Balrogs are the demons of the ancient world, fearful flame-wreathed servants of Melkor Morgoth, the first Dark Lord -- but what keeps Tolkien fans awake at night is not the fiery swords or whips, the shadow or the flame, but whether Balrogs had wings or not. Here I'd like to add my two penn'orth to the argument. I think that the answer, on balance, is 'not'.

The principal reason is that Tolkien is never explicit about the presence of wings in Balrogs, even though he goes into great detail about wings in other flying and fantastical creatures -- and that one should, as a matter of course, take Tolkien at his word, where possible. The second reason is based on aerodynamics: that is, if the wings were to have been used in flight (which is what the scanty hints that survive tend to lead one to believe) then they would have to have been ridiculously huge -- so huge, that not to have mentioned them at all would have been strange indeed.

The first time that most Tolkien readers will meet a Balrog is on the bridge of Khazad-Dûm in The Fellowship of the Ring, where the two references to wings are contradictory. This Balrog -- the Balrog of Moria -- is initially described as a 'shadow' in the midst of which is a humanoid form, perhaps greater than the height of a Man. It has a fiery mane, and holds a flaming sword and a whip. When the Balrog gains the Bridge, the shadow about it reaches out 'like two vast wings.' These wings are plainly described as the appurtenances of shadow, and they could be just that -- shadows cast when a great dark shape is illuminated from below, for example by the light cast by flames burning in a fiery chasm. The second reference seems more explicit, stating that its wings were spread from 'wall to wall' -- however, given the first reference, these could still be wings of shadow, a cloak of smoke neither more substantial nor capable of flight than Count Dracula's cape.

The case against wings in Balrogs is very strong. Balrogs go right back to The Book of Lost Tales, but none of the many references to them before The Lord of the Rings even hints that they had wings or were actually capable of flight. This is in complete contrast to dragons, where Tolkien makes a clear distinction between wingless dragons (such as Glaurung in the Silmarillion) and flying, winged dragons (such as Ancalagon, again in the Silmarillion, or Smaug in the Hobbit.) As an aside, the name Ancalagon actually appears in the real world -- as the formal name for a genus of sea-worm that lived in the Cambrian Period around 530 million years ago. The creator of the name, Professor Simon Conway Morris of the University of Cambridge, referred explicitly to Tolkien's dragon in its creation.

The vagueness about Balrog wings also contrasts with the very clear description of the appearance -- and the wings -- of the flying steeds of the Nazgûl (described rather ineptly, I think, as 'Fell Beasts' in all the publicity material for Peter Jackson's films.) In The Return of the King these creatures are described in some detail as featherless, with wings made of webs of skin stretched between horned fingers, in other words, somewhat like giant bats. However, in their great size, somewhat reptilian mien and, rather more notably, their description of having survived from some primeval age, these creatures are rather more like pterodactyls (more properly, pterosaurs) the order of flying reptile that became extinct at the same time as the dinosaurs, 65 million years ago.

As an interesting but possibly tangential aside, one of the protagonists in The Notion Club Papers (published in Sauron Defeated) is described as having written a book of poems entitled Experiments in Pterodactylics: given that The Notion Club Papers and The Return of the King were written just one before the other, I think that pterosaurs, rather than bats (still less Balrogs) was what Tolkien might have had in mind for the flying steeds of the Nazgûl.

However, Turgon reminds me that we have here an actual case in which we have Tolkien's own words on the models for these particular winged creatures. Responding to a direct question from a reader about whether these creatures were pterodactyls, Tolkien wrote (Letters, no. 211, 14 October 1958) that this was not, in fact, his explicit intention. Nevertheless, the creatures were -- again in his words -- 'pterodactylic' -- in that the allusion offered a way of portraying them as the last relics of a vanished prehistory. Tolkien also notes that his creation is probably less mythological than the reconstructions offered by palaeontology, that Ònew and fascinating semi-scientific mythology of the 'Prehistoric'Ó. As an ex-palaeontologist myself I must admit that Tolkien's reading of the state of palaeontological thought in the 1950s is sharply accurate (see the book In Search of Deep Time by my great friend Henry Gee for a more detailed discussion of palaeontological myth-making.) To emphasize the ambiguity, Tolkien refers to the creatures himself as 'Nazgûl-birds' (Letters, no. 100, 25 May 1945). This ambiguity should not be seen as authorial indecision or vacillation, but a deliberate strategy that maintains the mystery of the story, and therefore its excitement and allure. To my mind, it would have been a great pity had Tolkien simply admitted that 'Nazgûl-birds' were meant to be pterodactyls. Where's the fun in that? Of course, one could say the same thing about the wings of Balrogs -- part of the enjoyment lies in not being quite sure: on a larger canvas, the fun of scientific enquiry lies in the realization that discoveries never achieve definitive conclusions but only open up more, grander vistas for our exploration.

