Origins and Nature of the First Dragons

Since ancient times, they were seen as the fantastical embodiment of chaos and evil, and they seem to be an important part in legends and myths in almost every culture ever since the cradle of civilization. People have pointed at crocodiles, dinosaurs and whales as the creatures that might have inspired the dragon myth, but it is generally acknowledged that the snake is probably the creature that shares most of the characteristics of the first dragons. In fact, according to the Merriam-Webster dictionary the term “dragon” comes from the Greek word “derkesthai”, meaning “to see” or “to look at”, a reference to the perpetual gaze of the snake since it has no eyelids, and it might have in turn coined the word “drakōn”, a term that was used to indistinctly designate a snake or dragon. Once more, we must remember that in the ancient languages, people usually used the same term to refer to either a dragon or a snake: it happened with the Sumerian word “ušum”, the Sanskrit “ahî”, the Avestan “aži”, the Hittite “angu”, etc. However, there was evidently a point in which the dragon became a separate ideological entity that could no longer be identified with an ordinary snake, and when it happened, the first dragons actually represented something totally opposed to the evil and destructive image that people bestowed upon them in later periods.

It is indeed evident that the first dragons were linked to the symbolism behind the snake. After all, snake worship is one of the most ancient kinds of cults. A carved wall at the Rhino Cave in Bostwana (Figure 1) dating to the Late Pleistocene in the Middle Stone Age (12 600-25 000 BC) shows the zoomorphic features of a snake, which is linked with a form or ritualized behavior (1), and the Water Snake was a symbol of water among the Khoekhoen and /Xam San people, two of the oldest ethnic groups in Africa (2). For the ancient people, the image of the serpent was related to water, fertility, life and the underworld, particularly in those regions of Africa and the Ancient Near East that were deeply affected by the cycle of water and the seasonal changes of weather, two essential factors for their survival and development. In some bowls and beakers from the ancient city of Susa (now part of Iran) that date from the 5th millennium BC, the snake seems to be connected with water as a source of life (3). In Ancient Egypt, the snake is a symbol of a rather dual nature since it represents both life and the underworld (4): Atum, the god of creation in Egyptian mythology, is said to reverse in his form as a snake when the world comes to an end (5), and the goddess of harvest Renenutet is depicted as either a cobra or a woman with the head of a cobra (6); furthermore, the Egyptian emblem of immortality known as the Ouroboros is a serpent biting its own tail (7), and in the funerary text known as the Egyptian Book of the Dead (c. 1550 BC) there is a section in chapter 87 explaining that transformation into a serpent upon death gives new life to the deceased person (8).

Rhino Cave

Figure 1 — Rhino Cave southern wall lightened at night [from Coulson, Staurset, and Walker (2011) Ritualized Behavior in the Middle Stone Age: Evidence from Rhino Cave, Tsodilo Hills, Botswana], photo by Sheila Coulson.

The fact that the snakes shed their skin might explain why these animals were seen as symbols of life and renewal, a feature that also appears in the Mesopotamian Epic of Gilgamesh (c. 2100 BC), where a snake instantly rejuvenates and sheds its skin after eating the herb of life (9), thus taking away the hero’s hope for immortality. Similarly, the symbol of Ningizzida, the Sumerian deity of fertility and the underworld, was the snake, and he was also attributed the power to bestow fertility and heal diseases (10); he is sometimes depicted as either a snake or a man with two snakes protruding from his shoulders, a trait that he seems to share with Nidaba, a Mesopotamian goddess of harvest who is also represented as a woman with snakes coming out of her body (11). Harvest and fertility divinities that are represented together with snakes are also common in Ancient Canaan, where the findings of bronze serpents at temples in Meggido dating from 1650-1100 BC indicate that they were associated with fertility cults, and the fertility goddesses from Canaan are usually represented in art as women with snakes in their hands or with snakes entwined around their bodies (12). And even though in the Bible the snake represents evil when connected with the devil, in the Book of Numbers, God told Moses to raise a bronze serpent on a pole so that any person who was bitten by a snake and saw the bronze sculpture would live (21:4-9), thus becoming a symbol of God’s power until king Hezekiah ordered its destruction because some of the people of Israel had started to do offer to it (2 Kings, 18:4).

