I found this cryptic Sumerian proverb looking for ancient barley references:

"The cemuc barley is reserved for the necromancer."

There is an unrelated note that follows explaining how barley was used as commodity money in ancient Mesopotamia, not really relevant to necromancy. The closest I came to any definition of the word "cemuc" was to find a waterfall named "Cemuc" in Guatemala,but suspect that word might be Mayan.

My online search for botanical articles and cemuc barley yielded nothing. It could be a typo. Does anyone know what cemuc barley is, and why it would be reserved for necromancy?

2 Answers 2


The C in cemuc is how the ETCSL transcribes the letter that is more often transcribed as SH or Š, so you'll see cemuc spelled with these letters instead (i.e šemuš or shemush).

What it exactly is I don't think we really know. It's a type of barley, but which doesn't seem clear, or too important.

However, it was used in magical rituals, especially in the creation of a zisurrû:

Zisurrû, meaning “magic circle drawn with flour,” and inscribed ZÌ-SUR-RA-a, was an ancient Mesopotamian means of delineating, purifying and protecting from evil by the enclosing of a ritual space in a circle of flour. It involved ritual drawings with a variety of powdered cereals to counter different threats and is accompanied by the gloss: SAG.BA SAG.BA, Akkadian: māmīt māmīt, the curse from a broken oath, in The Exorcists Manual, where it refers to a specific ritual on two tablets the first of which is extant.

The circle is rationalized in commentaries as representing certain protective deities, LUGAL.GIR.RA and Meslamtae’a according to one. In other rituals a circle might be painted in whitewash or dark wash on either side of a doorway for apotropaic purposes. The choice of flour was crucial to the purpose of the ritual, with šemuš-flour reserved (níĝ-gig) for repelling ghosts, wheat-flour for rituals invoking personal gods and šenuḫa-barley to encircle beds, presumably to counter disease-carrying demons.


Let me continue the trail from where cmw left off

TL;DR: The actual Sumerian name for this type of barley was probably šeĝuš ("shengush"; although šemuš is also a possible reading), which was loaned into Akkadian as šigūšu, šegūšu or šeguššu. All we really know about it is that it was most likely some type of barley, although the fact that it's often mentioned in parallel with še'eštub ("spring barley", Akkadian arsuppu) suggests that it might be a variety harvested later in the year.

The obvious first thing to do, when faced with an unfamiliar Sumerian word, would be to check the Pennsylvania Sumerian Dictionary. However, the ePSD search interface is somewhat quirky (to be polite) and may require some familiarity and exploration to actually find what you need. In particular, just typing "cemuc" into the ePSD dictionary search brings up no matches.

However, with a bit of knowledge of Sumerian, and the fact that this word is supposed to describe some type of barley (and that "c" is actually supposed to be "š"), we can make the educated guess that this word is probably a compound of še (𒊺), meaning "grain" or "barley", and some cuneiform sign that can be read as muš, such as:

  • 𒈲 (MUŠ), with various uses, the most common and basic probably being muš = "snake";
  • 𒈹 (MUŠ3 = INANA), probably best knows as the logogram for the goddess Inanna (𒀭𒈹); or
  • 𒋀 (MUŠ5 = ŠEŠ), most commonly used to write the word šeš = "brother".

In particular, if you instead search ePSD for "ce-muc" (or "ce-muc5") with a hyphen to separate the signs, you'll get an entry for "še-muš = šeĝuš [BARLEY]", with the following line at the top:

šeĝuš [BARLEY] (20x: ED IIIb, Old Akkadian, Ur III) wr. še-muš; še-muš5 "a barley" Akk. šigūšu

Unpacking this dictionary entry a bit, we can tell from it that:

  • Yes, this is an actual attested Sumerian word.
  • The editors of the Pennsylvania Sumerian Dictionary believe it should actually be transcribed as šeĝuš rather than šemuš. (The letter "ĝ" in Sumerian transcription denotes a velar nasal consonant, more or less like the /ŋ/ sound commonly written "ng" in English.)
  • It is attested from 20 different sources, dated to the Early Dynastic IIIb, Old Akkadian and Ur III periods. (The table on the page below the dictionary entry shows how many times each spelling is attested in each period.)
  • It is written in cuneiform as 𒊺𒈲 (še-muš) or 𒊺𒋀 (še-muš5).
  • The corresponding Akkadian word is šigūšu (or presumably šigūšum in the Old Akkadian period, before final mimation was lost). Given the similarity to the Sumerian word and the lack of a transparent Akkadian etymology, it is presumably a straight phonetic loanword from Sumerian.
  • It means some kind of barley.

Further down on the page there are also citations to two articles by M. Powell, published in 1984 and 1987 in "BSA", which according to this handy list of assyriological abbreviations refers to the Bulletin on Sumerian Agriculture. Alas, I haven't managed to track down either of these publications yet. (I did find various PDF copies of the Bulletin online, but not those specific issues.)

