Watery Abyss

For technical reasons I’ve had to miss Sky & Telescope for several months. Now at last I have the current issue, which – though it’s now January – is for March. (Many features of fast-paced modern life are shifted into the future; it has something to do with relativity.)

In the magazine there is an interesting article, “The Very Ancient Origin of the Water Constellations,” by Craig Crossen, which I wish I’d had a chance to see before I wrote about Capricornus, the constellation that looks like a boat but is supposed to be a fish-tailed goat

As Crossen says, these watery constellations – which he lists from west to east as Capricornus, Delphinus, Aquarius, Piscis Austrinus, Pisces, Cetus, Eridanus, Argo, and Hydra – all in ancient times lay on or south of the celestial equator. And that makes it plausible that their association with water dates back all the way to the Sumerians, the ancient people of southern Mesopotamia (now southern Iraq). To the south of them lay the Persian Gulf; and the southern zone of the sky was sacred to Enki, their god of the water and of the abyss beneath the earth. An additional large watery constellation was Centaurus, which for them was a gigantic Marsh Boar, no doubt associated with the marshland where the Euphrates and Tigris debouch into the Gulf (and where live the Marsh Arabs – perhaps descendants of the Sumerians, and nearly extirpated by Saddam Hussein.)

Yes, these constellations that we sometimes call the “Ocean of the Sky” were, back then, along or below the equator. (Actually, all except Delphinus still are. And Delphinus only barely touched the equator, leaping from it – appropriately for a dolphin.)

I thought I’d plot a picture to clarify this. Here are the Watery Constellations (including Centaurus), drawn on the imaginary celestial sphere. The boundaries are the modern ones, fixed by nineteenth- and twentieth-century astronomers, but they fit the traditional areas. The vast ancient constellation of Argo the Ship was broken into Puppis, Vela, and Carina (its poop, sails, and keel).



After trying to picture all of this band of constellations in one sphere, and finding that it’s difficult to prevent those on the nearer side from obscuring those on the farther, I broke the sphere into halves – or, you could say, the sphere viewed from opposite directions and with its front cut away.

The positions are precessed to 4000 BC, when the Sumerian civilization was developing but had not yet invented writing. The stars and constellations stay where they are, and the ecliptic also stays virtually fixed; it’s the celestial equator that is slewed around by the long slow precession, or twisting, of Earth’s axis. In six thousand years the equator has slewed more than a sixth of the way around the sky. The point where the equator crossed the ecliptic – where the Sun is at the spring equinox – was, back in 4000 BC, toward the left (east) end of Taurus; it traveled though Aries and is now far along toward the right (west) end of Pisces.

I tried plotting the AD 2000 equator onto the same pictures, and even a horizon for a Sumerian evening in 4000 BC, but decided these would overwhelm clarity. I have to say it has taken me a full day, with countless reruns (seventy?), to reduce these pictures to sufficient correctness and simplicity. Ars celat artem.

Crossen adds that Capricornus was originally a boat with a goat’s-head prow, before it morphed into a goat with a fish tail. Such boats, sacred to Enki, are referred to in Sumerian religious texts from around 2200 BC. That’s what I wish I’d known a week ago. Middle Eastern archeology was my passion from age ten to about sixteen, but it’s long since I caught up on the literature.


Rough sketchmap of Sumer (also known as Sumeria). There are many more of these mounds that were once cities, rich in archeological information. Light blue represents the area that was sea and has, since ancient times, silted up (I don’t really know the accurate extent of it).  Rivers like the Euphrates and Tigris, in floodplains, often shift their courses; probably Eridu was at the mouth of the Euphrates, then was abandoned (still before the beginning of history) when the river shifted course. Babylon was a city of the Akkadians who coexisted with and then absorbed the Sumerians.

A further claim is that the mysterious name of Eridanus, the sky’s lengthy River, descends from Eridu, southernmost and most ancient of the Sumerian cities, and home of the worship of Enki. And that Eridanus earlier was regarded as ending not at the star Achernar (Arabic âkhir an-nahr, “last of the river”) but at brighter and more northerly Canopus. And that Canopus was the Sumerians’ “Star of Enki.” Eridanus, then, was the heavenly reflection of the Euphrates, ending at Eridu. “Eridanus, some scholars have pointed out, is phonetically similar to Eridu,” writes Crossen rather feebly.

It obviously is, and all this fits together with a compound plausibility and would be beautiful if true, though I don’t know whether it has been generally accepted. The first Greek letter of Eridanus is eta, the long open kind of e, which tends to alternate with a; the Akkadian version of Eridu was Iritu, which suggests a close vowel. The Greeks identified Eridanus with the Po, or an amber-bearing river somewhere northward in Europe, or with the world-encircling river of Ocean, or with a brook near Athens.

