From BHK to BHS and Back to the Codices
Abstract
The opening verse of 1 Chronicles 10 presents a minor but instructive discrepancy in the transmission of the Tiberian accentuation. Biblia Hebraica Stuttgartensia (BHS) prints two adjacent occurrences of Atnaḥ in the sequence בְיִשְׂרָאֵ֑ל וַיָּ֑נָס, whereas earlier editions of Biblia Hebraica, several modern Masoretic editions, Miqraʾ ʿal pi ha-Masora (MAM), and direct evidence from Codex Leningradensis support בְיִשְׂרָאֵ֑ל וַיָּ֤נָס, with Mahapakh on וַיָּנָס. The error was subsequently reproduced in some electronic Hebrew texts before being corrected through renewed consultation of the Leningrad manuscript. This note traces the reading through printed editions, manuscript evidence, and the digital textual tradition, while identifying an unresolved bibliographical question concerning the first 1975 BHS fascicle of Chronicles.
1. The Problem
The opening of 1 Chronicles 10:1 in Biblia Hebraica Stuttgartensia reads:
וּפְלִשְׁתִּ֖ים נִלְחֲמ֣וּ בְיִשְׂרָאֵ֑ל וַיָּ֑נָס אִֽישׁ־יִשְׂרָאֵל֙
The sequence relevant to the present discussion is:
בְיִשְׂרָאֵ֑ל וַיָּ֑נָס
BHS thus places Atnaḥ (אַתְנָח) both on בְיִשְׂרָאֵל and on the immediately following וַיָּנָס. The reading is conspicuous from the standpoint of the normal hierarchy of the Tiberian prose accents. Atnaḥ is one of the principal disjunctives of the verse, whereas Mahapakh (מְהֻפָּךְ; also transcribed mahapakh, mehuppach, etc.) normally functions as a conjunctive.[1]
The accentual anomaly by itself would warrant investigation, but it does not constitute sufficient evidence for emendation. The more important question is documentary: what do the relevant textual witnesses actually read?
The alternative form, found in numerous other editions, is:
בְיִשְׂרָאֵ֑ל וַיָּ֤נָס
The difference is very small graphically, but its history proves unexpectedly interesting.
2. BHK before the Stuttgartensia
Rudolf Kittel’s first Biblia Hebraica, published in 1906, prints:
בְיִשְׂרָאֵ֑ל וַיָּ֤נָס
with Atnaḥ on בְיִשְׂרָאֵל and Mahapakh on וַיָּנָס.[2]
[Figure 1. BHK, 1906, p. 1241: 1 Chr 10:1.]
The evidential significance of this edition must be qualified, however. Kittel’s first and second editions did not reproduce Codex Leningradensis as their base text. They used the received Masoretic text represented by the second Rabbinic Bible of Jacob ben Ḥayyim, published by Daniel Bomberg in 1524–25. It was only with the thoroughly revised third edition, completed in 1937, that the Biblia Hebraica adopted Codex Leningradensis B19A as its base manuscript. (Die Bibel)
The 1937 edition is therefore considerably more important for the present question. If the second Atnaḥ reflected Leningrad, one might expect it to appear when the Biblia Hebraica first changed its base text to that manuscript.
It does not.
BHK3 reads:
בְיִשְׂרָאֵ֑ל וַיָּ֤נָס.[3]
[Figure 2. BHK3, 1937, p. 1345.]
The significance of this observation is straightforward: the second Atnaḥ did not enter the Biblia Hebraica tradition as a consequence of the 1937 adoption of Codex Leningradensis. The first Leningrad-based BHK prints Mahapakh.
A 1954 BHK reprint that I also examined preserves the same reading:
בְיִשְׂרָאֵ֑ל וַיָּ֤נָס.[4]
[Figure 3. BHK, 1954: 1 Chr 10:1.]
