readings, one book at its own scale — the
counter-text to the last reading, a polemic that takes the efficacy out
of the practitioner and puts it in the act, and the first frame in this
series whose uncheckable part is the practice, and says soTwenty-second of the readings, and the answer to the
twenty-first. Proclus, The
Elements of Theology was the same school’s system as pure deduction:
two hundred and eleven propositions, each proved from the one before it,
and — on its editor’s own account — no reference to theurgy in the whole
of it. That reading ended on an absence. This is the other book of the
same tradition, and it says the opposite thing in the plainest words
available to it: it is not thought that unites, it is the act. Then it
names what the act is made of, and the name it gives is one that no
reader can inspect. The blog has spent this series asking where the seat
is — who or what is doing the work, and where the authority sits. Here
the seat is removed outright, and what replaces it is harder on a
reader, not easier: nothing is left to examine but an act and its
symbols, and the book states, without embarrassment, that the symbols
are understood by the gods alone.
Status. This post reads two editions of one work, and names the edition at every quotation. The primary text is Thomas Taylor’s translation, Iamblichus on the Mysteries of the Egyptians, Chaldeans, and Assyrians — the 1895 reprint of the translation first printed in 1821 — cited by Taylor’s own divisions, Section and Chapter (“Section II, Chap. XI”), and, where a passage comes from his front matter or his own additional notes, marked as his. The second is Emma C. Clarke, John M. Dillon and Jackson P. Hershbell’s Iamblichus: De Mysteriis (Society of Biblical Literature, 2003), Greek and English facing, with an Introduction and notes; it is cited by Book and section (“II.11”) as its running heads print them, and its Introduction and the editors’ notes are cited as the editor’s. Where a quotation is the editor’s note rather than Iamblichus, it is marked as such, and a cross-reference the editor gives is quoted as the editor’s. Both copies are machine readings of printed books and each is damaged in its own places. Line-break hyphenation is undone and spacing collapsed where a quotation crosses a line; typographic quotation marks are rendered here as straight ASCII; curly apostrophes, dashes and ellipses are normalised, and inner quotation marks are rendered as single quotes. Three classes of misreading in the 2003 printing are handled, and are named here: the digit-for-letter slips in which is is set as 15 and the capital I of Iamblichus is set as a lowercase l, both of which are restored; and footnote markers that the machine reading has dropped into the middle of a sentence, which are removed. Where Taylor’s reprint is corrupt — and at a few points it is — the quotation stops, or the damaged part is elided, rather than being mended. Nothing is spliced in from anywhere else, and elisions are marked with a bare ellipsis. The reading of this material through the blog’s model is an inference, offered as one. Nothing here claims the gods are real and nothing here claims they are not; nothing here claims theurgic efficacy is real and nothing here claims it is not; and nothing here validates, disqualifies or restricts anyone’s practice. Iamblichus’ claims are reported as his, Taylor’s as his, the editors’ as theirs. No living or dead person is diagnosed. As everywhere: describe structures, diagnose no one.
The first thing to say about On the Mysteries is that it is not a treatise but a reply, and the reply’s shape is the opponent’s. The Clarke/Dillon/Hersh edition names the genre exactly, and the name is the useful thing:
“it is essentially a series of replies to a set of aporiai proposed by Porphyry, about the nature of the gods, or of the divine realm in general, and the proper mode of our worship of them.”
— the editor, Introduction §5.1 The editor’s name for the genre is “Problems and Solutions” (the editors’ term).1 What set the questions was a letter of Porphyry, and the editor says what that letter was in a sentence that ought to be kept in view for the whole of this reading:
“The De mysteriis, however, is a point-by-point refutation of Porphyry’s Letter to Anebo, an epistle which launched a vicious attack on theurgy, more than likely aimed specifically at Iamblichus and his beliefs.”
— the editor, Introduction §2.2 And in a note on the same page, on why Porphyry changed his position:
“It was also a refutation of Porphyry’s own earlier dabblings in this field, as expressed in the Philosophy from Oracles. Porphyry’s personal association with Plotinus was more than likely the cause of his change of heart.”
— the editor, Introduction §2, note.3 So the book was written against someone, by a man who had been that someone’s pupil, on a subject where the two of them disagreed about the efficacy of ritual in particular. And the book opens by saying that the letter may be taken as addressed to him: “the letter sent to my student Anebo” (the Clarke/Dillon/Hersh text, I.1) — the persona’s student, which the editor reads as a joke at the addressee’s expense.4
The speaker is not Iamblichus but a persona, and the editions name him differently. Taylor’s translation heads the work:
“THE ANSWER OF THE PRECEPTOR ABAMMON TO THE EPISTLE OF PORPHYRY TO ANEBO, AND A SOLUTION OF THE DOUBTS CONTAINED IN IT.”
— Taylor’s translation, the heading of Section I as his volume prints it.5 Taylor prints the name Abammon, and gives the attribution in a note of his own at the foot of the first chapter, resting on Proclus:
“It is requisite to know that the philosopher Proclus, in his Commentary on the Enneads of the great Plotinus, says that it is the divine Iamblichus who answers the prefixed Epistle of Porphyry, and who assumes the person of a certain Egyptian of the name of Abammon, through the affinity and congruity of the hypothesis.”