As stated before, the terms dragon and serpent were often used interchangeably in ancient mythology to describe the same being, yet it was in Mesopotamian iconography where the physical features of the dragon —as different from a snake— were first conceived. The impressions of a cylinder seal (Figure 2) dating from the Uruk IV period (c. 3300-3100 BC) show pairs of monsters with long necks intertwined next to scenes of people who seem to be some kind of craftsmen. These creatures appear again in greater detail on another seal of the same period (Figure 3), this time with their tails also intertwined and with the lion-headed eagle flying above them (as shown on the pictures below). It is worth noting that these beings seem to be connected with ancient representations of intertwined snakes such as the libation vase that Gudea (Figures 4 and 5), king of Sumer, dedicated to the Sumerian god of healing Ningizzida, which displays a staff between two intertwined snakes with winged dragons on each side. As a matter of fact, it is highly likely that this motif would eventually inspire the Rod of Asclepius (a snake coiled around a staff) and the Caduceus (a staff between two intertwined snakes with a pair of wings on its top), the first one considered a symbol of health by the 6th century BC in Ancient Greece (13) while the second was a symbol of commerce during the Roman Empire (14). Therefore, it has been asserted that intertwined dragons and snakes were not seen as evil monsters, but rather as benign creatures in the mindset of ancient people (15).

Dragon Entwined Necks Morgan Museum

Figure 2 — Leather craftsmen working in two rooms divided by entwined snake-necked monsters [cylinder seal], Mesopotamia, Late Uruk period ca. 3500–3100 BC Pierpont Morgan Library, Morgan Seal 1.

Entwined Dragons Uruk Seal

Entwined Dragons Uruk Seal Impression

Figure 3 — Jasper cylinder seal: monstrous lions and lion-headed eagles, Mesopotamia, Uruk Period (4100 BC–3000 BC), Department of Oriental Antiquities, Richelieu wing, ground floor, room 1a, case 2, Louvre Museum, photo credits: Marie-Lan Nguyen (2010) and PHGCOM (2007).

Gudea vase steatite

Figure 4 — Sumerian libation vase in green steatite (2150 BC). The inscription reads: “To his god Ningizzida, Gudea priest-king of Lagash has dedicated this for the prolongation of his life.” Erich Lessing/Art Resource, New York.

Gudea vase drawing

Figure 5 — Drawing of the green steatite vase of king Gudea of Lagash of Sumer [from Touraj Nayernouri (2010) Asclepius, Caduceus and Simurgh as Medical Symbols Part I].

These long-necked creatures appear to be an attempt to blend the physical features of a leopard/lion and a snake (they are sometimes called “serpopards”, though this term is merely a modern invention since there is no mention of such a word in the ancient sources) so as to represent a particular ideological entity composed of discordant elements that should not strike the beholders as entirely bizarre (16); in this sense, the artists of Ancient Mesopotamia chose to weld the main characteristics of the real animals that best resembled the nature and attributes of this first dragon: the sinuous neck and forked tongue of the snake, which represented water, fertility and life; the body, paws and head of the lion, which was a symbol of power and strength (17); and, in time, the wings of the eagle.

