Most of us probably think we know what a dragon is, even if we’ve never seen one.1 A dragon has a scaly, reptilian body, but one that is incongruously combined with features of mammals (horns, whiskers, mane), or in some cases, of birds (feathers). It is usually larger than a person, sometimes much larger, although there are notable exceptions. Dragons usually have two legs, but may have four. They have wings, so they can fly. They breathe fire, and guard treasures.
Figure 1
European dragon
Illustration by Orlyn EsquivelThis brief description would likely satisfy most readers who lack the curiosity to inquire further, and it suffices as a common description of European dragons. But is it adequate as a description of all dragons? To someone steeped in the rich iconography of classical Chinese civilization, where dragons adorn almost every temple roof, are common subjects of paintings, and are stitched into the silk fabric of imperial robes, the European dragon sounds vaguely familiar, but it differs in a number of details from the one s/he knows.
First, the body of a Chinese dragon is typically far more sinuous than its European counterpart. One might characterize it as serpentine, or snake-like, as compared with the stocky, almost dinosaurian bodies of dragons in European art, where we see damsels in distress, and Christian knights rushing to their rescue, lance in hand, ready to slay the loathsome monster. Second, Chinese dragons almost always have four legs, while European dragons often do not. Third, they are portrayed with little, if any, indication of wings. Yet these wingless dragons fly with as much ease (or greater ease) than any of the winged dragons of Europe. In fact, they must, since Chinese dragons are universally characterized as controllers of the weather, which requires them to be in the sky among the clouds—something that goes almost completely unmentioned in descriptions of the dragons of European literature and folk tradition.
Figure 2
Chinese dragon
Illustration by Orlyn EsquivelThe latter point requires a bit of elaboration. European and Chinese dragons differ not only in their physical features, but also in their behavior, or in the attitudes that people have toward them. Even in Classical Greece, with few exceptions European dragons were regarded in a negative light. In later times they are clearly opposed to the cultural values represented by the likes of St. George, and are plausibly seen as representing the rejected belief systems of a pre-Christian past. Because they are seen as controllers of the weather, Chinese dragons are associated both with the positive and the negative aspects of rainfall, hence with fertility and with drought or flooding. In general, however, despite some ambivalence, they are auspicious creatures, not treated at all with the horror and rejection that pervades many medieval accounts of dragons in Europe. And, of course, the five-toed dragon became a symbol of the Chinese emperor, adding necessarily favorable features to its already basically positive image in connection with fertility. Europeans think of killing dragons as a noble sport, the common pastime of knights and saints, while to most Chinese this idea seems absurd, or as something deserving of punishment. Why would anyone destroy the source of the fertilizing rains that bring prosperity to the human community?
In short, European and Chinese dragons are not identical in imputed physical traits, in behavior, or in the attitudes surrounding them. What conclusion do we draw from this? Are there two partly overlapping, but separate categories of mythical creatures that just happen to share some striking traits despite the differences that distinguish them, both of which happen to be called ‘dragons’ in English? Or do we conclude that categories are inherently composed of variable members, and that category membership is determined by significant overlap of features rather than by identity?
To the average reader this discussion may seem unnecessary. Many may say that it is obvious that European and Chinese dragons are the same type of creature, so there is no need to argue the point. However, the question then becomes ‘Where do we draw the line, and on what general principles do we justify where we draw it?’ A belief in ‘dragons’ has been reported from many other areas, and here the deviations from a common pattern may become more extreme than those seen so far, although many popular books on dragons open the door wide to all of these variants. A brief survey of dragon reports from other areas will serve to introduce the range of variation.
Since Islam has tended to expunge most earlier folk beliefs, little information can be obtained about dragons from modern Islamic societies of the Middle East. However, pre-Islamic depictions of creatures that are called ‘dragons’ appear in various sources. Drawings of the ‘dragon of the Ishtar Gate of Babylon’ dating from the sixth century B.C. have been reproduced by many writers (Ingersoll 1928:25, Hogarth and Cleary 1979:12, Newman 1979:111, Jones 2000:11, to cite a few). This is a horned, four-legged creature with a long tail that looks thoroughly mammalian except that its body is covered with scales, and it extrudes a forked tongue. It thus shares with both European and Chinese dragons the property of being chimerical (that is, composed of the parts of different animals), but instead of having a basically reptilian body with mammalian or avian embellishments, its body is mammalian and the embellishments reptilian.