Achernar certainly was invisible to the Sumerians in 4000 BC, being even farther south than it is now by, I figure, 25 degrees – only 8 degrees from the south celestial pole. Canopus, which is now at declination -53, was farther south by 6 degrees, so from Sumeria’s latitude of about 30 it briefly peeped no more than a couple of degrees above the horizon. Catching sight of it would have been as much of a feat as it is now from Virginia. Could the great southern Star of Enki have been Rigel or Sirius?

For the Sumerians presumably the world was flat, so the celestial equator itself would have had no meaning for them, except possibly that the Sun rose and set on it, due east and west, at the equinoxes. But these constellations were the ones that struggled up from the southeastern horizon and stood in the south over the Marsh and the sea, and sank to the southwestern horizon. You could well have imagined that their home was Enki’s watery abyss that underlay the world.

That underworld, from which Enki emerged to found Eridu and civilization, was Abzu, and his temple in Eridu was E-Abzu, “house of the abyss.” Yet the word abyss seems to be simple Greek: a-byssos, “no bottom.” Is that another coincidence? Incidentally, since an abyss is not just deep but infinitely deep, it’s a bit of a misnomer when we term the floor of the ocean an “abyssal plain.”


3 thoughts on “Watery Abyss”

  1. The idea the watery constellations we currently recognise may have had a very ancient origin is certainly tempting, and has been a subject for speculation since the 19th century, when some of the Mesopotamian cuneiform texts dealing with astronomy were first translated. However there are problems with the concept, judging by the current star and constellation identifications from those cuneiform originals (taken here chiefly from Hermann Hunger & David Pingree’s Astral Sciences in Mesopotamia, published by Brill in 1999, which I think remains the most complete modern scholarly summary of the topic).

    The Mesopotamian versions of these constellations were:

    * Capricornus = The Goat-Fish.
    * Delphinus = perhaps The Dead Man.
    * Aquarius = The Great One, which had Front and Rear Basket stars, respectively perhaps Phi or maybe Chi Aquarii, and Lambda Piscium.
    * Piscis Austrinus = The Fish.
    * Pisces was split in three:
    * Northeastern Pisces = Anunitu. Anunitu was the name of a humanoid Babylonian childbirth goddess. The constellation though included a Fin (unidentified star or stars), and a Bright Star in the Ribbon of Fish (possibly Eta Piscium), perhaps indicative of a similar appearance to the later Pisces, if with only one fish.
    * Western Pisces + probably Epsilon, Zeta and Theta Pegasi + Alpha Equulei = The Swallow, which had stars for a Head (unidentified) and Tails (plural; maybe Lambda Piscium).
    * Lambda Piscium = possibly The Rear Basket of The Great One, as well as/instead of The Tails of The Swallow.
    * Cetus = no identifiable constellations.
    * Eridanus = no identifiable constellations.
    * Hydra = The Snake.
    * Argo Navis:
    * Carina = no identifiable constellations.
    * Part of Puppis + Pyxis + part of Canis Major + Canis Minor = The Arrow. Curiously, The Arrow was assigned an Elbow (perhaps Alpha Pyxis) and a Left Foot (maybe Pi Puppis).
    * K Puppis + Omega, Delta, Sigma and Epsilon Canis Majoris = probably The Bow. (Hunger & Pingree listed the four CMa stars as being in Canis Minor, and gave Kappa Pup too. From other sources, these seem to be in error, especially as the Greek-letter stars in CMi end with Eta. Alternatively, Kappa CMa, which lies near the modern border with Puppis, may have been intended by their “Kappa Pup”).
    * Part of Puppis + part of Vela = Eridu, which was allocated Hands (Gamma Velorum, perhaps).
    * Eastern Vela = The Harrow.
    * Vela probably including and near Phi Velorum = Ninmah. This was also the name of the goddess who was midwife at the birth of humankind. Her connection here is uncertain, though the constellation possessed a Hand (suggested as Phi Velorum).

    Elsewhere, the Mesopotamians recognised The Crab as our Cancer, and The Barque as probably Epsilon Sagittarii, of their other more likely water-related star patterns. Leaving aside the somewhat stranded Crab, this reduces the identifiable Mesopotamian “Sky Ocean” to just PsA, Cap, southwestern Sgr, maybe Aqr (The Great One’s Baskets could have been caulked for carrying water, if we’re being charitable), and the somewhat isolated northeastern Psc.

