before the sefirot
Merkavah and the Hekhalot literature
Six or seven hundred years before anyone drew a Tree, Jewish mystical writing was about a chariot. It grew out of the exegesis of one chapter of Ezekiel, it describes an ascent through seven palaces — which its own sources call a descent — and it has almost nothing in it that looks like the ten sefirot.
The disputed road to Kabbalah
A familiar history runs from Merkavah vision through the Hekhalot texts to the ten sefiroth of the Bahir and finally the Tree. The relationship is real enough to study, but not settled enough to draw as one continuous lineage.
What has happened over the last thirty years is that the field moved the whole corpus later. The editors of the standard recent collection describe a “revisionist” position that locates the formative stages of this literature in Byzantine Palestine and Sasanian or early Islamic Iraq between the fifth and ninth centuries — and that has, in their words, “largely displaced the contention, associated with the pioneering work of Gershom Scholem, that the Hekhalot texts record the esoteric doctrines, ritual practices, and mystical experiences of the earliest generations of rabbis.” If that is right, this literature is not the ancestor of the rabbis' mysticism; it is a later cousin.
That is not a settled question with a dead losing side. Scholem's position is still actively defended, and the re-assessment was launched, with some irony, in Scholem's own Festschrift. What can be said without taking a side is that the arrows in the popular chain are of different kinds. The first is exegesis. The middle one is disputed. And the last — the move from a decad of divine potencies appearing unsystematically in a late-twelfth-century Provençal anthology to the sefirot as the thirteenth century systematised them — is a change of kind rather than a transmission of the same doctrine. The editor of the critical edition of the Bahir says of it that it “should not be seen as a primer” of sefirotic symbolism, and that the word sefirah rarely appears in it at all.
The Jewish Encyclopedia reached much the same caution in 1904, which is worth quoting because it is old enough to be freely reproduced: the Talmudic doctrine “may well be connected with the old Jewish esoteric teachings of the time of the Second Temple… but the theosophic and cosmogonic portions of this literature can not with certainty be regarded as the source of the Talmudic doctrine, nor can the literature of the so-called geonic mysticism… be regarded as the immediate continuation of Talmudic mysticism.”
The chapter it all comes out of
Ezekiel 1 — the vision by the river Chebar — with Ezekiel 10 and Isaiah 6 behind it. The Mishnah restricts the subject: the chariot is not expounded even to one, unless he is wise and understands of his own knowledge. The Talmud's story of the four who entered the orchard is what the restriction is usually read through, and only one of the four came out in peace.
One correction worth making, because it is repeated everywhere: the noun מרכבה does not appear in Ezekiel 1 at all. Ezekiel has the living creatures and the wheels; the word merkavah in connection with the cherubim is in 1 Chronicles 28:18. The name of the subject and the vision it names come from two different books.
The texts
These works carry layers rather than single dates, and none has a named author. The corpus has no one critical text, and even its titles are late conventions. Each entry keeps the questions surrounding its date, shape and authorship.
Re'uyot Yehezkel
A mystical midrash on the single verse Ezekiel 1:1 that enumerates what Ezekiel saw, and gives an unusually detailed account of seven heavens with their names, dimensions and functions.
The work shows Hekhalot mysticism growing directly from an intense reading of Ezekiel's first chapter.
What is contested about it
The corpus has no critical text and arguably no originals. Schäfer's model denies that these macroforms ever had a single authoritative version, and the titles themselves are late conventions rather than ancient names. Dates below are ranges with the disagreement named, never a single year. Two Hebrew spellings of the title circulate, and no critical edition was located to decide between them. The established transliteration therefore stands alone.
Who wrote it, and who did not
A midrash rather than an ascent narrative, and the only text here that is not in Schäfer's Synopse.
1 edition and where to read them
- Jewish Encyclopedia articles on the Merkabah (1904)
- Public domain, and best read as historiography rather than a current account. Kohler identifies Merkavah mysticism with Mithraism, casts Metatron as Mithra the divine charioteer and speaks of 'the same hallucination at work'. That period thesis is no longer accepted; the companion article remains the more useful of the two. jewishencyclopedia.com ↗ jewishencyclopedia.com ↗
Hekhalot Rabbati היכלות רבתי
The longest and best known Hekhalot text follows R. Ishmael and R. Akiva through seven guarded palaces, where seals and divine names open the way. The journey is called a descent to the chariot, yeridah la-merkavah. Throne hymns, a martyrology and a rite for obtaining perfect memory of Torah interrupt and deepen the passage.
Much of the later image of the chariot mystic grows from this work, including the story of R. Nehunyah being drawn back from trance when his disciples render him ritually impure.
What is contested about it
The corpus has no critical text and arguably no originals. Schäfer's model denies that these macroforms ever had a single authoritative version, and the titles themselves are late conventions rather than ancient names. Dates below are ranges with the disagreement named, never a single year. The martyrological section is now argued to be a later insertion reflecting Byzantine imperial ceremonial rather than an early stratum.