Similarly, the Babylonian creation epic depicts an Armageddon between the god Marduk (portrayed in human form), and his mother Tiamat, who is a four-legged non-human mammal with wings, body scales, horns and fangs (Hogarth and Cleary 1979:15). Tiamat is in charge of the release of waters from the sky, and so is reminiscent of the Chinese dragon’s control of the weather. Notably, as will be seen in due course, Marduk wins the battle by attacking his monstrous mother, not with earthly weapons, but with thunderbolts.
Figure 3
Ancient Near Eastern dragon
Illustration by Orlyn EsquivelFigure 4
Vedic Indian nāga
Illustration by Orlyn EsquivelAnother variant of the dragon is attested on the Indian subcontinent. These creatures, called nāgas, are sometimes described as semidivine beings that are half human and half cobra. They thus differ from most other dragons in lacking horns, hair, feathers, or other features of non-human mammals or birds. However, their human part makes them—like all dragons—a fusion of a cold-blooded and a warm-blooded animal. A second feature of the Indian nāga that sets it apart from other dragons is that its serpent form specifically resembles the cobra, the most salient snake of this part of the world, sometimes being described as a multi-headed cobra rather than a cobra with a human upper torso. Finally, the discussion of the Indian nāga as a dragon is complicated by its historical intersection with Buddhism, which may have played a role in introducing some Indian ideas about the dragon into China. Despite being snake-like beings, the nāgas clearly had human behavioral traits, and are described in classical Indian accounts as being among the worshippers of Buddha (de Visser 1913:1). Allen and Griffiths (1979:41) further note that
The Indian Nāgas, although serpentine, were sufficiently dragon-like to interbreed, at least mythologically, with the Chinese, and later the Japanese dragons .... The nāgas were water-deities. They were sometimes associated with sacred pools or wells, in which case they were presumed to live underwater at the base of an adjacent tree; and were sometimes sea-gods. Their palaces were opulent and beautiful and often took the form of whole cities, encrusted with jewels, there they occasionally entertained men who managed to find their way there.
This association with pools or wells is one that will come back again repeatedly in looking at both rainbow serpents and dragons around the world, for when they aren’t in the sky as rainbows, these mysterious creatures lurk in deep pools, lakes or rivers, and somewhat surprisingly, they often guard treasures.
Even this brief discussion shows enough differences between the European dragon, the dragon of Babylon, the Indian nāgas and the Chinese dragon to prompt some ‘splitters’ to divide them into several distinct categories. However, there are also enough similarities that are unlikely to be purely arbitrary that most writers on the topic have united them into a single ‘dragon’ category. As Ingersoll (1928:26–27) put it nearly a century ago:
This wonder-beast ranges from Western Europe to the Far East of Asia, and in the view of a few extremists, even across the Pacific to America. Although in the different localities a great number of most varied ingredients enter into its composition, in most places where the dragon occurs the substratum of its anatomy consists of a serpent or a crocodile, usually with the scales of a fish for covering, and the feet and wings, and sometimes also the head, of an eagle, falcon, or hawk, and the forelimbs and sometimes the head of a lion. An association of anatomical features of so unnatural and arbitrary a nature can only mean that all dragons are the progeny of the same ultimate ancestors.
Ingersoll’s statement that the dragon idea has a single origin, whether through shared history or common psychology (he doesn’t distinguish the two) is perceptive. However, his remark about ‘extremists’ shows that he himself regarded the dragon as an essentially Eurasian phenomenon, reaching from the Atlantic to the Pacific in varied forms, but not extending beyond this continental landmass. However, most other writers do not seem to share this restrictive view. Hogarth and Cleary (1979:200–201) briefly note reports of dragons in North America, various parts of Africa, and even New Zealand, Huxley (1979:8–10, 22, 91) repeatedly refers to the Mesoamerican plumed serpent as a dragon, and G. Elliot Smith (1919) famously included both the North American horned water serpent and the Mesoamerican plumed serpent as types of dragons in connection with his radical diffusionist theory that the dragon idea spread around the planet from Egypt.