— Taylor’s note to Section I, Chap. I.6 The 2003 edition prints the same name as Abamon, one m lighter, and its Introduction gives the same attribution in its own voice: Iamblichus, the editors write, is “writing under the assumed guise of the Egyptian prophet” Abamon.1 The one-letter difference is the smallest instance of a difference that runs through this whole reading: the two texts are two readings of one book, and they do not agree in small things.
What the persona is for is stated by the editor as a problem, and answered:
“If Iamblichus was unconcerned with the pressures of the pagan-Christian debate, why did he write in the guise of the purported Egyptian prophet Abamon? The decision was an interesting one, and sparked by Porphyry’s attack to which he is making his reply.”
— the editor, Introduction §3.4 The book’s own first chapter answers it by saying the speaker’s identity is not the point, and both renderings make that move in nearly the same words:
“dismiss from your mind the speaker, whether he be better or worse, and consider what is said, whether it be true or false, rousing up your intellect to the task with a will.”
— the Clarke/Dillon/Hersh text, I.1.7 Taylor’s:
“dismiss the consideration whether the speaker is an inferior or a superior character, but direct your attention to what is said, so as readily to excite your mind to survey whether what is asserted is true or false.”
— Taylor’s translation, Section I, Chap. I.8 So the book’s first instruction is to stop asking who is speaking — the earliest thing in this series that resembles the blog’s own question, and it is answered by a manoeuvre rather than an answer: the authority is placed in the argument’s content, and the speaker is retired.
Then the book states its own subject, and both editions state it for it. Taylor’s Introduction to his volume is explicit about the claim the book is making:
“it is the most copious, the clearest, and the most satisfactory defence extant of genuine ancient theology. This theology, the sacred operations pertaining to which called theurgy are here developed”
— Taylor’s Introduction to his translation.9 The editor of the modern edition says the same thing in more neutral language, and adds what the author’s motive was:
“a masterful attempt to combine the teachings of revelation literature with those of Neoplatonism, and to give theurgic rites a philosophical basis.”
— the editor, Introduction §3.4 And the thesis is stated as the book’s subject, not a topic in it:
“Iamblichus was essentially interested in re-awakening and preserving man’s contact with the ancestral gods, and in arguing that theurgy (or ‘god-work’) rather than theology (or ‘god-talk’) was the only way of achieving this.”
— the editor, Introduction §3.4 Theurgy against theology; the act against the talk. That is the book, and it is the exact inversion of the last reading, where the school’s most complete book was all talk and no rite, and its editor could say so with a straight face.
The blog should also record, at the outset, something the 2003 edition says about the divisions of the book, because every citation below depends on them:
“The present division of the work into books dates only from the Renaissance.”
— the editor, Introduction §5.10 The editor credits the division to Scutellius’ Latin of 1556, after Ficino’s of 1497, and records that “this arrangement was adopted by Thomas Gale in his editio princeps of 1678”,10 and then says what the arrangement costs:
“The division into ten ‘books’ (of very unequal length), while not seriously misleading (as being based largely on natural breaks in Iamblichus’s exposition), does, however, somewhat obscure the original structure of Iamblichus’s work.”
— the editor, Introduction §5.10 The addresses this post cites are a sixteenth-century editor’s, not the author’s. The same shape of problem as the last reading’s — a text established for a reader by someone else — and worse here, because the modern edition classifies the book by its true title (The Reply of the Master Abamon…)11 and then cites it by a later one.
The passage at the centre of this book, and of this reading, is a sentence about what does not unite. The 2003 text renders it:
“it does not follow on this that the offerings made to the gods and divine works are invalid, for it is not pure thought that unites theurgists to the gods. Indeed what, then, would hinder those who are theoretical philosophers from enjoying a theurgic union with the gods? But the situation is not so: it is the accomplishment of acts not to be divulged and beyond all conception, and the power of the unutterable symbols, understood solely by the gods, which establishes theurgic union.”
— the Clarke/Dillon/Hersh text, II.11.12 Taylor’s rendering of the same passage is why this post quotes both editions here — it is the same claim, and it is visibly a different claim:
“For a conception of the mind does not conjoin theurgists with the Gods … For the perfect efficacy of ineffable works, which are divinely performed in a way surpassing all intelligence, and the power of inexplicable symbols, which are known only to the Gods, impart theurgic union.”
— Taylor’s translation, Section II, Chap. XI.13 Read it as a removal. The claim is negative first: thought does not unite. The rhetorical question turns that into an argument — if thought united, the theoretical philosophers would be united, and they are not. Then the positive claim names what does unite, and the name is not a person, not a state of mind, not a discipline and not a virtue. It is “the accomplishment of acts” and “the power of the unutterable symbols” (the Clarke/Dillon/Hersh text, II.11).12 Nobody is standing in this sentence.
Nor can the theurgist be re-inserted at the next step. The same section gives the mechanism and takes the operation from both sides at once — the human and the divine:
“Hence, we do not bring about these things by intellection alone; for thus their efficacy would be intellectual, and dependent upon us. But neither assumption is true. For even when we are not engaged in intellection, the symbols themselves, by themselves, perform their appropriate work, and the ineffable power of the gods, to whom these symbols relate, itself recognises the proper images of itself, not through being aroused by our thought.”
— the Clarke/Dillon/Hersh text, II.11.12 Three clauses, and each takes something from the operator: the efficacy is not intellectual, not dependent upon us, and operates even when we are not engaged in intellection. Taylor’s rendering is harder on the operator still:
“we do not perform these things through intellectual perception … the intellectual energy of them would be imparted by us; neither of which is true.”