When dragons were eventually portrayed in art together with other divinities, particularly with weather and fertility gods, they were not represented as their antagonists but rather as their companions, their servants and even as their own representations; as  matter of fact, since the Early Dynastic Period (2900-2350 BC) the dragon became one of the four symbolic beasts that were associated with storm gods such as Enlil, Ningirsu, Ninurta, Iskur, etc. —the others being the bull, the lion, and the lion-headed eagle (18). We can see the dragon in terms of amity with other deities in several artistic manifestations throughout the region. A stone impression dating to the Akkadian Period (2334-2154 BC) found at Tell Asmar shows two weather gods driving a plough that is being drawn by a snakelike dragon and a lion, both working as their servants (Figure 6). Also, at this very same place archaeologists found a clay impression depicting a storm god enthroned on a dragon before an altar, together with a minor deity who is introducing a worshipper to this god (Figure 7). This scene of the storm-god on top of a dragon is a recurrent motif that is meant to highlight the close relationship between the deity and the dragon as well as their inherent supernatural attributes. We see a similar scene on another impression, also of the Akkadian period, in which two divinities are introducing a pair of worshippers to a god holding a whip on his hand who is standing on the back of a four-legged horned dragon (Figure 8).

Dragon plough

Figure 6 — Two gods plowing; one holding plow, the other driving span (consisting of snakelike dragon and lion) with left hand, which either holds or is shaped like a scorpion [from H. Frankfort (1955) Stratified Cylinder Seals from the Diyala Region, University of Chicago, Plate 62 No. 654].

Figura 6 Dragon Tishpak

Figure 7 — Ancient impression on clay; god (perhaps Tishpak) enthroned on dragon before fire altar; minor deity introducing worshipper [from H. Frankfort (1955) Stratified Cylinder Seals from the Diyala Region, University of Chicago, Plate 62 No. 649].

Figure 7 mussel shell impression

Figure 8 — Seal impression of the Akkadian period. Two divinities introducing a pair of worshippers to a god holding a whip on his hand who is standing on top of a dragon [from E. Douglas Van Buren (1946) The Dragon in Ancient Mesopotamia, GBPress-Gregorian Biblical Press, Plate I, Figure 5].

Since the Akkadian Period, dragons are depicted with two or three lines coming out of their tongue that look like flames; however, it has been argued that these flames are composed by two or three lines that actually are meant to depict the forked tongue of a snake (19). Whatever the case, from this period on certain features of the dragon, such as its tongue/flames and sinuous neck and tail, were possibly used to further reinforce the relationship between the weather deities and the dragon. Storm gods riding on a dragon usually appear with a three-pronged lightning and a whip on their hands, which might be alluding to the roaring thunder that heralds the upcoming rain, and sometimes a fertility or rain goddess appears between the wings of the dragon holding similar symbols (Figures 9, 10 and 11); therefore, it has been suggested that the artistic representation of these two objects were inspired by the forked tongue, and the undulating body of the dragon (20). Hence, it stands to reason that a dragon, the living symbol of water and streams, is an appropriate attendant of the storm god who brings forth the rain (21).

Chariot dragon

Figure 9 — Dragon drawing the chariot of a god with a nude female deity between its wings [from H. Frankfort (1939) Cylinder Seals, plate XXIIa].

Chariot dragon II

Figure 10— God and goddess riding winged dragons [from H. Frankfort (1939) Cylinder Seals, plate XXIId].

L'iconographie dragon chariot

Figure 11 — Mythical fire-spitting, leonine, winged dragon with lowered head, harnessed to the Storm-god’s chariot [From Vanel, A.  (1966) L’iconographie du dieu de l’orage dans le Proche-Orient ancien jusqu’au VIIe siècle avant Jesus-Christ, p. 23, fig. 5].

