Saturday, May 14, 2022

One Secret of the Seals

A scene from Ingmar Bergman’s The Seventh Seal

As so often here in the land of blogs, it may seem like it’s  about who receives the proper credit for getting something started. So before I start to give my answers on the origin and meaning of the word kha-tham that you see just below, I’d like to say something about how I got up to this point. What makes me do it?

A detail (complete page illustrated below)

Based on my past experience, whenever you have a two-syllable word (excluding endings) in Tibetan in which the two syllables don’t successfully (or very successfully) suggest a Tibetan-language etymology, you can feel justified and often vindicated in supposing it to be a foreign borrowing.* But I should add that these borrowings are liable to be subjected to what I call a ‘Tibetanizing’ way of spelling that makes them look more as if they were native Tibetan terms. In some cases you can see this disguising process at work, observing different spellings at different times in the history of the language. I don’t want to go too much into this right now, just to say past experience explains why seeing a word like kha-tham immediately sets my mind to wondering if it isn’t ultimately foreign. If you want more justification for such hunches, have a look at one example in the earlier blog about a Central Asian Turkish word that ended up in Tibetan in quite early times: thu-lum (ཐུ་ལུམ་).

(*I realize I’m neglecting, and unjustifiably overlooking, calques and direct borrowings from Sanskrit, which at times may undergo their own Tibetanizing spelling transformations, whether at the moment of borrowing or further on down the road. If you need a quick example, think of bskal-pa for Sanskrit kalpa, meaning ‘eon.’)

But before giving my positive answer about where kha-tham comes from, I’d like to give the internally Tibetan derivation a good old college try, if nothing else to demonstrate that it doesn’t work out so well. If it doesn’t interest you, you can skip over the blue. 

I mean, the kha could mean mouth or surface or face, or the mouth-like opening of a vessel, the edge of a lake, a shore (and several quite obscure and rare but possible meanings). Kha tends to be used in Tibetan scriptures when translating the Sanskrit word mukha with its similar but not identical range of meanings.

Now the syllable tham[s] causes us more difficulties. The only usage as a one-syllable term I know of is archaic and quite obscure, with a meaning to be agreed upon. Equally old and obscure is the bisyllabic tham[s]-lag, that is glossed by man-ngag, or secret instructions, in Sanskrit upadeśa

Finally, putting the two syllables kha and tham together doesn't easily lead anywhere that makes sense. This kind of failure is my usual trigger for searching abroad, but which abroad? How far abroad do we need to go?

Particularly in Buddhist contexts India is place of first resort when you feel the need to think about borrowings. In our case we at least need to consider this possibility since we cannot go forward in our argument without getting it out of the way. First of all I may have to say that Indian languages don’t confuse consonants with their aspirates. The letter ‘k’ is not ‘kh’, and ‘t’ is not ‘th’, and this sharp distinction holds whether it’s an Indian language that we are talking about, or Tibetan itself. Foreigners might confuse them or regard them as the same, but that’s entirely their problem. That’s why the Sanskrit word katha, meaning ‘story,’ cannot even be brought into the picture. Even if it were worth considering, we would have to reject it because the meaning ‘story’ bears no sense in the context (more on this context in a moment). The hypothetical *khatha doesn't seem to exist at all as a Sanskritic form.

Zhijé Collection, vol. 1, p. 158 (click to enlarge)

This very same passage was discussed over a decade ago in a comment to Sam van Schaik's blog Early Tibet. You may want to look back at that, but I should say, at the time I was more concerned about the Seven Seals, and had no idea to offer about the kha-tham let alone its weird ending “ṃḥ” that makes it look so much like a mantra.* But that’s a technical point that would require bringing in some arcane Sanskrit grammar as well as little-known mantric science, so we won’t go down that curious track right now. 

(*... an anusvāra followed by a visargaas if that were something that could happen. Well, it does happen in the Kālacakra system, look here.)