This is consistent with the history of BHK3’s typesetting. The German Bible Society notes that the BHK3 setting could subsequently be altered only to a limited extent and that later reprints based on it continued into the mid-1970s. (Die Bibel)
The same reading is found in a considerably broader range of Hebrew Bible editions that I consulted, including editions associated with Letteris, Ginsburg, Snaith, Koren, Breuer, ArtScroll, Dotan, JPS, and the Jerusalem Crown.[5] These editions are, of course, neither textually independent of one another in every respect nor witnesses of equal weight. Their agreement should therefore not be treated as a simple numerical vote. Nevertheless, they demonstrate how widespread and stable the Mahapakh reading has been in the printed Masoretic tradition.
3. The Change in BHS
The 1977 one-volume Biblia Hebraica Stuttgartensia presents a different reading:
בְיִשְׂרָאֵ֑ל וַיָּ֑נָס.[6]
[Figure 4. BHS, 1977, p. 1480.]
The second Mahapakh of the earlier BHK tradition has become an Atnaḥ.
This is more significant than an ordinary difference between two printed Hebrew Bibles. BHS, like BHK3, was intended as a diplomatic presentation of Codex Leningradensis. The German Bible Society describes BHS as having been prepared between 1967 and 1977 and states that its Masoretic text is reproduced from the Leningrad Codex. (Die Bibel)
The question is therefore not whether BHS follows a different editorial theory of accentuation, but whether its second Atnaḥ actually represents what is written in its base manuscript.
It does not.
4. Codex Leningradensis
The most decisive evidence comes from renewed examination of Leningrad itself.
The UXLC change record contains an entry for 1 Chronicles 10:1.4, dated 22 February 2021 and supplied by Stephen Salisbury of the Groves Center. The note states:
“Change the etnachta on the yod to a mahapakh.”
The manuscript location is given as folio 333B, column 2, line 7, and the correction is displayed explicitly as:
וַיָּ֑נָס → וַיָּ֤נָס.[7]
The change log further identifies this group of corrections as having been provided from WLCU 4.22. (Tanach)
This is particularly strong evidence because the correction does not depend upon another printed edition. It represents a renewed transcriptional judgment based upon the manuscript itself.
Codex Leningradensis therefore supports:
בְיִשְׂרָאֵ֑ל וַיָּ֤נָס
and not the double-Atnaḥ reading of BHS.
The 2021 date should be interpreted carefully. It securely dates the UXLC record and incorporation of the correction; it should not automatically be taken as proof that the Groves Center first discovered or adopted the correction on that exact date. The underlying WLCU 4.22 data may have its own editorial history.
That qualification notwithstanding, the textual result is unambiguous: BHS does not accurately reproduce Leningrad at this point.
5. The Aleppo Codex and MAM
The Aleppo Codex provides an important independent control.
Chronicles survives complete in Aleppo. The documentation for Miqraʾ ʿal pi ha-Masora explicitly states that, where the Aleppo Codex survives in the Prophets and Writings, the edition bases its consonantal text, vocalization, accents, and related details directly upon it. For Chronicles specifically, MAM states that the book is complete in Aleppo and that its edition is based upon it, though Leningrad is also checked at numerous locations. (Wikisource)
MAM prints 1 Chronicles 10:1:
וּפְלִשְׁתִּ֖ים נִלְחֲמ֣וּ בְיִשְׂרָאֵ֑ל וַיָּ֤נׇס אִֽישׁ־יִשְׂרָאֵל֙ …[8] (Wikisource)
Again, the accent on וַיָּנָס is Mahapakh.
Inspection of the Aleppo photographs adds an interesting codicological detail. First Chronicles 10:1 falls across the transition between folios 211a and 211b. MAM’s Aleppo index assigns 1 Chr 9:13–10:1 to 211a and 1 Chr 10:1–11:11 to 211b. (Wikisource)
Thus the two words at issue are effectively separated by the folio transition.
[Figure 5a. Aleppo Codex, fol. 211a: …נלחמו בישראל.]
[Figure 5b. Aleppo Codex, fol. 211b: וינס איש ישראל….]
The photographed sign on the continuation is considerably fainter than one would wish for independent palaeographical analysis from a reproduction alone. Accordingly, the most prudent formulation is that the photograph is consistent with, while MAM’s direct transcription explicitly gives, Mahapakh. MAM does not treat 1 Chronicles 10:1 as an exceptional or doubtful case in its documentation of the surviving Aleppo text.