Beliefs about the horned water serpent were traditionally ubiquitous in aboriginal North America. Briefly, this creature was viewed as a large snake with horns, which are sometimes described as ‘multicolored’, as among the Creek Indians from the American Southeast. Unlike Eurasian dragons the horned water serpent is never portrayed as flying, but like them it is a jealous guardian of springs, lakes or other water sources. It is thus purely terrestrial, but is always aquatic, and—like all dragons of Eurasia—has a body that combines reptilian and mammalian features.
Figure 5
North American horned water serpent
Illustration by Orlyn EsquivelUnlike the horned water serpent of North America, but like Eurasian dragons, the plumed serpent of Mesoamerica is said to fly. However, unlike each of the above, it normally lacks horns. And, like the Chinese dragon, but unlike the European dragon or horned water serpent, it is a controller of the rains (Wisdom 1974:392–394, Huxley 1979:9–10). In the American Southwest the two creatures seem to overlap, as among the Zuni Indians of western New Mexico, where it is said that “In the sea and underground waters dwells the water serpent, conceived of as a plumed or horned snake. This monster can bring rain, and may impregnate bathing women” (Hultkrantz 1987:97).
Here, the weather-controlling function reminiscent of the Chinese dragon appears again, along with the body cover typical of the Mesoamerican plumed serpent, and the horns characteristic of the North American horned water serpent.
Figure 6
Mesoamerican plumed serpent
Illustration by Orlyn EsquivelHow can we unite all of these creatures? Or should we keep them apart? Despite local variations in details of body form or behavior, there is clearly a common thread that runs through this population of fabulous beasts. All Eurasian dragons, the North American horned water serpent, and the Mesoamerican plumed serpent share the property of combining reptilian physical traits with those of mammals, birds, or both. Everywhere except in the ancient Near East the basic body form is that of a reptile, with mammalian or avian embellishments; in those limited examples we have from the Babylonian creation epic and early visual portrayals from Mesopotamia, the basic body form is instead that of a mammal, with reptilian embellishments. Huxley (1979:10) has perhaps put it best: “This combination of a hot-blooded and a cold-blooded animal aptly sums up the dragon’s compound nature of fire and water.”
There is one apparent exception to this statement that all dragons and only dragons are chimera that combine the body parts of warm-blooded and cold-blooded animals. This is the Scottish nixie, or ‘water horse’, a fabulous creature that has other similarities to the dragon, but has no reptilian component in its physical make-up. However, as will be seen in due course, the nixie can be included in the dragon category both because it guards terrestrial water sources and preys on young women like several other dragons, and also because its connections with a reptilian body are indirectly attested through the dragon-horse of China and Japan, which have the physical characteristics of both dragons and horses.
It is surprising how perceptive various writers have been in identifying the contradictory nature of dragons, without however seeming to understand how those contradictions could have arisen from observations of the natural world. As Huxley says, the dragon’s ‘compound nature’, mingling the visual attributes of warm-blooded and cold-blooded animals, can be reduced to the statement that the dragon is, in some sense, a blend of “fire and water.” But how can a mythical animal that breathes fire and controls rainfall in widely separated parts of the earth be made of fire and water, which are mutually incompatible? Doesn’t water extinguish fire? And doesn’t fire cause water to boil? This would be puzzling enough if such a belief were restricted to one geographical region, where it might be assumed to be an arbitrary invention of a particular culture. But, as many writers have pointed out, the dragon idea is universal in the sense that its geographical distribution cannot plausibly be explained by common origin among demonstrably related peoples, or by borrowing. We are therefore left to conclude either that it is a creation of the human mind that has mysteriously persisted in many of the world’s cultures for tens of thousands of years, or that something has motivated the invention of this creature again and again throughout human history.