By the way, just like you or so I suspect, I’m still fascinated by the whole idea of the Seven Seals. It brings to mind not only one of Ingmar Bergman’s most famous movies — remember that scene of playing chess with death? Of course you do — but the Book of Revelation that came two millennia before the film. Buddhologists will think of the Seven Seals in a story told in the Perfection of Wisdom sutras that could be about the same age as Revelations or a couple of centuries earlier, Who knows? I hope to blog on that particular topic soon. Today I’ll just point to the complete set of them named in that page from the Zhijé Collection (its actual scribing dated ca. 1245), followed by the text-ending expression “That's all,” before you get the word kha-tham. There is no observable difference in penmanship, so arguments  dismissing it as idle scribbling by a later hand hold no water.*

(*And besides, if you are conversant with written Tibetan, you can do an experiment and put it in the search-box at TBRC/BDRC/BUDA. In case you don’t know how to do that, I'll do it for you, just click here. There is a lot to contemplate, but you can see that the kha-tham is very often used in conjunction with sealing terms, particularly in Nyingma texts found by tertöns.)

Ready to hear my solution? 

The word means ‘seal,’ but in what language? Not Tibetan, not Indic, but Semitic. That’s right, it’s a word that three Middle Eastern languages share: Arabic, Hebrew and Aramaic. It’s the Hebrew I’m most familiar with, where the infinitive lakhtom has the dictionary ‘root’ form (=3rd person singular past) khatam, and it’s similar with Arabic and Aramaic. It’s the very word used in the well-known expression ‘seal of the prophets,’ applied to Prophet Muhammad peace be upon him: khātam an-nabīyīn. Of course the Semitic word was borrowed into Persian, which might seem a more likely transit station, especially if it arrived in Tibet via India. However that may be, it has even more ancient roots in old Egypt, where it was being used far before the time of Solomon three thousand years ago, in fact over four thousand years BP.*

(*See Schultz listed below. Oddly, although Mesopotamia is justly famed for cylinder seals, my etymological dictionaries tell me nothing about corresponding cognates being used in either Sumerian or Akkadian.)

Tibetan doesn’t possess an uvular consonant such as you find in the Semitic terms (well, perhaps some northeastern dialects do), but Tibetan language’s best way of representing it would have been the aspirated ‘kh.’

I recognize that there is bound to be a certain amount of resistance from people who haven’t sufficiently realized what can happen, who might overestimate Tibet’s historical isolation. I know it might seem like the philological counterpart to archaeology's misplaced artifact. What is that doing here?  My point exactly. What is that doing here? In any case, it’s there.

I think this foreign word for ‘seal’ is acting as a sealing expression, every bit as much as the seven sealing expressions that came before it. It is saying, ‘This seals it.’ It is also saying, ‘If you aren’t the intended recipient of this communication, you had better not read it.’ 

Not that this quite seals my argument, since if you still remember I promised that there would be a connection between two words. That means maṇḍala al-mandal moving from east to west, and now khatam → kha-tham moving from west to east. Al-mandal in the Arabic magical text from Yemen (and Ethiopia) mentioned in the last blog is done in wax. I’m no expert in the field, but in the course of the magical operation a golden seal might be impressed onto the center of the design, and may also be used to mark the wax candles. So there is something connecting the two word borrowings, as both seem to have a lot to do with the history of Solomonic magical lore, where the Seven Seals have a well established home as well. But we’ll save that seven-fold set for another day, especially since I can’t even imagine how to sort out all the problems associated with showing where they had their start, how they wandered, got reinterpreted, and became so widespread, leaving their stamp all over the place. No need to overstretch ahead of time, is there? I think we have enough to think about for now.

So rest up and save some energy for our next blog. It will be about the very same page of the Zhijé Collection, if all goes according to plan, about those Seven Seals that appear just above the kha-tham.