— Taylor’s translation, Section II, Chap. XI.13 And on the symbols, Taylor leaves the Greek word standing in his English:
“the ineffable power of the Gods itself knows, by itself, its own images. It does not, however, know them, as if excited by our intelligence; for neither is it natural that things which comprehend should be excited by those that are comprehended, nor perfect by imperfect natures, nor wholes by parts.”
— Taylor’s translation, Section II, Chap. XI.13 The word he leaves untranslated is synthemata, and his note glosses it: “The inexplicable theurgic signs or symbols.”14 He keeps the Greek term so that the reader sees the doing done by the thing and not by the person — and the modern edition does the opposite, rendering synthemata as “the symbols themselves” (the Clarke/Dillon/Hersh text, II.11).12
The gods are not moved by the request. Taylor, on invocation:
“Neither do the invocations which implore the Gods to incline to us, conjoin the priests to them through passion; but procure for them the communion of an indissoluble connexion, through the friendship which binds all things together.”
— Taylor’s translation, Section I, Chap. XII.15 The 2003 text’s rendering of what invocation actually does is the sentence this reading leans on most, because it relocates the operation:
“rather in virtue of the divine love which holds together all things that they provide a union of indissoluble involvement—not, as the name seems immediately to imply, inclining the mind of the gods to humans, but rather, as the truth of things itself desires to teach us, disposing the human mind to participation in the gods, leading it up to the gods and bringing it into accord with them through harmonious persuasion.”
— the Clarke/Dillon/Hersh text, I.12.16 Nothing is inclined downward. The gods are not persuaded or brought; it is the human mind that is disposed, led, and brought into accord. The leading is done by the names and symbols, and Taylor’s rendering of the next sentence carries a technical term into the English and glosses it:
“such names of the Gods as are adapted to sacred concerns, and other divine symbols, are able, as they are of an anagogic or elevating nature, to connect invocations with the Gods themselves.”
— Taylor’s translation, Section I, Chap. XII.15
So far, so Platonic: a hierarchy, an ascent, symbols that carry it. What makes this book the counter-text to the Elements is the next move, about prayer, which denies that prayer is a communication between two parties at all:
“it is neither through faculties nor through organs that the gods receive into themselves our prayers, but rather they embrace within themselves the realisations of the words of good men, and in particular of those which, by virtue of the sacred liturgy, are established within the gods and united to them; for in that case the divine is literally united with itself, and it is not in the way of one person addressing another that it participates in the thought expressed by the prayers.”
— the Clarke/Dillon/Hersh text, I.15.17 Taylor’s rendering ends on the same structural point:
“The Gods, therefore, do not receive prayers in themselves, through any corporeal powers or organs, but rather contain in themselves the energies of pious invocations; and especially of such as, through sacred ceremonies, are established in, and united to, the Gods. For then, in reality, a divine nature is present with itself …”
— Taylor’s translation, Section I, Chap. XV.18 There is no addressee in the ordinary sense. The prayer is not carried to a hearer; the gods “embrace within themselves” what is offered (the Clarke/Dillon/Hersh text, I.15); the divine is “united with itself” (both editions, at I.15 and Section I, Chap. XV).1718 The blog’s marker for a frame has been the seat: an endpoint that is named, and someone or something stationed there whose authority does the work. Here the endpoint is named and the seat holds no occupant the reader could examine — not because the book is modest, but because it has moved the efficacy into the act, and the act’s power is the gods’ own. Taylor says it outright at Section II, Chap. XI: “things pertaining to the Gods, are moved by themselves, and do not receive from any inferior nature a certain principle in themselves of their own proper energy.”13 The answer to where the agency sits is: nowhere a reader can point, and everywhere the gods are — the question has been refused, not answered.
That refusal is repeated where a reader would most expect a seat. Take Porphyry’s sharpest question — how it is that theurgists invoke the gods as superiors and then give them orders as inferiors. The 2003 text splits the act in two:
“the whole of theurgy presents a double aspect. On the one hand, it is performed by men, and as such observes our natural rank in the universe; but on the other, it controls divine symbols, and in virtue of them is raised up to union with the higher powers, and directs itself harmoniously in accordance with their dispensation, which enables it quite properly to assume the mantle of the gods.”
— the Clarke/Dillon/Hersh text, IV.2.19 Taylor’s rendering of the same split is the blog’s question in miniature:
“in all theurgical operations the priest sustains a twofold character; one, indeed, as man, and which preserves the order possessed by our nature in the universe; but the other, which is corroborated by divine signs, and through these is conjoined to more excellent natures, and is elevated to their order by an elegant circumduction, this is deservedly capable of being surrounded with the external form of the Gods.”
— Taylor’s translation, Section IV, Chap. II.20 The modern text says the theurgy has a double aspect; Taylor says the priest has a twofold character — one puts the doubleness in the rite, one in the man, and the difference decides where a reader looks for a seat. The divergence continues in the next clause, where the authority is worn rather than owned:
“and again, he commands these powers, because through arcane symbols, he, in a certain respect, is invested with the sacred form of the Gods.”
— Taylor’s translation, Section IV, Chap. II.20 Taylor also appends a note making the hierarchy explicit in his own voice: “And these irrational spirits, so far as they contribute to wholes, are more excellent than we are, though through being irrational they are inferior to us.”21 The seat, where it exists at all here, is a borrowed costume — the hieratic role of the gods, put on by the rite, worn by whoever is standing there.