To answer the question posed at the beginning of this chapter, then, a minimal definition of what a dragon is would be ‘a mythical creature compounded of bodily elements from cold-blooded and warm-blooded animals’. This suffices for European, ancient Near Eastern, Chinese, North American and Mesoamerican creatures that have been called ‘dragons’ in various publications over the past century and more. Other imaginary creatures are also chimerical in combining body parts of different animals, as the unicorn. However, in this case both the equine body and the horn are traits of mammals. What cannot be stressed enough is that dragons differ from other chimerical creatures in showing a fusion of body parts from both cold-blooded and warm-blooded models in nature. No other creature shows this peculiarity of form, which can be seen as the physical embodiment of a contradictory nature.
This unique characteristic provides us with a minimal definition of what makes a dragon a dragon, but surely, we would like to add more detail to it. At first it might seem that the most straightforward way to do this would be to draw up a list of traits that have been reported for dragons anywhere in the world. To keep the model maximally simple in the interest of clarity let’s consider only European and Chinese dragons. We start with the skeletal framework already established, namely that the creature we are calling a ‘dragon’ combines reptilian with mammalian or avian traits. What can we add to this? Among their better-known traits, European dragons 1. fly, 2. breathe fire, 3. guard treasures, 4. prey on young women, etc. Similarly, Chinese dragons 1. fly, 2. control the weather, 3. often carry a pearl in their mouths, 4. are fond of eating roasted swallows, etc. Based on this preliminary comparison we can compile a list of traits as follows:
-
the dragon combines reptilian with mammalian or avian traits
-
it flies
-
it breathes fire
-
it guards treasure
-
it preys on young women
-
it controls the weather
-
it often carries a pearl in its mouth
-
it is fond of eating roasted swallows
A moment’s reflection will show that something should trouble us in thinking of using this list of traits or any other list like it to define the nature of dragons. Just because a trait is attributed to a dragon in a single culture does not imply that it is in any way intrinsic to the nature of a dragon. Points 7 and 8 make this particularly clear. These traits have never been reported outside Chinese culture, or the larger ‘sinosphere’ (the area in which Chinese cultural influence was historically dominant, as Vietnam, Korea or Japan). Why then should they be considered traits of dragons in general rather than as features of dragons that were independently invented in China? Based only on this preliminary (and deliberately simplified) inventory, the only traits that can safely be attributed to dragons are 1 and 2: they are mythical creatures that combine features of cold-blooded and warm-blooded animals, and they can fly.
While this gets us further than simply relying on point 1 as the defining feature of dragons, we can surely add further detail. Prodding deeper we find other points of agreement that are less apparent on the surface. European dragons breathe fire in a very conspicuous way, something that is much less commonly seen with Chinese dragons. Nonetheless the same trait is reported in the Chinese classics, where we are told that dragons breathe fire, but not fire of the ordinary kind. As de Visser (1913:67) expresses it, “Dragon fire and human fire are opposite. If dragon fire comes into contact with wetness it flames, and if it meets water it burns. If one drives it away by means of fire, it stops burning and its flames are extinguished.”
Similarly, European dragons are well-known for guarding treasures, especially hoards of gold. Familiar examples are the fire-breathing dragon that guarded the golden fleece sought by Jason and the argonauts, and the later dragon Fafnir, who guarded a golden treasure trove in the medieval Icelandic Völsunga saga and its Middle High German equivalent, the Nibelungenlied. Although Chinese dragons are not commonly portrayed in this capacity, Ingersoll (1928:88) points out that in Korea a terrestrial dragon was formerly thought to be the protector of mines and gems “and the intense regard for it is perhaps the chief reason why mines have been so little worked in Chosen, the people superstitiously fearing that disasters may follow disturbance of the metals which they believe are peculiarly the treasure of the jealous earth-spirit.” Whether this is a native Korean belief, or a product of cultural diffusion from China remains unclear, but in any case, the much-debated pearl that Chinese dragons carry in their mouths may well be equivalent to a treasure in European accounts of the dragon (Allen and Griffiths 1979:39).
Without looking further we can perhaps stop here and conclude tentatively that dragons:
-
combine reptilian with mammalian or avian traits
-
fly with or without wings
-
breathe fire
-
guard treasures
Points 5 and 6 from our longer list will eventually be justified as intrinsic traits of dragons as well, but to show this it will be necessary to expand our comparison to a wider area.