Now, it is worth noting that there seem to be two main types of dragon in Mesopotamian art: the snakelike dragon and the leonine dragon, though both of them still share some similarities. The snakelike dragon was the first to appear in iconography, and it can be traced back to the Uruk IV period (c. 3300-3100 BC). The typical features of these dragons are its elongated neck and tail, its flat serpent head and its scales; it sometimes has the body of a snake, though it is also often represented with four legs. The muš-huš, mušmah or mušḫuššu, meaning “fierce snake” or “reddish snake” in Sumerian (22) is usually identified with this kind of dragon. In contrast, the leonine dragon is frequently depicted with a lion’s face, a snake or scorpion’s tail, lion paws and a pair of wings on its back. This dragon replaced the winged lion of the Early Dynastic III Period (c. 2600-2370 BC) as a royal and divine symbol with the advent of the Akkadian period (c. 2334-2154 BC), and it is identified with the ušumgal, meaning “great snake” or “great dragon” in Sumerian (23). Literary evidence of these two dragons can be found in the Epic of Gilgamesh (24) and the inscriptions of Gudea’s Seal (25) which were composed around 2100 BC, thus making them the earliest dragons in ancient literature that can be clearly identified in iconography as such (some may claim that Kur, the Sumerian underworld, was also the first dragon in ancient literature, yet in the ancient texts he is never explicitly called a dragon or even a snake, and his description hardly matches that of a dragon). Both dragons usually have four legs and scaly bodies, they serve as the attendants of different weather gods, and their forked tongues or “flames” seem to match the lightning symbol that the storm gods hold on their hands in subsequent centuries (Figures 9-12).


Figure 12 — Cylinder seal and clay imprint of a storm god (probably Assur) riding a dragon. Steatite, Assyria, 9th-8th centuries BC., Louvre Museum, Department of Oriental Antiquities, Richelieu, ground floor, room 6; photo credits: Marie-Lan Nguyen (2006).

As for the deities that are often associated with dragons, the storm god Ningirsu of the Sumerian city of Lagash is perhaps one of the first to be identified with the dragon. Ningirsu was first mentioned in Eannatum’s Incription in the Early Dynastic III Period (2600-2350 BC.), and since his symbolic artifact is the plow (26), it is easy to link his image with that of the dragon driving his plow or even the storm god’s chariot. Gudea’s cylinders describe his destructive mace šar-ur like “a mušmah (fierce snake/dragon), like the water of the clouds out of the mountains of cedar” (27), and his legendary drum is called the “usumgal-kalama” or “great dragon of the land” (28). Under the name of Ninurta, in the Lugal-e or “The Exploits of Ninurta” (late 3rd millennium BC) he is referred as an “indefatigable serpent hurling yourself at the rebel land” and as a “dragon who turns on himself, strength of a lion snarling at a snake, roaring hurricane” (29), so it might even be plausible that Ningursu/Ninurta could have been represented as either a storm-god or a dragon, or that the dragon was the ideological embodiment of his weapons, the šar-ur and the usumgal-kalama. Other Mesopotamian deities that are associated with dragons in iconography are Tishpak (Figure 7), an Akkadian god of storm and sky and the tutelary deity of the city of Eshnumma; Assur (Figure 12), the Assyrian equivalent of the Sumerian storm god Enlil; and Marduk (Figure 13), the patron deity of  Babylon, yet in this last case Marduk appears confronting a monster that is probably Tiamat (though it has the features of the usumgal or the monstrous Anzud).


Figure 13 — Marduk (Bel) and Tiamat drawn from a bas-relief from the Palace of Ashurbanipal, King of Assyria, 885-860 B.C., at Nimrûd. British Museum, Nimrûd Gallery, Nos. 28 and 29.

So, when is it that the concept of the dragon fighting the storm-god is introduced? Well, the Mesopotamiam myth known as The Exploits of Ninurta or Lugal-e tells us that even though Ninurta was associated with the dragon in the text, he is also known as a fierce warrior who faced several legendary beings:

“Hero, pitfall (?), net of battle, Ninurta, King, celestial mace …… irresistible against the enemy, vigorous one, tempest which rages against the rebel lands, wave which submerges the harvest, King, you have looked on battles, you have …… in the thick of them. Ninurta, after gathering the enemy in a battle-net, after erecting a great reed-altar, Lord, heavenly serpent, purify your pickaxe and your mace! Ninurta, I will enumerate the names of the warriors you have already slain: the Kuli-ana, the Dragon, Gypsum, the Strong Copper, the hero Six-headed Wild Ram, the Magilum boat, Lord Saman-ana, the Bison bull, the Palm-tree King, the Anzud bird, the Seven-headed Snake — Ninurta, you slew them in the Mountains.”