§   §   §

Relevant references

Regine Schulz, “Seals and Scarabs,” contained in: Ian Shaw and Elizabeth Bloxam, eds., The Oxford Handbook of Egyptology, Oxford University Press (Oxford 2020), pp. 367-408, at p. 368: 

“The ancient Egyptian language has several function-oriented, partly interchangeable terms for seal: ḫtm, which derives from the verb ḫtm ‘to lock, close, seal’, names the ‘sealdevice,’ the ‘impression,’ and also the ‘sealed item.’ In the Old Kingdom (c. 2686–2125 BCE), the term described cylinder seals, but was later used for all seal types, including signet rings.”

The hieroglyph looks like this: 

It’s supposed to depict a cylinder seal
attached to a necklace so it doesn’t get lost or ‘borrowed.’

The following little essay also points out the Egyptian origins of the word:  Zsofia Buda, “Speaking to Angels: Charaktêres in Jewish Magical Manuscripts – Part I,” posted at Rylands Blog, on January 25, 2022.

This blog by Jonas Sibony has a handy list of the word forms that appear in Semitic tongues, supplying them in their original scripts, which is useful and delightful.

Quite a few instances of the Tibetan form of the word may be found by simply doing a TBRC search for "kha taM" without the quote-marks, or a general internet search with them.

I highly recommend consulting the lengthy entry “K̲h̲ātam, K̲h̲ātim” in the 2nd ed. of the Encyclopaedia of Islam, although I have to say, even if this encyclopedia is rightly regarded as reliable most of the time, it is technically off the mark in finding Aramaic origins for the word, as we know Egyptian usage is attested from a time before Aramaic was known to exist:

K̲h̲ātam, K̲h̲ātim  (a.) (P. muhr), seal, signet, signet-ring, the impression (also k̲h̲atm) as well as the actual seal-matrix; it is applied not only to seals proper... indeed anything with an inscription stamped upon it may be called k̲h̲ātam. The word k̲h̲ātam is said by Nöldeke, Mandäische Grammatik, 112 to be of Aramaic origin, and in this he is followed by Fraenkel...”

The dozen-year-old page in Sam’s blog is here:  I haven't yet seen his recent book Buddhist Magic: Divination, Healing, and Enchantment through the Ages, but hope to read it before long. It’s possible to view an hour-long lecture he gave on his book, and the 950-ish CE book behind his book, h...e...r...e. Toward the end of it, in the Q&A section, you can hear him saying how magical technologies such as mandalas, mantras and mudras were appropriated to Buddhist soteriological purposes, clearly placing the magic prior to the Buddhist usages of them. In terms of explaining Vajrayāna origins, we could call it the magical primacist position.

  • I looked into Berthold Laufer’s “Loanwords,” and didn’t locate anything relevant, so that’s time that could have been better spent. I’m acutely aware that “loanwords” itself is a misuse of language, as no language has ever had an honest intention of giving them back. In fact, loanwords are nothing but purloined words, English has gotten very fat on them. But even so the source language doesn’t suffer the least impoverishment, doesn’t particular miss the words that were stolen from it, it just goes on to use them all the more. So perhaps we ought to invent a new concept of sharewords. You think we can get that started?

I noticed that Gwendolyn Hyslop’s (et al.) new and actually not yet published dictionary of Kurtöp-Dzongkha-English has this entry on p. 5:

“katham n. བཀའི་ཐིའུ་; kai thiu; royal seal.”

I think this Kurtöp word "katham" may be most simply accounted for as a loan from chancellery Persian as used in India in more recent centuries, particularly under the Mughals. I wouldn’t lend it much more significance than that. Still, I suppose if we are weighing possibilities about routes of transmission from Egypt to Tibet, Indian employment of Persian could be one piece of the puzzle. If you are listening to me, it was more likely transmitted in a magical context than one of bureaucratic officialdom.

Here are the details on the ca. 1245 CE manuscript as it has been put into print (the Nepal-German Manuscript Preservation Project made a microfilm directly from the original that is far better to work with):

Zhijé Collection  —   Kun-dga’ et al., The Tradition of Pha Dampa Sangyas:  A Treasured Collection of His Teachings Transmitted by T[h]ug[s]-sras Kun-dga', Kunsang Tobgey (Thimphu, Bhutan), in 5 volumes, with English preface by Barbara N. Aziz.