Although it may not be immediately clear, there is an important point of scientific method that governs this process of identifying the traits that count as intrinsic traits of dragons (or any other phenomenon that involves comparative data). Traits that are unique to one area must be considered independent inventions in that area until further evidence shows that they are not, while shared traits that are widely separated in space or time or both, as the four cited above, must be attributed to 1) common origin in one ethnolinguistic group that passed them on to its descendants, 2) diffusion, or borrowing from one unrelated group into another, or 3) parallel development. Intellectually, the most challenging of these options for many has been the last: why would the same seemingly arbitrary ideas arise repeatedly in different parts of the world? Evolutionary biologists refer to this phenomenon as ‘convergence’. Examples include the evolution of various kinds of anteaters (aardvark, armadillo, tamanduá, echidna) from very different ancestral animals. In biology convergence is explained by parallel adaptations to a similar environment, but what could motivate convergence in belief systems?
As will be seen in greater detail below, it was this problem that led G. Elliot Smith (1919) to his theory that the dragon was an invention of the Egyptian priesthood that was carried around the planet as the Egyptians supposedly crossed continents and oceans, spreading their religious ideas as they went. This approach, which assumes transcontinental or transoceanic contact influences, is generally known as radical diffusionism. We will look at the merits of radical diffusionism as an explanation for the widely-shared traits of the dragon in a later chapter. However, if this alternative is eliminated as a plausible explanation for the distribution of draconic traits, and if common origin is similarly eliminated (a relatively easy matter in cases such as the European and Chinese dragons, given the clear biological and linguistic differences of the populations concerned), we are left with what cultural anthropologists traditionally call ‘independent invention’.
In some cases independent invention is not a vexing issue. That the bow may have been invented independently in many places from cultures that previously used only spears does not trouble most researchers: clever individuals, wherever they were, recognized at some point that a smaller version of the spear could be propelled to greater lengths if released from a tensed wooden lath. Similarly, it is not surprising that the idea of a supreme being has arisen again and again in human societies, since this serves to answer such basic questions as ‘where did we and the world around us come from?’ But why would an idea like that of the dragon emerge repeatedly in pre-Christian Europe, the ancient Near East, India, China, North America and Mesoamerica, to mention only some of the areas in which it is found, without any convincing evidence that it was spread by contact? Needless to say, this question has generated extensive speculation over the past 130 years, spawning theories that have varied from radical diffusionism, to psychological archetypes imprinted on the pre-human mind when our remote mammalian ancestors were simple arboreal primates.
It is important that we treat all theories seriously, whether we find them convincing or not. What will guide our decisions about the value of theories is their ability to explain observations in a straightforward manner—nothing more and nothing less. The observations that must be explained are those that, like 1–4 above, describe traits of dragons that are shared across widely separated cultures, and hence require a foundation rooted in perceptions of the natural world. In other words, there must be a reason why dragons in geographically distant areas combine reptilian with mammalian or avian physical features, fly, breathe fire and guard treasures, among many other traits yet to be discussed. Fanciful as these traits may seem at first, if they are widely shared (as these are) they must be motivated by observations that are available to all human beings regardless of location, language, or culture. Stated differently, they cannot be products of freewheeling psychology or imagination, but must derive from some aspect of a universal human response to the observable world.
We have now reached our first plateau—a tentative definition of what a dragon is. Every other conclusion reached from this point on must be based on the same type of logic, namely are widely-shared features of dragons most likely to have arisen from 1. a single cultural community that spread out over the world, carrying the dragon idea with it, 2. the borrowing of ideas about dragons from one community to another across continents until they encircled the planet, or 3. separate, parallel developments motivated by the similar reactions of human beings everywhere to features of the natural environment that they sought to explain without the benefit of science?