Who wrote it, and who did not
The text speaks in the names of the tannaim, though scholars do not accept them as its authors. In Schäfer's redactional model, no single original text ever stood behind the surviving versions.
2 editions and where to read them
- Morton Smith translation, corrected by Scholem (2009) locator only
- Copyright restricted; cited by locator and not quoted. This complete English of chapters I–XXX is keyed to Schäfer's section numbers and posted online by permission of Smith's literary executor under 'all rights reserved'. digital-brilliance.com ↗
- Synopse zur Hekhalot-Literatur (1981) locator only
- Copyright restricted: cite by locator and do not quote. The section numbers commonly used for this corpus (§§81–306 for the Greater Palaces, §§1–80 for 3 Enoch) come from this edition, which is why it is listed against a text it does not translate.
Hekhalot Zutarti היכלות זוטרתי
Shorter, denser and more magical than the Greater Palaces, turning on the disclosure of divine names. It carries the water vision: the warning that when the one who descends reaches the stones of pure marble he must not cry 'water, water'.
The same warning appears with R. Akiva in Hagigah 14b, one of the few visible points of contact between a Hekhalot text and a Talmudic passage.
What is contested about it
The corpus has no critical text and arguably no originals. Schäfer's model denies that these macroforms ever had a single authoritative version, and the titles themselves are late conventions rather than ancient names. Dates below are ranges with the disagreement named, never a single year. Morray-Jones, who defends the continuity case, nonetheless concedes of his own edition that 'it cannot be claimed that this version represents an original or authoritative text' and that 'Schäfer's contention that no such text ever existed is basically correct'.
Who wrote it, and who did not
Centred on R. Akiva, the one of the four who entered the orchard and returned in peace.
Ma'aseh Merkavah מעשה מרכבה
The most liturgical of the Hekhalot texts: a collection of prayers and hymns inside a dialogue between R. Ishmael and R. Akiva, which embeds the ascent instructions in recited prayer and treats the recitation itself as the technique.
Its many spoken prayers make the practical side of Hekhalot literature unusually clear: this is a text meant to be recited as well as read.
What is contested about it
The corpus has no critical text and arguably no originals. Schäfer's model denies that these macroforms ever had a single authoritative version, and the titles themselves are late conventions rather than ancient names. Dates below are ranges with the disagreement named, never a single year. In the Mishnah, 'ma'aseh merkavah' names a subject of study restricted from public exposition. This late-antique book borrowed the phrase centuries later, so citations must distinguish the subject from the text.
Who wrote it, and who did not
A dialogue in form; an anthology of prayers in substance.
Shi'ur Qomah שיעור קומה
An enumeration of the measurements of the divine body, given in figures so vast they defeat comprehension, which is generally taken to be the point of them.
It is the most theologically scandalous material in the corpus and was attacked as such from the Middle Ages onward. Its existence is a large part of why later Jewish thought treated this literature at arm's length.
What is contested about it
The corpus has no critical text and arguably no originals. Schäfer's model denies that these macroforms ever had a single authoritative version, and the titles themselves are late conventions rather than ancient names. Dates below are ranges with the disagreement named, never a single year. Origen attests the tradition in the second century, but the surviving texts are much later. The age of the tradition and the date of these versions are separate questions.
Who wrote it, and who did not
Not a freestanding book so much as a body of material found inside the Greater and Lesser Palaces, 3 Enoch and Merkavah Rabbah.
3 Enoch ספר היכלות
Third Enoch is a central source for the tradition in which Enoch is elevated and transformed into Metatron. It develops Metatron's heavenly offices and titles, including Prince of the Divine Presence, within the world of Hekhalot mysticism.
Primary mystical evidence for Metatron's identity and offices, independent of the much later Kether correspondence.
What is contested about it
Its layers are difficult to date, and the text combines more than one strand of Metatron tradition.
Who wrote it, and who did not
The title '3 Enoch' is modern. Hugo Odeberg coined it for his 1928 edition and acknowledged the choice in his preface: 'The Venerable the Archdeacon of Westminster, Dr R. H. Charles, has authorized the writer to give the present book the title "3 Enoch".' The work calls itself Sefer Hekhalot, the Book of Palaces, and occupies §§1–80 in Schäfer's numbering.
1 edition and where to read them
- 3 Enoch, or The Hebrew Book of Enoch (1928)
- Public domain in the United States since 2024, but still protected in the EU and Sweden until 2044 because Odeberg died in 1973. The more restrictive status governs quotations here. archive.org ↗ archive.org ↗
Two names that mislead
Ma'aseh Merkavah is both a subject and a book, and they are eight centuries apart. The Mishnah restricts a subject of study. The Hekhalot text that carries the same name borrowed the phrase much later. Attaching the restriction to the book would make a second-century rule apply to something that did not exist.
3 Enoch is not what the book calls itself. Hugo Odeberg coined the title for his 1928 edition, and recorded in his preface that he had been authorised to do so. Its own name is Sefer Hekhalot, the Book of Palaces.