· · ·

I might have to apologize to some of you for not remaining silent, but I simply must mention an amazing crystal seal associated with Mar Mani (216-276 CE), apparently one made to be used by Mani himself, that you can see and learn a little about by tapping here

The concept of Manichaean primacy is briefly mentioned in an article that is mostly about the interesting developments and variations in ideas about what the seal of the prophets means in Islamic traditions’ ways of thinking: 

Simeon Evstatiev, “On the Perception of the Khātam al-Nabiyyīn Doctrine in Arabic Historical Thought: Confirmation or Finality,” contained in S. Leder, et al., eds., Studies in Arabic and Islam, Orientalia Lovaniensia Analecta no. 108, Peeters (Leuven 2002), pp. 455-467. 

However, it isn’t so simple that we can just say Manichaeism set a precedent for Islam in this regard and leave it at that. Gedaliahu G. Stroumsa has written this fascinating paper: 

“Seal of the Prophets: The Nature of a Manichaean Metaphor,” Jerusalem Studies in Arabic and Islam, vol. 7 (1986), pp. 61-74. 

According to him, Manichaeans had an idea of “three seals,” those of mouth [diet, abstaining from evil words], hand [abstaining from killing anything at all] and purity [chastity].* They also had “four light seals”: love, faith, fear and wisdom (with wisdom herself very significantly called the “seal of the burxan [prophets]”). 

The Manichaean phrase in question means ‘seal of [Mani's] apostles,’ likely referring to those who would follow him in the future (not necessarily to his predecessors, nota bene). An apostle, for him, was ranked higher than a prophet or a [mere] teacher. Mani never really called himself “seal of the prophets,” but what he did say was that wisdom was the seal of the burxan...**

(*This probably ought to be mouth, hand and body, in other words and in a different order, body, speech and mind, a very important triad in Buddhist thinking, particularly in relation to ethics as we have it here. Manichaeans drew a lot of inspiration from Buddhist ideas, or at least found ways to work with them. **If you need to know what burxan means, see this brief article by Berthold Laufer.)

Now I see Wikipedia has an entry for “Sealstone of Mani.” Have a look there, too. I haven’t yet been able to read the full article about it by Zsuzsanna Gulácsi. It seems she has the latest word on the subject, and her work on Manichaean art history is outstanding.

Wednesday, May 04, 2022

Maṇḍalas of Medieval Jewish Magic


Book of Secrets manuscript, New York
Public Library no. 190, page 167

This blog ought to add some persuasive force to the preceding one on Maṇḍalas of Medieval Arabic (and Latin) Magic. After posting it, I wrote to a professor at Tel Aviv University, Gideon Bohak, author of an impressive book with the title Ancient Jewish Magic. His quick answer took my breath away. He not only knew of a Jewish example of a maṇḍala in a magic book, he sent the manuscript page with a sketch for one, labelled with the word mandal in larger Hebrew letters in the upper right corner. I feel humbled but then again slightly intelligent just because I knew enough to write the right person with my question. He not only knew of this manuscript — as he informed me it is available online — he studied it and published an edition of it.

The manuscript you see here is not in Hebrew, no matter how much it may appear to be. It’s in what is sometimes known as Judaeo-Arabic. That means it was written by an Arabic speaker more comfortable writing their Arabic in Hebrew letters. It has the title Book of Secrets. The language is Arabic, although Hebrew terms might appear here and there, and even, as we see here, at least one Sanskrit word. Oh, and some Greek terms, too. It appears it is little more than a lengthy book of magical prescriptions.

The title makes us think of a different magic book: Noah received the transmission of a Book of Secrets from an angel prior to the Great Flood and inscribed it on a sapphire slate. This he placed in a gold box he took aboard the ark, and after the flood was over passed it on to his descendants until it finally reached the hands of King Solomon, who evidently was the most remarkable in ability to make use of its magic. It dates from the Second Temple, at least in its essentials, preserving elements from that period, but possibly dating as late as the 7th century, according to G.B. 