The book’s account of the ascent has the same shape. In Book X, on the one road to happiness:
“what other reasonable mode of ascent to it can there be apart from the gods? For if the essence and accomplishment of all good is encompassed by the gods and their primal power and authority, it is only with us and those who are similarly possessed by the greatest kinds and have genuinely gained union with them that the beginning and the end of all good is seriously practised.”
— the Clarke/Dillon/Hersh text, X.1.22 The editor glosses the first clause with a bare note: “That is, other than theurgy.”23 So the only ascent is theurgy, and the theurgy is the gods’ own power and authority, which those who attain it are “possessed by”.22 And the book’s last chapter is not a technique but a prayer:
“It remains, therefore, at the end of this discourse, for me to pray to the gods to grant both to me and to you the unalterable preservation of true thoughts, to implant in us the truth of eternal things forever, and to grant to us a participation in the more perfect conceptions of the gods in which the most blessed end of good things is placed before us, along with the sanction of the harmonious friendship between us.”
— the Clarke/Dillon/Hersh text, X.8.24 Taylor’s closes the same way, shortening the address:
“It remains, therefore, at the end of this discussion, that I should beseech the Gods to afford me an immutable guard of true conceptions, to insert in me truth eternally, and to supply me abundantly with the participation of more perfect conceptions of the Gods, in which the most blessed end of our good is posited, and the confirmation of our concordant friendship with each other.”
— Taylor’s translation, Section X, Chap. VIII.25 The modern text prays “both to me and to you”, and the editor reads the closing line as aimed: “A final put-down of Porphyry—combined, perhaps, with something of an olive branch?”26 The book does not end by handing the reader a method. It ends by asking the gods for something, and the request is also the last move in an argument with the man it was written against.
One more place a reader would look for a seat is the mediator — a person standing between the practitioner and the god. There is no such person. The book has priests, a caste, and an address to an Egyptian prophet, and every one of those is a persona, a rank, or an addressee. What stands between practitioner and god is a symbol — not a person, not an office, and not learnable the way a technique is learnable.
The blog’s apparatus has been asking where the seat is, and the readings have given it several answers. The owned cosmos named the endpoint and stationed someone at it. The cosmos that refuses kept the correspondence and dropped the endpoint. The Elements proved a whole cosmos and stationed nobody, and the blog corrected itself on that one: the operator was absent from the text, not from the system, and the lesson taken was that the marker of a frame is the claim, and that a frame in proof-form relocates its uncheckable part to the axioms without labelling them as axioms.
This reading supplies a third arrangement, and it is the one the instrument cannot currently see. Here the seat is not hidden, not relocated, and not occupied by a substitute. It is removed, and what replaces it is a practice whose operative part the book declares unavailable to anyone but the gods. The declaration is not tucked away; it is inside the sentence that states the doctrine:
“the power of the unutterable symbols, understood solely by the gods, which establishes theurgic union.”
— the Clarke/Dillon/Hersh text, II.11.12 Taylor’s rendering is the same claim in other vocabulary: “the power of inexplicable symbols, which are known only to the Gods, impart theurgic union.”13 Unutterable and inexplicable; understood solely by the gods, known only to the gods. The symbols are the efficacy, and the symbols are by construction not available for inspection. The book does not apologise and does not explain it away. It states it as the mechanism.
Nor is that a single sentence’s slip. The book says the same about the ground of the whole system at I.21, in a passage whose two renderings differ just enough to set side by side:
“unutterable truths are expressed through secret symbols, beings beyond form brought under the control of form, things superior to all image reproduced through images, and all things brought to completion through one single divine cause, which itself so far transcends passions that reason is not even capable of grasping it.”
— the Clarke/Dillon/Hersh text, I.21.27 Taylor’s:
“which is in so great a degree separated from passions, that reason is not able to come into contact with it.”
— Taylor’s translation, Section I, Chap. XXI.28 One says reason cannot grasp it, the other that reason cannot come into contact with it. Both say the same thing about the reader. And the passage goes on to explain why people get the frame wrong, in a phrase that names exactly the error the blog’s instrument makes — “This, therefore, is nearly the cause of our aberration”28 — where the aberration, on both renderings, is judging divine things by human ones. The book’s position is that the standard is not available: not that the argument is hard, but that its terms are the gods’ own, and reason is named as unable to reach them.
The same structure shows in the treatment of the sacred names, where a reader can measure it because the text is talking about words:
“which may indeed be called ‘inexplicable’ and ‘barbarous,’ but which are in fact wholly suitable for sacred rituals.”
— the Clarke/Dillon/Hersh text, VII.5.29 The claim is not that the names are difficult but that they must not be altered, and the editor’s note attaches the injunction to the Chaldaean Oracles and to a line of ritual practice running back through the magical papyri.30 The book that argues the act is the efficacy also holds that the act’s words are untranslatable by design — a pointed finding to state against the blog’s own method, which is reading the argument in English.
The modern edition’s note at I.15, cross-referring to the central passage, makes the same point from the other side, and it is the editor’s sentence rather than Iamblichus’, which is why it is marked as the editor’s:
“and it is a point often reiterated by ‘Abamon’ (see e.g. II.11.96-97)—not even the most accomplished sage, so long as he maintains a purely intellectual approach (as does Porphyry, and as did Plotinus), can attain to the highest levels of theurgic union.”