The last thing we need to do in this chapter is perhaps to provide a preliminary indication of how many dragon traits must be regarded as inherent because they are found in at least two widely separated regions. The most convenient way to do this is in a table. This is done below for six geographical regions that can be considered the ‘core’ areas from which dragons have been reported, although as will be seen, chimerical serpents that are indistinguishable from features of the natural environment are also known in tribal societies elsewhere, and when these are considered, further traits will be proposed that are found in just one of the core areas below plus a non-core area. Note that EUR = Europe, ANE = the ancient Near East, SA = South Asia, CEA = Central and East Asia, NAM = North America and Mexico, and CSA = Central and South America:2
Table 1
Preliminary geographical distribution of draconic traits
EUR |
ANE |
SA |
CEA |
NAM |
CSA |
|
---|---|---|---|---|---|---|
1) |
? |
+ |
+ |
+ |
+ |
+ |
2) |
+ |
+ |
+ |
+ |
+ |
|
3) |
+ |
+ |
+ |
+ |
+ |
|
4) |
+ |
+ |
+ |
+ |
+ |
|
5) |
+ |
+ |
+ |
+ |
+ |
+ |
6) |
+ |
+ |
+ |
+ |
+ |
|
7) |
+? |
+? |
+ |
+ |
||
8) |
+ |
+ |
||||
9) |
+ |
+ |
+ |
+ |
+ |
|
10) |
+ |
+ |
+ |
|||
11) |
+ |
+ |
+ |
+ |
+ |
|
12) |
+ |
+ |
+ |
+? |
+? |
|
13) |
+ |
+ |
+ |
|||
14) |
+ |
+ |
||||
15) |
+ |
+ |
||||
16) |
+ |
+ |
+ |
+ |
||
17) |
+ |
+ |
||||
18) |
+ |
+ |
+ |
|||
19) |
+ |
+ |
+ |
+? |
||
20) |
+ |
+ |
+ |
+ |
||
21) |
+ |
+ |
+ |
|||
22) |
+ |
+ |
+ |
|||
23) |
+ |
+ |
+ |
+ |
+ |
|
24) |
+ |
|||||
25) |
+ |
+ |
+ |
|||
26) |
+ |
+ |
+ |
+ |
+ |
+ |
27) |
+ |
+ |
||||
1) = is a giver/withholder of rain |
||||||
2) = is a guardian of springs or other bodies of water |
||||||
3) = lives in caves |
||||||
4) = can fly |
||||||
5) = has scales |
||||||
6) = has horns |
||||||
7) = has hair (mane, whiskers, etc.) |
||||||
8) = has feathers |
||||||
9) = is opposed to thunder/lightning or the sun |
||||||
10) = is bisexual/androgynous |
||||||
11) = is colorful/red |
||||||
12) = guards a treasure |
||||||
13) = has a jewel/other valuable on its head |
||||||
14) = lives in waterfalls |
||||||
15) = encircles the world |
||||||
16) = terrifies women/can impregnate them with demonic child |
||||||
17) = is offended by menstruation |
||||||
18) = is connected with hoofed mammals |
||||||
19) = has fiery breath |
||||||
20) = has fetid or toxic breath |
||||||
21) = causes earthquakes |
||||||
22) = causes whirlwinds and storms |
||||||
23) = causes floods |
||||||
24) = is a sign of war |
||||||
25) = causes sickness, disease or trouble |
||||||
26) = has human traits |
||||||
27) = can be personified |
Some of these traits are widespread, as 1) ‘giver/withholder of rain’, 2) ‘guardian of springs or other bodies of water’, 3) ‘lives in caves’, 4) ‘can fly’, 5) ‘has scales’, 6) ‘has horns’, 9) ‘is opposed to thunder/lightning or the sun’, 11) ‘is colorful/red’, 12) ‘guards a treasure’, 23) ‘causes floods’, and 26) ‘has human traits’, all of which are found in at least five of the six regions considered here. Other traits are rare (or perhaps under-reported), as 8) ‘has feathers’, reported only in the western part of North America and in Mesoamerica, 17) ‘is offended by menstruation’, reported in the areas covered here only in the Americas (where it is widespread, although it also occurs in other areas that we will examine later), 24) ‘is a sign of war’, which is unique to Europe (but which we will find in other locations in Chapter 7), and 27) ‘can be personified’, which is found only in Europe and South Asia, among the areas considered here. As will be seen in later sections, there is another fundamental division within this list, namely that traits 1–15 are readily, or at least plausibly, explainable by reference to features of the natural environment, while traits 16–27 are not.