The Exploits of Ninurta, lines 122-134

As we can see, Ninurta was known to have defeated a dragon and a seven-headed snake (which reminds us of the hydra), and the Anzud bird. This last creature is of special interest since there is a seal impression of Ninurta slaying the Anzud (Figure 14), which is depicted as a winged lion-bird very similar to the usumgal (which also appears below the storm-god with the clear intention of assisting his master in the fight) yet it has the tail of a bird, while the usumgal shown on this same image has the distinct forked tongue or flames coming out of its mouth and the long tail of a scorpion or scaly snake. And though we must acknowledge that this impression is from a rather later period (800-750 BC), it supports the idea that the dragon was more of a supportive servant of the storm-god than an aggressor, while the Anzud bird was seen as his enemy in the ancient texts and in iconography.

Ninurta Anzud Usumgal

Figure 14 — Ninurta slaying the Anzud, grey or brown chalcedony cylinder seal in the modelled style, Neo-Assyrian (800BC-750 BC), Middle East Department of The British Museum.

It is actually in the Anatolian region (Asia Minor) where the dragon conclusively took the role of the enemy of the storm-god. The Hattian spring festival known as Puruli was dedicated to the destruction of the serpentine dragon Illuyanka by the storm-god Tarhunt (known as Tarhunna by the Hitites), which reflected the annual event where the dragon, an evil influence that brought drought and famine (this being an area that was deeply affected by insufficient moisture or destructive floods that endangered the livelihoods of their people), had to be slayed in order to have a more plentiful season (30). This theme was developed by the Hattians, the Hurrians and the Hittites, who inhabited the region between 2300-1180 BC, and it can be compared with other Chaoskampf myths such as the battles between Enlil and Kur, Marduk and Tiamat, Set and Osiris, Zeus and Typhon, etc., and though this motif of the struggle between order and chaos entities did not originate in the Anatolian region, it seems to be the first one to attest the conflict between the storm-god and the dragon.

As we have seen, the dragon was regarded an independent benign influence that was associated with water, streams and rain in Ancient Mesopotamia before becoming an attendant of the weather deities while preserving its original attributes. At some point in the late third millennium, it slowly started to become the antagonist of the storm god and civilization (especially in Anatolia), though the iconographical evidence proves that we was still considered a benevolent creature in Mesopotamia until the first half of the first millennium BC. It is possible that the cultural blending that was to be intensified as a natural product of progress and trade incorporated new elements into the dragon myth that eventually turned it into the evil contestant of the storm god, and it is this view that has sturdily persisted in western civilizations.



(1) Coulson, Staurset, and Walker (2011) Ritualized Behavior in the Middle Stone Age: Evidence from Rhino Cave, Tsodilo Hills, Botswana, PaleoAnthropology Society, pp. 45-49.

(2) Hoff, Ansie (1997) The Water Snake of the Khoekhoen and /Xan, South Africa: South African Archaeological Society, The South African Archaeological Bulletin Vol. 52, No. 165, pp. 21-27.

(3) Bonanno, Anthony (1985) Archaeology and Fertility Cult in the Ancient Mediterranean, First International Conference on Archaeology of the Ancient Mediterranean, University of Malta, B.R. Grüner Publishing Company, p. 309

(4) Lurker, Manfred (1980) The Gods and Symbols of Ancient Egypt, New York: Thames and Hudson, p. 108.

(5) Frankfort, Henri (1962) Kingship and the Gods, Chicago: University of Chicago Press, pp. 145-146.

(6) Pinch, Geraldine (2003) Egyptian mythology: a guide to the gods, goddesses, and traditions of Ancient Egypt, New York: Oxford University Press.

(7) Hornung, Erik (1999) The Ancient Egyptian Books of the Afterlife, English translation by David Lorton, New York: Cornell University Press, pp. 38, 77-78.