— the editor’s note to I.15.31 The editor names Porphyry and Plotinus as the two men whose approach cannot reach the top of the practice, and prints that as a restatement of the author’s position rather than a criticism. The editor records elsewhere that the author asserts “all wonders or demonstrations of power are their work alone” — the gods’ — and that by the editors’ summary “wonder-working by man was at best impious, at worst an example of meaningless sorcery.”2
Put those together and the finding is plain. The seat is removed, the efficacy sits in acts and symbols, and the operative content of the acts and symbols is declared unknowable to anyone but the gods. The blog’s instrument looks for where the authority sits; the answer here is in the rite, and in the words of the rite, which cannot be checked by the people performing them, by the philosophers reasoning about them, or by the reader reading them. The uncheckable part of this frame is not a person and not a claim. It is the practice — and the text says so on the page, in the same sentence as the doctrine, without hedging. That is the third case, and a more consequential one than the first two, because a person can be watched and a claim can be tested, and a rite whose meaning belongs to the gods has been placed beyond both by construction.
Where the last reading’s correction was that a frame in proof-form relocates its uncheckable part to the front and mislabels it, this reading’s is stronger and less comfortable: a frame can be entirely candid about its uncheckable part and remain uncheckable. Candour is not a safeguard. It is at most a disclosure — and this book discloses more than the Elements did.
The blog has been running one instrument through this series since the owned cosmos: find the endpoint, find the route, and ask who is stationed at the end doing the work. The reading of Proclus’ Elements sharpened it into a second question — does the frame state its premises or hide them? — and the blog’s conclusion there was that the axiomatic form relocates the uncheckable part rather than removing it.
The correction this reading owes is that both of those questions assume the uncheckable part is something a reader could, in principle, put a finger on. A named seat, a stated axiom, a revealed text, a proof: each offers the reader a place to look, even when what is found there turns out to be an assumption. That is the ground of the last reading’s correction — the axiomatic form relocates the uncheckable part to the front. On the Mysteries relocates nothing. So:
The instrument’s questions all locate the uncheckable part somewhere a reader can reach: a person whose authority is borrowed, a premise that is not demonstrated. Iamblichus supplies the arrangement the instrument has no question for. He removes the seat, puts the efficacy into an act, and then declares the operative content of the act to be the gods’ knowledge — so that the frame’s uncheckable part is no longer a person and no longer a claim. It is a practice, and there is nothing in it for a reader to inspect. The absence of a seat is not a safeguard and not an improvement. It is a third location for the same problem, and in this book the location that is hardest to reach.
The consequence for the blog’s test is a third question, and it is the one this reading earns. The series has asked: is the endpoint named, and is anyone stationed at it? Then: is the frame’s first principle offered as a postulate or as a fact? The question this book adds is not about either end of the frame but about its middle: what is this frame’s uncheckable part made of — a person, a claim, or a practice — and does the frame say which? The Elements answered with a claim and did not label it. On the Mysteries answers with a practice and labels it twice in the sentence that states the doctrine. Both are uncheckable; only one of them is honest about it; and honesty changes nothing about the reader’s position. That is a harder finding than the last one, and the blog should hold it without softening it: a frame can tell you exactly where it is uncheckable, precisely, in plain language, and be uncheckable anyway.
Three, and all three belong in the body, because each one cuts against the finding above rather than decorating it.
First: this is a polemic, and the shape of its answers is not its own. The book is a reply, and the reply is organised by the opponent’s letter. The editor names it a “point-by-point refutation of Porphyry’s Letter to Anebo”.2 That has two consequences a reader should carry. One is that nothing in the book is a neutral description of practice: every position is stated where Porphyry’s question forces it, so the book’s proportions are the questioner’s. The book says almost nothing, for example, about how a rite is to be performed, and a great deal about whether it can work — because Porphyry asked the second and not the first. The other consequence is the one this post has already recorded as a fact about its own citations: the book’s division into ten books is a Renaissance arrangement, adopted into the editio princeps of 1678.10 The evidence for the finding in §3 is therefore itself editorially assembled. That does not weaken the finding — the sentences are Iamblichus’ — but it does mean that the blog’s “Book.Section” citations are citations to a shape the sixteenth century imposed, and a reader who wants to know where a passage sat in the author’s own arrangement cannot be told.
Second: the uncheckable ground is the argument, and the cost should be stated plainly. The cost of the finding is that there is no version of this book from the outside. A reader cannot adopt the frame and check it, because the terms in which it would be checked — the symbols, the names, the efficacy of the act — are held by the book to be the gods’ knowledge and not the reader’s. The book states this as its strength; from where the blog sits it is a limit that cannot be argued away. The blog’s job is to describe it, not adjudicate it: the practice is uncheckable by construction, the construction is the claim, and the claim is not the sort of thing a non-participant can test. The one thing the blog must not do is convert “uncheckable” into “false”. Nothing here establishes that theurgic efficacy does not work; what is established is that this book’s account of it cannot be verified from outside the practice, and that the book says as much. That is a statement about the frame’s structure, and about the reader’s access to it, and about nothing else.
Third: the two renderings differ measurably, and an apparatus adds what a translation cannot. That is the subject of §6. The short version: Taylor’s English and the modern text disagree where the doctrine is densest; the modern edition supplies book-and-section addresses, the editor’s cross-references and the editor’s own notes; Taylor’s volume has an Introduction and Additional Notes keyed to his printed pages. Neither volume is the book. Both are readings, and this post has quoted both where they diverge.