The evidence from the two principal Tiberian codices therefore converges:
| Witness | Accent on וַיָּנָס |
|---|---|
| Aleppo Codex as represented by MAM | Mahapakh ֤ |
| Codex Leningradensis | Mahapakh ֤ |
| BHK 1906 | Mahapakh ֤ |
| BHK3 1937 | Mahapakh ֤ |
| BHK reprint, 1954 | Mahapakh ֤ |
| Breuer, Dotan, Koren, Jerusalem Crown, etc. | Mahapakh ֤ |
| BHS 1977 | Atnaḥ ֑ |
This distribution makes it very difficult to regard the BHS form as evidence for a competing Masoretic accentual tradition. The most economical explanation is that the second Atnaḥ in BHS arose as a transcriptional, editorial, or typographical error in the production history of that edition.
6. Accentual Considerations
The accentual syntax independently supports the manuscript evidence.
The opening of the verse may be represented:
וּפְלִשְׁתִּ֖ים נִלְחֲמ֣וּ בְיִשְׂרָאֵ֑ל | וַיָּ֤נָס אִֽישׁ־יִשְׂרָאֵל֙ …
Atnaḥ effects the major internal division after בְיִשְׂרָאֵל. The clause beginning with וַיָּנָס then belongs to the second principal portion of the verse, and Mahapakh serves conjunctively within the ensuing accentual structure.[9]
By contrast,
בְיִשְׂרָאֵ֑ל וַיָּ֑נָס
places another principal disjunctive immediately after the first and does not correspond to the normal hierarchical analysis of the verse.
This observation should remain secondary to the manuscript evidence. One should not “correct” a manuscript merely because its accentuation appears unusual. In the present case, however, accentual syntax and documentary evidence point in the same direction.
7. Transmission into Electronic Hebrew Texts
The case becomes especially instructive in the history of electronic editions.
An older Westminster text—Westminster Hebrew Morphology 4.18, as preserved in the Logos resource examined during this study—still displays:
וַיָּ֑נָס
That is, the electronic text reproduces the BHS reading.
The current Accordance module Hebrew Bible Tagged (Leningrad Codex), by contrast, identifies its underlying text as Groves-Wheeler Westminster Hebrew Morphology 4.22 and displays Mahapakh at 1 Chronicles 10:1:
וַיָּ֤נָס
Accordance describes the current 4.22-based module as including documented divergences from BHS and BHQ “in favor of the manuscript.” (Accordance)
[Figure 6. Westminster 4.18 in Logos and Westminster 4.22 in Accordance.]
The UXLC change record supplies the documentary bridge between these forms: the older וַיָּ֑נָס is explicitly replaced with וַיָּ֤נָס after consultation of Leningrad. (Tanach)
This provides a useful caution regarding digital nomenclature. A resource labelled “Leningrad Codex” need not represent a wholly fresh transcription of every graphic feature of the manuscript. Electronic texts possess their own transmission histories. They inherit earlier transcriptions, editorial judgments, and occasionally errors; those data may subsequently be corrected as manuscripts are re-collated.
The history at 1 Chronicles 10:1 may therefore be schematized:
Leningrad → BHK3 correctly transcribed → BHS error → older electronic descendants → renewed manuscript collation → corrected electronic text.
That sequence should not be understood as implying that every electronic text passed through precisely the same pipeline, but it describes the documentary relationship evident in the witnesses examined here.
8. The Remaining Bibliographical Question
The textual question is, in my judgment, essentially settled. The bibliographical question is not.
We can presently establish two secure chronological points:
1954 BHK: וַיָּ֤נָס
1977 BHS: וַיָּ֑נָס
The error therefore entered the Biblia Hebraica tradition somewhere between the later BHK reprints and the completed BHS.
The most important missing witness is the independently published BHS fascicle of Chronicles. The Society for Old Testament Study’s 1976 Book List records:
Biblia Hebraica Stuttgartensia. Editio Funditus Renovata. Ed. K. Elliger and W. Rudolph. Fasc. 15. Liber Chronicorum, praeparavit W. Rudolph. 1975. Pp. x, 116.