    Centaurus reinterpreted as The Marsh Boar is intriguing, given that most of Centaurus (and possibly Crux) seemed to have been named as the untranslatable Habasiranu, in Sumerian Entenabarhum, who had a Head (Delta Cen?) and a Left Foot (unidentified star) at least. The “Left Foot” need not have implied a bipedal form, however. Either variant was possibly a personal name, although the reading of the cuneiform is not certain, and some modern authors have suggested it may have meant “mouse”. If correctly identified, Epsilon or Nu Cen, and Mu Cen were respectively called Hanish and Shullat, the names given elsewhere to two minor deities who were ministers of the Storm God, Sumerian Ishkur, Akkadian Adad. Another, rather obscure, possible Storm God, Numushda, was the name for a further star in Centaurus, perhaps Eta or Kappa. Mesopotamian storm deities tended to be associated with a winged lion-dragon as their preferred totemic animal, not a boar.

    Presumably the idea of a boat with a goat’s-head prow originated from the name given to Enki’s processional barge called the “Ibex of the Abzu”, although there are no descriptions of this boat suggesting it had a goat-headed prow. The Mesopotamian Goat-Fish constellation (more literally “Carp-Goat” from the Sumerian “Suhur-mash”) seems not to have had any earlier known boat-like precedent though. It was provided with a Horn Star (Beta Cap?), a Front (Gamma Cap?), Middle (unidentified) and Rear Star (Delta Cap).

    Odd that Capricornus’ watery nature, if known to them, was apparently dropped by the Greeks to become a Wild Goat, Aigokereos in Aratus’ Phainomena. His text did warn against sea voyages in the winter solstice month when the Sun passed through this constellation, albeit by Aratus’ time, the 4th-3rd centuries BC, precession meant this had been no longer physically true – solstice point in mid-constellation – for 1500 years or so. Conversely, he reinforced Aquarius’ watery connection by providing as the adjacent constellation Hydor, The Water.

    Sumerian Nunki was the name for Eridu celestially, although whether this linked the Eridu constellation and the Mesopotamian city is less assured. Hunger & Pingree did not mention Canopus however, but other authors have done so in this context, and the star would have risen for a short time as seen from Eridu city by circa 3000 BC (refraction might have made it a little easier). Whether that could have made it a, or the, Star of Enki is another matter.

    There were several groupings of constellations into the “Paths of” three leading gods in Mesopotamia by the second millennium BC, a regimentation of the sky which fitted quite poorly with the apparently pre-existing constellations, the Paths of Ea (the Akkadian equivalent of Enki), Anu and Enlil. While Ea’s Path always occupied the more southerly regions of the sky in these, its constellations were not always the same. Thus some versions included in the Path of Ea The Arrow (presumably with Sirius) and The True Shepherd of Anu (Orion, thus probably with Rigel, although as Lepus was perceived as The Rooster, it’s conceivable Rigel may have been incorporated with those stars instead), but omitted the heavenly Eridu.

    The somewhat later (perhaps late second, or more definitely early first, millennium BC), and more comprehensive, Mul Apin constellation compendium set Eridu in the Path of Ea, but The Arrow and True Shepherd of Anu were assigned to the Path of Anu. In addition, Mul Apin linked deities with various of the constellations, so The Fish (PsA), The Great One (Aqr) and Eridu in the Path of Ea were associated with the god Ea, but The Field (now the Square of Pegasus) was called “The Seat of Ea”, yet it lay in the Path of Anu! So any of these latter four star groupings could be considered “Stars of Ea” (i.e. Enki), but then so might any of the constellations considered by one or other Mesopotamian source as in the Path of Ea. Mul Apin’s list of these comprised, in the sequence as written: The Fish (PsA), The Great One (Aqr), Eridu (parts of Pup + Vel), Ninmah (part of Vel around Phi Vel), Habasiranu (most of Cen), The Harrow (eastern Vel), Shullat and Hanish (Mu, and Epsilon or Nu Cen respectively), Numushda (Eta or Kappa Cen), The Mad Dog (Lup + part of Sco), The Scorpion (Sco), Pabilsang (Sgr + part of Oph, who had a Left Hand on the Bow – Delta Sgr – a Right Hand on the Arrow – Phi Sgr – a Bright Star on the Tip of the Arrow – Theta Oph – and a Sting – unidentified star – thus was an archer, and possibly one of scorpion-man form, as known from Mesopotamian art), The Barque (Epsilon Sgr), and the Goat-Fish (Cap). Thus it may be in the Mul Apin codification of the three sky Paths that the first glimmerings of what would become the Ptolemaic “sky-ocean” constellations originated. Ancient, but not that ancient, perhaps.

    1. Yes, and I used the word “plausible” rather than “probable” because I don’t know to what extent Craig Crossen’s suggestions can be accepted. I suspect that there is,more depth to the scholarship of Hunger and Pingree (and McBeath).

      While writing about this I started belatedly to learn a little Sumerian.

  2. Nice piece!
    And good timing, since I am teaching Tor about Mesopotamia and consequently reading about Enki and Ninhursaja.

Leave a Reply to K Cancel reply