However, perhaps the most useful thing in the ePSD entry is the Akkadian translation, since having it means that we can look it up in an Akkadian dictionary such as the Concise Dictionary of Akkadian (CDA) or the far more comprehensive Chicago Assyrian Dictionary (CAD), both of which can be downloaded in PDF form for free.

Looking up šigūšu in the CDA just gives us this brief entry, which is not much more informative than the one from ePSD:

šigūšu, šegūšu, šeguššu 1. O/jB (a type of barley) [ŠE.MUŠ5] bread, flour of š. 2. jB lex. (a kind of apple tree)?; < Sum.?

Unpacking this again, we get the following additional information:

  • a couple of alternative transcriptions (šegūšu, šeguššu), which may reflect actual historical variation in pronunciation and/or ambiguities in the Akkadian cuneiform writing system;
  • a note that the word could also be used to describe bread or flour made from this type of barley;
  • a possible secondary meaning ("a kind of apple tree") only attested in lexical lists (basically a type of ancient dictionary) in late Standard Babylonian (a.k.a. jungbabylonisch = jB), whereas the primary meaning ("a type of barley") is also attested in Old Babylonian (OB); and
  • a confirmation at the end that this is indeed likely a loanword from Sumerian.

However, the CAD (specifically volume 17: Š (part 2)) gives a lot more information. Looking up šigūšu on page 414 just redirects us to the spelling šeguššu, which the CAD editors apparently prefer, but the entry for šeguššu on pp. 261–262 gives a trove of details and attestations.

I'm not going to try to quote the whole entry here, but it starts with:

šeguššu (šegūšu, šigūšu) s.; 1. (a cereal), 2. (a kind of apple tree); OB, MB, SB; Sum. lw.; wr. syll. and ŠE.MUŠ5 (ŠE.MUŠ)

Again we have the variant transcriptions, two meanings, attestations from Old, Middle and Standard Babylonian, and a note that this is a Sumerian loanword and that it could be written either syllabically (according to the Akkadian pronunciation) or with the "sumerograms" 𒊺𒋀 (ŠE.MUŠ5) or 𒊺𒈲 (ŠE.MUŠ).

This is followed by a couple of lexical attestations, such as "še-mu-uš ŠE.ŠEŠ = ši-gu-šú" from the Diri V lexical list.

As noted above, these lexical lists were essentially ancient dictionaries, typically listing a bunch of semantically similar Sumerian words (e.g. plants, animals, cities, deities, etc.), their approximate Sumerian pronunciation written with syllabic signs, and often a translation or a gloss in Akkadian. These lists were frequently copied by student scribes as part of their training, which is why they have survived in many (but often incomplete) copies.

The entry from the Diri V list, for example, indicates that according to this particular list, the Sumerian word written as 𒊺𒋀 (ŠE.ŠEŠ = ŠE.MUŠ5) was pronounced (approximately) as šemuš (or šeĝuš; the sign 𒈬 = MU = GU10 could be read as either mu or ĝu in Sumerian) and corresponded to "šigūšu" in Akkadian.

What these lexical lists do not usually include, however, is any details about what these words actually meant. Presumably the meaning of all the words would've been known to the scribal students copying such lists, or at least something they could ask their teacher to explain. Alas, those teachers have been dead for thousands of years, and often the best we can do now is try to infer the meaning of uncommon words from the (typically semantically similar) words listed near them in the same lexical lists, or from occasional non-lexical attestations. That's why modern Sumerian and Akkadian dictionaries have so many entries that are just translated as e.g. "a plant", "an animal" or "a type of priest", because all we really know about them is that they appear in a list of such things.

For šeguššu, however, the CAD does provide several non-lexical attestations as well. From these, we can infer e.g. that this šeguššu:

  • was grown on fields,
  • could be ground into flour and used to make bread,
  • could be used to make beer,
  • could be loaded on wagons,
  • could be fed to horses,
  • was often often mentioned together with arsuppu (a type of early barley), inninu / enninu (also a type of barley) and sometimes other cereals (wheat, emmer, etc.), and
  • also had plenty of ritual and/or medical uses (to which about half the CAD entry is devoted).

In particular, a note at the end of the CAD entry for šeguššu says:

Since the word often occurs beside arsuppu early barley, šeguššu may denote the late barley, see Civil, Nippur 11th Season (= OIC 22) 130.

("OIC" here refers to the Oriental Institute Communications of the University of Chicago, which are also conveniently available online, so this is a reference I can actually check. However, looking at issue 22, pp. 129–130, there really isn't much there. Civil just translates a fragment of a model contract, reading "30 gur še-eštub / 20 gur še-muš5", as "30 gur of early grain and 30 (sic!) gur of late grain", noting that "the grain še-muš5, almost always next to še-eštub, is unusual but not unknown in administrative texts", with citations to several earlier publications that presumably include these further attestations.)

  • Thank you for providing not only an extensive reference, but a marvelous resource for future research. Feb 27 at 17:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.