(8) Lurker, Manfred (1987), “Snakes”, in The Encyclopedia of Religion, ed. Mircea Eliade, New York: Macmillan, 13:373.

(9) A. R. George (2003), The Babylonian Gilgamesh Epic: Introduction, Critical Edition and Cuneiform Texts, Oxford: Oxford UniversityPress, XI:305–307.

(10) Van Buren, E. Douglas (1934), “The God Ningizzida”, Iraq 1, April 1, p. 89.

(11) Langdon, Stephen Herbert (1964) The Mythology of All Races, Volume V: Semitic, ed. John Arnott MacCulloch, New York: Cooper Square Publishers, p.78.

(12) Münnich, Maciej (2008) The Cult of Bronze Serpents in Ancient Canaan and Israel, in The Bible and Its World, Rabbinic Literature and Jewish Law, and Jewish Though Volume 1, IGGUD, Selected Essays in Jewish Studies, Jerusalem, pp. 40-42., retrieved from

(13) Farnell, RL. (1921) The Kult of Asclepius, in: Farnell, RL. Greek Hero Cults and Ideas of Immortality, Oxford: Clarendon Press.

(14) Nayernouri, Toraj (2010) Asclepius, Caduceus, and Simurgh as Medical Symbols Part I, Archives of Iranian Medicine, Volume 13, Number 1, pp. 66.

(15) Van Buren, E. Douglas (1946) The Dragon in Ancient Mesopotamia, Orientalia, NOVA SERIES, Vol. 15, GBPress- Gregorian Biblical Press, p. 4.

(16) Van Buren, E. Douglas (1946) The Dragon in Ancient Mesopotamia, p. 3.

(17) Green, Alberto Ravinell Whitney (2003) The Storm-god in the Ancient Near East, Biblical and Judaic Studies Volume 8, University of California, p. 17.

(18) Green, Alberto Ravinell Whitney (2003) The Storm-god in the Ancient Near East, p. 13.

(19) Van Buren, E. Douglas (1946) The Dragon in Ancient Mesopotamia, p. 10.

(20) Amiet, Pierre (1980) La glyptique mésopotamienne archaïque, 2e éd., Paris: Centre nationale de la recherche scientifique, p. 142.

(21) Frankfort, Henri (1939) Cylinder Seals: A Documentary Essay on the Art and Religion of the Ancient Near East, London: Macmillan and Co., p. 125.

(22) The Electronic Text Corpus of Sumerian Literature, (ETCSL), University of Oxford, Faculty of Oriental Studies

(23) The Electronic Text Corpus of Sumerian Literature, (ETCSL), University of Oxford, Faulty of Oriental Studies.

(24) Epic of Gilgamesh: Gilgamesh and Huwawa, lines 34-47, translation by The Electronic Text Corpus of Sumerian Literature, (ETCSL), University of Oxford, Faculty of Oriental Studies

(25) The building of Ninĝirsu’s temple (Gudea, cylinders A and B), lines 726 and 126 respectively, translation by The Electronic Text Corpus of Sumerian Literature, (ETCSL), University of Oxford, Faculty of Oriental Studies.

(26) Green, Alberto Ravinell Whitney (2003) The Storm-god in the Ancient Near East, pp. 30, 42.

(27) Green, Alberto Ravinell Whitney (2003) The Storm-god in the Ancient Near East, p. 46.

(28) The building of Ninĝirsu’s temple (Gudea, cylinders A and B), lines 162 and 188, translation by The Electronic Text Corpus of Sumerian Literature, (ETCSL), University of Oxford, Faculty of Oriental Studies,

(29) The exploits of Ninurta, lines 1-16, translation by The Electronic Text Corpus of Sumerian Literature, (ETCSL), University of Oxford, Faculty of Oriental Studies,

(30) Green, Alberto Ravinell Whitney (2003) The Storm-god in the Ancient Near East, pp. 149-151.

2 thoughts on “Origins and Nature of the First Dragons

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s