This section is a comparison at named loci and makes no claim beyond them.
The central passage is the cleanest case, because both editions render the same Greek and the difference is doctrinal. On the first clause, the modern text has “it is not pure thought that unites theurgists to the gods”;12 Taylor has “a conception of the mind does not conjoin theurgists with the Gods”.13 Four substitutions in one clause. Pure thought against a conception of the mind: the modern text names a faculty in its purity, Taylor a mental event. Unites against conjoin: the modern verb is the technical one, Taylor’s plainer and slightly more physical. To the gods against with the Gods moves the relation from an approach to an association. And the mood changes with the construction: the modern text is a flat negative assertion, Taylor’s the same assertion carried by a noun subject doing something. The two sentences make the same point and would be used to make different arguments — one by a philosopher distinguishing faculties, the other by a Platonist describing what attaches.
The second clause is a sharper difference, because the modern text makes explicit what Taylor leaves as a technical term. The 2003 edition has “the accomplishment of acts not to be divulged and beyond all conception, and the power of the unutterable symbols”;12 Taylor has “the perfect efficacy of ineffable works, which are divinely performed in a way surpassing all intelligence, and the power of inexplicable symbols”.13 Accomplishment of acts against perfect efficacy of works; unutterable against inexplicable; and, in the same Greek clause, not to be divulged against divinely performed — two different words, the one the modern text renders and the one Taylor does. Taylor’s “divinely performed” answers a word his Greek carries, which the modern text prints in the facing Greek but not in its English at that clause, so the divine performer is not an agent Taylor added. The modern text leaves the seat emptier than Taylor’s does, at the exact point where the passage is saying the seat is empty. That is a translation difference that bears on the finding of §3, and it is why this post has quoted both renderings rather than choosing the cleaner one.
The third locus puts a person where the modern text puts a process. On the division of the theurgic act: the modern text reads “the whole of theurgy presents a double aspect”,19 Taylor “in all theurgical operations the priest sustains a twofold character”.20 Taylor’s subject is a man and the modern text’s subject is the rite, and they assign the doubleness to different objects. The same divergence appears in the vocabulary of authority on the next clause: the modern text has the art “both naturally invokes the powers from the universe as superiors, inasmuch as the invoker is a man, and yet on the other hand gives them orders, since it invests itself, by virtue of the ineffable symbols, with the hieratic role of the gods”,19 where Taylor has a he who “is invested with the sacred form of the Gods”.20 One is a possession and one is a costume; one has a wearer and one does not.
There is a documented reason for the divergence on Taylor’s side. His own Introduction states his view of the work’s condition and of his chief source:
“The difficulties attending the translation of this work into English are necessarily great”
— Taylor’s Introduction to his translation.9 They arise, he says, not only from the work’s “sublimity and novelty” but also “from the defects of the original”.9 And on the annotations he used:
“he shows himself to be an inaccurate, impertinent, and garrulous smatterer.”
— Taylor’s Introduction, of Gale.9 Taylor is not describing a text he trusts. His notes record where his text was short: of one detail in the discussion of prayer, “This particular respecting the apples of gold is added from the version of Scutellius, who appears to have translated this work from a more perfect manuscript than that which was used by Gale.”32 So Taylor’s English is partly a translation of a Latin translation of a source that, on his own account, was not the fullest available — the same class of finding the last reading made about his Proclus, now in a second volume by the same translator. The renderings differ measurably because they are made from different texts by a man who said so.
What an apparatus adds that a translation cannot is easiest to state by listing what the modern edition supplies at these loci. It gives every passage a book-and-section address, so a reader can find the same sentence in another edition — which is how this post caught the address problem: Dodds cites this passage as De mysteriis I. 11, and the modern edition prints it at II.11. It gives the editor’s cross-references: the note at I.15 pointing back to II.11.96-97,31 the list of the author’s assertions in the Introduction, which repeats the address among several other book references,2 and the gloss reading one clause of X.1 as “other than theurgy”.23 It gives the provenance of the book’s divisions, and says plainly that they are a Renaissance editor’s.10 And it gives the editor’s judgement of the closing line.26 Taylor’s volume gives none of that: a translation, an Introduction, and Additional Notes keyed to page numbers — and where it does help, the help is Taylor’s opinion of his source, which is not an apparatus. A reader with only the 1895 volume has the sentences and no way to place them; a reader with only the modern one has the whole apparatus and a rendering so fluent that the divergences from Taylor’s disappear, including the one at the central passage. This post needed both, and the reason is not completeness. It is that the two editions disagree where the book is arguing.
One last difference belongs here because it is not a rendering at all. Taylor’s volume carries the publisher’s note explaining that the 1895 reprint is “an almost exact facsimile of the original edition, which was first printed in 1821”33 — so the copy this post read descends photographically from Taylor’s own book. The modern edition carries the history of the readings that followed, and the range is wide: the editor records Nilsson calling it a “basic book for religion in late antiquity”, “E. R. Dodds considered it ‘a manifesto of irrationalism’”, and Des Places “a breviary of paganism in decline”.4 The same book has been read as the basic document of a religion and as a manifesto of irrationality, by competent scholars in the last hundred years, and the modern editor records both before turning to the more recent scholarship that reads the book otherwise.
