrew, and the other Greek translations. Both of these works, the Hexapla and the Tetrapla, continued to be housed at the Library of Caesarea after Origen’s death in 254 CE. The work on the Tetrapla would be continued by Pamphilus and Eusebius over the coming century.[1]


Although the Hexapla was originally a massive work, taking up tens of volumes on a library shelf, it seems to have been lost or destroyed at some point between the fourth and seventh centuries CE. The result of this, the most tragic loss for the field of Greek Bible text criticism, is that almost all of the remains of the Hexapla and/or Tetrapla have been splintered and scattered across many sources. There are thousands of quotations of the Hexapla/Tetrapla sprinkled throughout the church fathers. The margins of LXX manuscripts also often contain scholia that make mention of the various Hexaplaric renderings of a particular word or phrase. The Syro-Hexapla, a Syriac rendering of the Tetrapla from late antiquity, provides the longest and most substantial witness of this ancient text, albeit in translation.


There are actually only about four or five manuscript witnesses that are in any way “direct” witnesses to the Hexapla. I use the term “direct” loosely, meaning only that these witnesses preserve the original columnar format of the Hexapla.


This finally brings us to our fragment: T-S 12.182. This sixth- or seventh-century palimpsest happens to be the oldest direct witness to the Hexapla extant today. It contains a portion of Psalm 22 (vv. 15–18 and 20–28). We do not know much about the history of the text, but it was probably an early copy of the Hexapla (or the Hexaplaric Psalter) that circulated in Christian circles in Palestine and/or Egypt in the latter half of the first millennium. Eventually, at least, it found its way to Egypt, whether before or after being written over with the Hebrew liturgical poetry of Yannai in the tenth century. If the Hexapla’s loss or destruction occurred at the uttermost end of the range given above (4th–7th c. CE), it remains possible, albeit a longshot, that this fragment may actually have been copied from the original or a copy quite close to the original. There are indeed some who have suggested that the Hexapla was not actually destroyed until the Islamic conquest of 638 CE.


In terms of analyzing the original format of the Hexapla, this is the most important witness we have. Although only portions of columns II (Hebrew in Greek letters), III (Aquila), IV (Symmachus), and V (LXX) remain, the proportions and measurements of the fragment, according to Jenkins, prove that this palimpsest originally contained both column I (Hebrew in Hebrew letters) and column VI (Theodotion). This is partially due to the fact that you can still see the original gutter (i.e., “crease”) that marked the centre of the codex.[2] As noted earlier, the original format of the codex probably looked like this:

title of image here


title of image here



When it was being prepared for reuse in the tenth century for the Hebrew liturgical poetry of Yannai, it was cut so that one side (fol. A verso and fol. B recto) contained only column III, column IV, and a portion of column V:


Hexapla with Genizah manuscript overlay front



The other side (fol. B verso and fol. A recto), then, contained only the remains of the furthest right edge of column II, column III, and column IV:

Hexapla with Genizah manscript overlay back



The back side of the palimpsest is especially important due to its preservation of part of the second column, which is the most sparsely attested of all the extant columns of the Hexapla across the spectrum of witnesses. Although only several letters remain per line, by measuring the average width of each particular letter in this scribal hand and determining a sort of “plumb-line” for the left edge of the column based on a word whose reconstruction is rather certain, we can actually arrive at a tentative reconstruction of the rest of the column (Ps. 22.25–28; Tiberian realization in parentheses):

reconstructed Hexapla, 2nd column


A detailed explanation of the rationale for all of these changes is outlined in my forthcoming critical edition of the second column of the Hexapla.[3] Even from the highly fragmentary remains of the second column, we can still notice several interesting linguistic features. For example, the 2MSG suffix ends in a consonant (-αχ) rather than a vowel as in Tiberian (ךָ). This, of course, has parallels in Rabbinic Hebrew and other ancient transcriptions. Also, the ending of the infinitive form וּֽבְשַׁוְּע֖וֹ is transcribed with -αυεω, which suggests that Greek script often required the addition of a vowel (in this case ε) either to preserve the syllable structure or to best approximate the guttural /ʕ/ sound (note that shewa is normally silent in this position in the second column).


All in all, it is difficult to overestimate the importance of this manuscript for the study of the compositional history of the Hexapla. Though scholars have been aware of its existence for over a century, it continues to yield new insights even to this day.

View T-S 12.182 on Cambridge Digital Library


[1] Peter Gentry, “Origen’s Hexapla,” in The Oxford Handbook of the Septuagint (ed. A. Salvesen and T.M. Law; Oxford: Oxford University Press, forthcoming).

[2] R.G. Jenkins, “The First Column of the Hexapla: The Evidence of the Milan Codex (Rahlfs 1098) and the Cairo Genizah Fragment (Rahlfs 2005),” in Origen’s Hexapla and Fragments: Papers Presented at the Rich Seminar on the Hexapla, Oxford Centre for Hebrew and Jewish Studies, 25th July–3rd August 1994 (ed. Alison Salvesen; Tübingen: Mohr Siebeck, 1998), 88–102.

[3] Benjamin Kantor, τὸ ἑβραϊκόν | TO HEBRAIKON: A Critical Edition of the Second Column (forthcoming)


Gentry, Peter. Forthcoming. “Origen’s Hexapla.” In The Oxford Handbook of the Septuagint. Edited by A. Salvesen and T.M. Law. Oxford: Oxford University Press.

Jenkins, R.G. 1998. “The First Column of the Hexapla: The Evidence of the Milan Codex (Rahlfs 1098) and the Cairo Genizah Fragment (Rahlfs 2005).” In Origen’s Hexapla and Fragments: Papers Presented at the Rich Seminar on the Hexapla, Oxford Centre for Hebrew and Jewish Studies, 25th July–3rd August 1994. Edited by Alison Salvesen. Tübingen: Mohr Siebeck. Pages 88–102.

Kantor, Benjamin. Forthcoming. τ βραϊκόν | TO HEBRAIKON: A Critical Edition of the Second Column (Secunda) of Origen’s Hexapla.

Sokoloff, Michael and Joseph Yahalom. 1978. “Christian palimpsests from the Cairo Genizah.” Revue d’Histoire des Textes 8: 109–32.

Taylor, Charles. Hebrew-Greek Cairo Genizah palimpsests from the Taylor-Schechter collection. Cambridge: 1900.


Cite this article

Kantor, B. (2019). The Oldest Fragment of Origen's Hexapla: T-S 12.182. [Genizah Research Unit, Fragment of the Month, March 2019].

If you enjoyed this Fragment of the Month, you can find others here.

The manuscripts found in this article is included in the Cambridge Digital Library. To see these and other items visit

Contact us:

The zoomable images are produced using Cloud Zoom, a jQueryimage zoom plugin:

Cloud Zoom, Copyright (c) 2010, R Cecco,

Licensed under the MIT License