This is already a rather bewildering variety of traits that need to be reconciled with one another, which is perhaps why many general books on the dragon mention only a few of them and ignore the others. No dragon has ever been reported with all of these traits, or even a majority of them. Rather, the need to posit them as inherent features of dragons is required by their appearance in widely-separated geographical areas where their presence cannot plausibly be a result of borrowing or common origin in a single ancestral community.
Before concluding this chapter it is important to stress that although my narrow aim here is to define what is or is not a dragon, my broader aim is to define how any category is formed and justified, and to better understand the arguments that follow, it will be worthwhile to take a moment to consider this point. Categories, by their nature, are collections of entities that vary to differing degrees. The Austrian philosopher and logician Ludwig Wittgenstein (1889–1951) is remembered, among other things, for his concept of ‘family resemblance’, which holds that members of a category may be connected by overlapping similarities, where no feature is common to all of them (Wittgenstein 1953). He models this in several ways, one of which is the following:
Item 1 |
A B C D |
Item 2 |
B C D E |
Item 3 |
C D E F |
Item 4 |
D E F G |
Item 5 |
E F G H |
In Wittgenstein’s terms, then, it is clear that items 1 and 5 are connected through a linking relationship, even though they share no similarity that is revealed by direct comparison. This type of connection was used by the linguist Joseph Greenberg (1957:41) to argue that the demonstration of a historical relationship between any two languages may face shortcomings that can be overcome by including a larger number of units of comparison. Reduced to its minimal expression Greenberg argued that ‘If A is related to B, and B is related to C, then A is related to C’, even where evidence of relationship has been entirely lost through change. In this way he argued that both the Hadza and Sandawe languages of Tanzania, which were traditionally treated as isolates (languages having no known relatives) are each related to the Khoisan (‘click’) languages of southern Africa, and that they are therefore related to one another. Subsequent linguistic scholarship has deconstructed the ‘Khoisan family’ into several language families, and has returned the classification of Hadza to its original position as a language isolate, but none of this affects the logical integrity of the argument that ‘If A is related to B, and B is related to C, then A is related to C’, as the same formula can be applied in cases where it may turn out to be true.
Ruhlen (1987:118) calls this linking of category members by overlapping shared properties a relationship of ‘transitivity’ (if item 1 is related to item 2, 2 to 3, 3 to 4 and 4 to 5, then item 1 must be related to item 5, despite the absence of any common property). In general, what is important in establishing a category is to find a common thread that unites all its members by some distinctive feature that sets them apart from all non-members, regardless of how different they are in other respects. Few people would have a problem including sparrows, robins, crows and eagles in the category ‘bird’, but the definition of ‘bird’ becomes more difficult when penguins, kiwis and cassowaries are added to the mix. Nonetheless, all and only birds have feathers. For the most part all dragons can be defined in this way: they are mythological creatures that combine the body parts of cold-blooded reptiles with those of warm-blooded mammals, or less commonly, birds, and no other creature, real or mythological, shares this property. Nonetheless there are some ‘outlier’ dragons, such as the Scottish nixie or ‘water horse’, and I will appeal to Wittgenstein’s concept of ‘family resemblance’, or the principle of transitivity, to show that they belong to the category ‘dragon’. Because of its fundamental importance, this question of category membership will also be discussed briefly in later sections, in particular at the end of Chapter 5, and again in the concluding chapter of the book.
For now it is perhaps enough to say that the patchwork pattern shown in Table 1, which includes overlapping shared traits extending from Europe through South and East Asia to the Americas, suggests that the idea of the dragon is either an extremely ancient inheritance, or that it arose repeatedly in different areas through observation of common features of the natural environment and similar modes of thinking about man’s relationship to it. Before examining the origin of the dragon idea, however, it will be worthwhile to first survey the wide range of theories that have been proposed to account for the global distribution of this belief, and then to document in at least minimal detail some of the traits that are simply checked off in the above distribution matrix, which no existing theory has accounted for, or even addressed.