But no, that older Book of Secrets is a different one from this one with the mandal. I say that because it is likely to confuse other people as it also confused me until G.B. set me straight. The two books of secrets share the same title. They have nothing especially in common besides the title and of course the general subject matter.

Title page of the Book of Secrets in
the New York Public Library

The New York Public Library’s Book of Secrets is a collection of recipes put together by the 15th-century scribe. Its significance right now is this: It supplies us with yet another example of the medieval Indian Buddhist maṇḍala moving far to the west, not just as a word, but as a corresponding object, a device used in ritual contexts. That is enough of a point to make for now.

§  §  §

On the frontispiece: The Digital Collection of the New York Public Library is the source of the page from the manuscript that you see at the top of this blog. You should go have a look at the link to find out more, but just let me say that the scribing was done in the year 1468 CE by one Mosheh ben Yaʻaḳov ben Mordekhai or משה בן יעקב בן מרדכי בן יעקב בן משה.  I assume this scribe must have been equally responsible for the sketch illustrated above, even if he likely copied from an earlier book.

A side note:  The title Sefer ha-Razim (ספר הרזים), or Book of Secrets, makes use of a Hebrew word borrowed from Persian raz, ‘secret.’ It occurs to me that one of many words in Sanskrit that mean ‘secret’ is rahasya, although I don’t propose to prove any linguistic connection, just to suggest the possibility. And now that I check into it, I’m hardly the first, since a lexicon by Georg Rosen entitled Elementa Persica beat me to it by one hundred and eighty years.

More to read

Gideon Bohak, Ancient Jewish Magic: A History, Cambridge University Press (Cambridge 2008). The brief part about the earlier Book of Secrets on pp. 170-177 is most recommended, and is the main source of my knowledge about it.

Gideon Bohak, A Fifteenth-Century Manuscript of Jewish Magic: MS New York Public Library, Heb. 190 (Formerly Sassoon 56) - Introduction, Annotated Edition and Facsimile [in Hebrew], Sources and Studies in the Literature of Jewish Mysticism no. 44], Cherub Press (Los Angeles 2014), in 2 volumes.

Michael A. Morgan, Sepher Ha-Razim: The Book of the Mysteries, Society of Biblical Literature, Scholars Press (Chico CA 1983). An inexpensive English translation of the more ancient and well known Book of Secrets.

• Prof. Bohak told me about some more relevant publications I had neglected to mention, with yet another article suggested by Gal Sofer, a doctoral candidate at Ben Gurion University. There are still more to be found, I know there are, but here are the ones I think to be most relevant in case you want to look into the matter of westward moving magical maṇḍalas some more:

Vajra Regan, “The De  consecratione lapidum: A Previously Unknown Thirteenth-Century Version of the Liber Almandal Salomonis, Newly Introduced with a Critical Edition and Translation,” The Journal of Medieval Latin, vol. 28 (2018), pp. 277-333. I haven’t yet seen this apart from an abstract on the internet. Notice the author’s first name Vajra, of Sanskrit origins.

Anne Regourd, “Al-Mandal as-Sulaymānī appliqué: Une section interpolée dans le Ms. Sanna 2774?”  The Arabist: Budapest Studies in Arabic, vol. 37 [Studies in Memory of Alexander Fodor] (2016), pp. 135-152.

Anne Regourd, “Images de djinns et exorcisme dans le Mandal al-Sulaymānī” [with text edition, and translation], contained in: Jean-Patrice Boudet, Anna Caiozzo and Nicolas Weill-Parot, eds., Images et magie: Picatrix entre Orient et Occident, Honoré Champion (Paris 2011), pp. 253-294.

Julien Véronèse, L’Almandal et l’Almandel latins au Moyen Âge: Introduction et éditions critiques, Micrologus’ Library no. 46, Salomon Latinus no. 2, SISMEL Edizioni del Galluzzo (Florence 2012).

Note: On May 8, 2022, I received some corrections from Gideon Bohak that lead me to remove and rewrite this blog before reposting it.

Follow me on