One. On the Mysteries is the counter-text to the Elements of Theology: the same tradition’s answer from the other side. Where the Elements derives a whole cosmos and performs nothing, this book performs a rite whose efficacy it denies to thought — “it is not pure thought that unites theurgists to the gods”.12 The joint between the two books is the joint between a proof and a practice.
Two. The seat is removed, not relocated. Three moves take efficacy from the practitioner — the rite’s power is not intellectual, is “not … dependent upon us”, and operates “even when we are not engaged in intellection” — and give it to acts and symbols that “perform their appropriate work” by themselves.12 The gods are not inclined toward the worshipper; the worshipper is disposed toward the gods.16 The divine “is literally united with itself”, and prayer is not one person addressing another.17 Where the Elements had an absent operator standing beside the text, this book has an operator told, in the text, that the work is not his.
Three. The uncheckable part of the frame is the practice, and the book says so. The symbols that carry the efficacy are “unutterable” and “understood solely by the gods”;12 the ground of the system is that “reason is not even capable of grasping it”;27 the sacred names must not be altered.29 This is the third case the instrument was missing: not a seat, not a claim, but an act whose operative content is declared to be the gods’ knowledge — uncheckable by construction, with the construction in the doctrine’s own sentence.
Four. The correction: the blog’s questions all locate the uncheckable part somewhere a reader can reach — a person whose authority is borrowed, a premise that is not demonstrated — and this frame has neither. Iamblichus removes the seat, puts the efficacy into an act, and declares the act’s operative content to be the gods’ knowledge. The test gains a third question: is the frame’s uncheckable part a person, a claim, or a practice, and does the frame say which? This book answers “a practice”, and says so, and is uncheckable anyway — which is the harder half: candour is not a safeguard, and it does not restore anything to the reader.
Five. The counterweights, in the body, are three. The book is a point-by-point refutation whose proportions are its opponent’s, and its division into books is a Renaissance arrangement, so the addresses cited here are not the author’s. The ground is uncheckable as the argument requires — a structural cost to describe, not adjudicate: the book’s account cannot be verified from outside the practice, and nothing here claims it is false. And the renderings disagree measurably where the book argues: on what does the uniting (pure thought against a conception of the mind), on whether the acts have a performer (divinely performed against not to be divulged), and on whether the doubleness belongs to the man or the rite. Neither edition is the book.
What the blog takes from the twenty-second reading is the completion of a run of four. The Orphic Hymns perform a cosmology and assert nothing. The Chaldean Oracles assert one and station no one. Proclus’ Theology of Plato argues one and gives every being its place. The Elements proves one and performs nothing. And On the Mysteries performs one — and tells the reader, in the same breath as the doctrine, that the part of the performance which does the work is the part the reader cannot have. The frame that removes the seat is not the frame that leaves something to check. It is the frame that puts the whole of its efficacy in an act, and then says, in the sentence that states it, that the act is understood by the gods alone. The blog has spent this series asking where the authority sits. This book answers that nobody’s authority is doing the work — and that this is not the safer arrangement, because a rite nobody owns is a rite nobody can examine either.
Marking: a reading, and a measured comparison. Documented — every passage in quotation marks, taken from one of two named editions of On the Mysteries: Thomas Taylor’s translation, the 1895 reprint of the version first printed in 1821, cited by his own Section and Chapter and, where the words are his, marked as his Introduction or his notes, checked character for character against the copy read; and Clarke, Dillon and Hershbell’s Iamblichus: De Mysteriis (2003), cited by Book and section as its running heads print them and, where the words are the editors’, marked as the editor’s, checked the same way. Attributed to the editor, not to Iamblichus — the characterisation of the work as a point-by-point refutation of Porphyry’s Letter to Anebo and the paragraph on Porphyry’s association with Plotinus; the account of the pseudonym, the work’s genre and its true title; the sentence placing theurgy against theology and the sentence on wonder-working by man; the list of the author’s assertions that all wonders are the gods’ work alone; the account of the Renaissance division of the work into books; the note naming Porphyry and Plotinus as unable to attain the highest levels of theurgic union; the gloss reading the opening of Book X as “other than theurgy”; the note on the closing line; the note on the barbarian names and its cross-references; and the cross-reference to II.11 in the note to I.15. Attributed to Taylor, not to Iamblichus — the translation wherever it is quoted, the headings of his Sections and Chapters, and the Introduction and footnotes attributed to him above. Attributed to this post — the reading of II.11 as the removal of a seat rather than its relocation; the finding that the frame’s uncheckable part is the practice rather than a person or a claim, and that the book discloses it; the correction stated in §4 and the third question it adds to the blog’s test; the three counterweights in §5; and the measured comparisons in §6, including the reading of the two renderings of II.11 as leaving the seat emptier in the modern text and of the two renderings of IV.2 as putting the doubleness in the rite or in the priest. The passages are addressed as they are in the editions named: Taylor’s volume carries its own Section and Chapter divisions, and the modern edition’s Book-and-section addresses are, on the editors’ own account, a Renaissance arrangement rather than the author’s, which is why every citation above names its edition as well as its address. The editions are damaged and the damage is declared. Line-break hyphenation is undone and spacing collapsed; typographic quotation marks are rendered as straight ASCII; curly apostrophes, dashes and ellipses are normalised, and inner quotation marks are rendered as single quotes. In the 2003 copy the digit-for-letter misreading of is as 15 is restored, as is the capital I of Iamblichus where it is set as a lowercase l, and footnote markers dropped into the middle of a quoted sentence are removed; two of Taylor’s sentences are damaged in the reprint, and at those points the damaged part is elided rather than being mended. Elisions are marked with a bare ellipsis. The cosmology is not endorsed — nothing here claims the gods, the daimones, the symbols, the theurgic union or the efficacy of any rite are real, and nothing here claims they are not; no one is diagnosed; and nothing here validates, disqualifies or restricts anyone’s practice.