(The Society for Old Testament Study)
Library catalogues also distinguish the 1975 Liber Chronicorum from an editio secunda emendata, prepared through the work of W. Rudolph and H. P. Rüger.[10] (NIIGATA University)
That creates a particularly interesting test.
If the original 1975 fascicle already reads:
וַיָּ֑נָס
then the error was introduced during the original preparation of the BHS text of Chronicles.
If the first issue instead reads:
וַיָּ֤נָס
while the editio secunda emendata or the completed 1977 BHS reads Atnaḥ, then the error must have entered during a later stage of revision or production.
Until the relevant page of the original 1975 fascicle can be examined, the precise point of entry remains unknown.
This distinction is important methodologically. We can identify the BHS reading as erroneous without pretending that we have yet reconstructed every stage by which the error arose.
9. Conclusion: Ad fontes
The accentuation of one word in 1 Chronicles 10:1 provides a compact case study in the relationships among manuscripts, printed critical editions, and electronic biblical texts.
No consonantal variant is involved. No difference in lexical meaning is at stake. The translation of the verse is unaffected. The problem concerns a single cantillation sign.
Yet this very small discrepancy illustrates several principles of broader significance.
A diplomatic edition, however carefully prepared, remains a human transcription of a manuscript. The statement that BHS is “based on Codex Leningradensis” does not entail that every sign in BHS must therefore be present in Leningrad. At 1 Chronicles 10:1, direct manuscript evidence demonstrates otherwise.
Nor does the title of an electronic resource guarantee that every detail has been freshly transcribed from the named manuscript. Digital texts, like printed ones, have genealogies. They may preserve errors from earlier stages and may subsequently be corrected through renewed collation.
In the present case the investigation moved successively from a modern critical edition, through earlier printed editions, through electronic transcriptions, and ultimately back to the medieval codices themselves.
The result is clear:
בְיִשְׂרָאֵ֑ל וַיָּ֤נָס
Atnaḥ on בְיִשְׂרָאֵל; Mahapakh on וַיָּנָס.
The historical question of precisely when the erroneous second Atnaḥ entered BHS remains to be answered. But the textual question illustrates in miniature the principle from which this blog takes its name:
Ad fontes—back to the sources.
Notes
[1] On the structure and hierarchy of the Tiberian accents, see William Wickes, Two Treatises on the Accentuation of the Old Testament: On Psalms, Proverbs, and Job; On the Twenty-One Prose Books (Oxford: Clarendon Press, 1887); and Israel Yeivin, Introduction to the Tiberian Masorah, trans. and ed. E. J. Revell, Masoretic Studies 5 (Missoula, MT: Scholars Press, 1980). Terminology varies considerably in English-language literature: אַתְנָח may appear as atnaḥ, athnach, or etnaḥta, and מְהֻפָּךְ as mahapakh, mahpak, or mehuppach. I use Atnaḥ and Mahapakh here except when quoting another source.
[2] Rudolf Kittel, ed., Biblia Hebraica, 2 vols. (Leipzig: J. C. Hinrichs, 1906), 2:1241. The German Bible Society confirms both the 1906 publication and the use of the Ben Ḥayyim textus receptus as the base of the first edition. (Die Bibel)
[3] Rudolf Kittel, ed., Biblia Hebraica, 3rd ed., completed by A. Alt and O. Eissfeldt, Masoretic text prepared by P. Kahle (Stuttgart: Württembergische Bibelanstalt, 1937), 1345. On BHK3’s adoption of Codex Leningradensis and its diplomatic editorial principle, see the historical account of the Deutsche Bibelgesellschaft. (Die Bibel)
[4] Biblia Hebraica (Stuttgart: Württembergische Bibelanstalt, 1954), 1 Chr 10:1. The copy examined is a later BHK reprint derived from the BHK3 tradition. The precise numbering of the postwar BHK impressions should be cited from the title page of the individual copy rather than inferred merely from its date.
[5] Editions consulted in addition to BHK include: Meir Letteris, Hebrew Bible (London, 1852), 1290; Christian D. Ginsburg, Hebrew Bible (London, 1894; later reprint), 1698; Norman H. Snaith, ed., Hebrew Bible (London, 1958), 1274; Koren Tanakh (Jerusalem, 1966); Mordechai Breuer, Tanakh (Jerusalem, 1989); ArtScroll Tanach (Brooklyn, 1996), 1898; Aron Dotan, Biblia Hebraica Leningradensia (Peabody, MA: Hendrickson, 2001), 1146; JPS Hebrew-English Tanakh, 2nd ed. (Philadelphia: Jewish Publication Society, 2004), 1911; and Jerusalem Crown (Jerusalem, 2004), 819. These witnesses have different editorial genealogies and are therefore cited as corroborating evidence rather than treated as nine independent manuscript witnesses.
[6] Karl Elliger and Wilhelm Rudolph, eds., Biblia Hebraica Stuttgartensia (Stuttgart: Deutsche Bibelstiftung, 1977), 1480. The Deutsche Bibelgesellschaft describes BHS as a 1967–1977 revision and as a diplomatic edition reproducing the Masoretic text of Codex Leningradensis. (Die Bibel)
[7] “Changes Made from UXLC 1.1 to UXLC 1.2,” entry 2021.02.22–27, 1 Chr 10:1.4, UXLC / tanach.us. The entry gives “Folio_333B Column: 2 Line: 7,” identifies Stephen Salisbury as the contributor, and explicitly changes וַיָּ֑נָס to וַיָּ֤נָס. (Tanach)
[8] Miqraʾ ʿal pi ha-Masora (MAM), 1 Chr 10:1, Hebrew Wikisource. The accented text reads בְיִשְׂרָאֵ֑ל וַיָּ֤נׇס. (Wikisource) On MAM’s use of the Aleppo Codex for the surviving Prophets and Writings, see its methodological introduction. (Wikisource)
[9] See Wickes, Two Treatises, and Yeivin, Introduction to the Tiberian Masorah. The accentual argument here is corroborative rather than constitutive: the manuscript evidence establishes the reading independently of theoretical expectations regarding accent hierarchy.
[10] Biblia Hebraica Stuttgartensia, fasc. 15, Liber Chronicorum, praeparavit W. Rudolph (Stuttgart: Württembergische Bibelanstalt, 1975), x + 116 pp. The original issue is independently catalogued, while a parallel catalogue record identifies an Editio secunda emendata / opera W. Rudolph et H. P. Rüger. (The Society for Old Testament Study)
Selected Bibliography
Elliger, Karl, and Wilhelm Rudolph, eds. Biblia Hebraica Stuttgartensia. Stuttgart: Deutsche Bibelstiftung, 1977.
Kittel, Rudolf, ed. Biblia Hebraica. 2 vols. Leipzig: J. C. Hinrichs, 1906.
Kittel, Rudolf, ed. Biblia Hebraica. 3rd ed. Masoretic text prepared by Paul Kahle; edition completed by Albrecht Alt and Otto Eissfeldt. Stuttgart: Württembergische Bibelanstalt, 1937.
Rudolph, Wilhelm, ed. Liber Chronicorum. Biblia Hebraica Stuttgartensia, fasc. 15. Stuttgart: Württembergische Bibelanstalt, 1975.
Dotan, Aron. Biblia Hebraica Leningradensia. Peabody, MA: Hendrickson, 2001.
Wickes, William. Two Treatises on the Accentuation of the Old Testament: On Psalms, Proverbs, and Job; On the Twenty-One Prose Books. Oxford: Clarendon Press, 1887.
Yeivin, Israel. Introduction to the Tiberian Masorah. Translated and edited by E. J. Revell. Masoretic Studies 5. Missoula, MT: Scholars Press, 1980.
Miqraʾ ʿal pi ha-Masora. Hebrew Wikisource. Especially 1 Chronicles 10:1 and the methodological introduction concerning the Aleppo Codex. (Wikisource)
“Changes Made from UXLC 1.1 to UXLC 1.2.” UXLC / tanach.us. Especially entry 2021.02.22–27, 1 Chronicles 10:1.4. (Tanach)
Deutsche Bibelgesellschaft. “Die Biblia Hebraica von Rudolf Kittel (BHK)” and “Biblia Hebraica Stuttgartensia.” (Die Bibel)