Clarke, Dillon and Hershbell, Iamblichus: De Mysteriis — the editors’ Introduction. The genre sentence and the characterisation of the work’s dating and authorship are from §3 and §5 of that Introduction; the phrase “writing under the assumed guise of the Egyptian prophet” is the editors’ own, in §3 of the same Introduction.↩︎
The same editors’ Introduction, §2 (the chapter on Iamblichus himself). The sentence about “a point-by-point refutation”, the paragraph on Porphyry’s change of heart, the list of the author’s assertions that all wonders are the gods’ work alone, and the sentence on wonder-working by man are all from that section and its notes.↩︎
The editors’ note to the Introduction §2 paragraph on Porphyry’s change of position.↩︎
The same editors’ Introduction, §3 (the chapter on the work as a defence of theurgy and an answer to Porphyry’s letter). The paragraph on the pseudonym and the pressures of the pagan-Christian debate, the sentences on theurgy against theology, the characterisation of the work as a combined defence of revelation literature and Neoplatonism, and the range of modern assessments (Nilsson, Dodds, Des Places) are all from that section and its notes.↩︎
Taylor’s translation, the heading of Section I as his volume prints it. Taylor’s division of the work into Sections and Chapters is his own volume’s, and each of his Sections corresponds to one of the modern edition’s Books.↩︎
Taylor’s own note at the foot of the first chapter of Section I, quoting Proclus’ testimony to the work’s authorship and to the persona of the speaker.↩︎
The Clarke/Dillon/Hersh text, Book I, section 1.↩︎
Taylor’s translation, Section I, Chap. I. The rendering of the passage on dismissing the speaker is quoted from the same chapter, with the clause that follows it omitted.↩︎
Taylor’s Introduction to his translation: on what the work is and what it defends, on the difficulties attending the translation and the state of the text he worked from, and on the annotations he used. The judgement of Gale is quoted from the same paragraph.↩︎
The same editors’ Introduction, §5 (on the nature and contents of the work). The sentence on the Renaissance division, the attribution to Scutellius and Ficino, the record of Gale’s adoption of the arrangement, and the sentence on what the division obscures are all from that section.↩︎
The editors’ note giving the work’s true title.↩︎
The Clarke/Dillon/Hersh text, Book II, section 11. Every quotation given from this section — on the offerings and divine works, on the accomplishment of acts, on the unutterable symbols, and on the efficacy being intellectual and dependent upon us — is from the same section, and the renderings given as Taylor’s are his rendering of the same Greek.↩︎
Taylor’s translation, Section II, Chap. XI. This is Taylor’s rendering of the Greek rendered in the modern edition at II.11, and is quoted at several points in this post. The elision in the first quotation given from it marks the point at which the copy is damaged, and nothing is supplied to bridge it.↩︎
Taylor’s own footnote in Section II, Chap. XI, glossing the Greek word he leaves untranslated in his text.↩︎
Taylor’s translation, Section I, Chap. XII.↩︎
The Clarke/Dillon/Hersh text, Book I, section 12. The quotation given from it is the passage on what the invocation does, and the sentence that follows it in the same section is paraphrased above rather than quoted.↩︎
The Clarke/Dillon/Hersh text, Book I, section 15.↩︎
Taylor’s translation, Section I, Chap. XV. The quotation breaks off before Taylor’s sentence closes, and the ellipsis marks the cut; the break is this post’s, not a damage in the copy.↩︎
The Clarke/Dillon/Hersh text, Book IV, section 2.↩︎
Taylor’s translation, Section IV, Chap. II.↩︎
Taylor’s own footnote to Section IV, Chap. II.↩︎
The Clarke/Dillon/Hersh text, Book X, section 1.↩︎
The editors’ gloss on the opening clause of X.1.↩︎
The Clarke/Dillon/Hersh text, Book X, section 8.↩︎
Taylor’s translation, Section X, Chap. VIII.↩︎
The editors’ note on the closing sentence of X.8.↩︎
The Clarke/Dillon/Hersh text, Book I, section 21.↩︎
Taylor’s translation, Section I, Chap. XXI. Two fragments are quoted from it: the clause on what reason cannot do, and the phrase naming “our aberration”, whose sequel is paraphrased above rather than quoted.↩︎
The Clarke/Dillon/Hersh text, Book VII, section 5.↩︎
The editors’ note on the barbarian names at VII.5, which cites the Chaldaean Oracles and the Hermetic literature, among others.↩︎
The editors’ note to I.15. The note’s own cross-reference to II.11 is quoted with it, and is the editors’ address, not this post’s.↩︎
Taylor’s own footnote in Section V, Chap. XXVI, on the detail he supplies from Scutellius’ Latin version.↩︎
The publisher’s note prefixed to the 1895 reprint of Taylor’s translation, on the relation of that reprint to the edition of 1821.↩︎