Back to the Balrogs. Although Tolkien refers to Balrogs repeatedly as parts of his assemblage of assorted monsters, they feature very strongly in just two episodes in the Silmarillion tradition -- their rescue of Melkor from Ungoliant, and their prominent role in the Fall of Gondolin.

Melkor, returning to Middle-earth after the Rape of the Silmarils, is almost overcome by the spider Ungoliant, but is rescued by Balrogs lying in wait for their master's return. The 1951 version of the Silmarillion has the Balrogs passing 'with winged speed' to their master -- a provocative phrase, but which need only refer to the speed of the Balrogs rather than their physical means of achieving it. In any case, the Silmarillion as published is more enigmatic still, saying that the Balrogs rushed to Melkor as a fiery tempest, with no mention of wings at all, even as adjectives. In a thematically similar passage in Appendix A of The Lord of the Rings (under 'Durin's Folk'), the Balrog is described as 'flying' from Thangorodrim, but without an explicit description of wings, this might be no more than flight in the sense of rapid and desperate escape -- a use as metaphoric as 'winged speed'. In any case, Balrogs were Maiar, beings of the same supernatural order as wizards and Sauron, and so could have had abilities to transcend those of the prosaic and corporeal. One presumes that they could have sped through the air much like Superman, who, it has been claimed, flies faster than a speeding bullet, for all that he has only a rather camp costume in which a flimsy red cape is all that passes for wings.

Balrogs lead the final assault on Gondolin, but in all the intense detail of the battle for that city (as described in The Book of Lost Tales), in which the hero Ecthelion dies in the act of slaughtering Gothmog, Lord of Balrogs, and where Balrogs are described as being heavily armored and twice the height of an Elf (and remembering that the Elves at that time were conceived as people of heroic stature), no mention at all is made of Balrogs having wings or being able to fly. When the refugees from Gondolin, led by the brave Glorfindel, are ambushed by a Balrog in the mountain passes, the monster is described as 'leaping' onto the sorry convoy, not once, but twice -- and wings are never mentioned. This is particularly significant in that the eagles of the Crissaegrim join the battle at that point, and rescue Glorfindel from the chasm into which he and the Balrog had fallen, locked in battle (a precursor, perhaps, of the whole Gandalf-Balrog-Eagle episode in The Lord of the Rings) -- and yet, despite this opportunity, the eagles retain air superiority, and there are no airborne dogfights between eagles and Balrogs as there are, in various tales, between eagles and (explicitly winged) dragons, or eagles and (explicitly winged) flying Nazgûl.

Balrogs remain resolutely wingless in early drafts for The Lord of the Rings itself (see The Treason of Isengard) where the Balrog of Moria is described as 'no more than Man high' but similar in every other way to the published creation, except that wings are not mentioned, not even as phantasmagoric, metaphoric or ephemeral shrouds of shadow.

The shape and attributes of Balrogs are established almost from the first -- demons of fire, humanoid in shape and ranging in height between that of a Man, and twice that, with the characteristic fiery sword and whip. Nowhere are they explicitly described as having wings, in contrast to the clear descriptions of wings in the other flying creatures described by Tolkien. Because of this, the wings of the Balrog of Moria, if they have any substance at all, are wings of smoke and shadow, rather than flesh and bone.

Not that this purely literary argument is likely to convince diehards in the Balrogs-have-wings camp, so I shall have to resort to experiments in aerodynamics, and -- to a small extent -- pterodactylics. A traditional way of proving a contention in mathematics or logic is to admit the opposite, and follow it to its remorselessly illogical conclusion. In which case, let us suppose, for a moment, that Balrogs really did have wings. What would they have been good for? Could they have been used for flight? My answer to that one is, again, probably not -- or, at least, not for the purposes of rapid airborne pursuit, which is all that is allowed from the tiny hints that Tolkien gives us.

Compared with birds, Balrogs were big. The largest birds to fly today are the Kori bustard (Ardeotis kori) of Africa, with a mass of 16 kg; the wandering albatross (Diomedea exulans) has a mass of 10-12 kg, and the South American condor (Vultur gryphus) is 14 kg. These were matched in the very recent geological past by Haast's Eagle (Harpagornis moorei) of New Zealand (up to 13 kg), a raptor that preyed on the giant moas -- a group of flightless birds akin to the ostrich, and which, like their flying predator, are all now extinct.

The Ice Age of the Americas saw the evolution of some very large, vulture-like birds of prey akin to condors, known as teratorns. Merriam's teratorn, (Teratornis merriami) had a mass of around 13 kg and a wingspan of more than three meters, comparable with that of a modern condor. It preyed on animals trapped in treacherous asphalt seeps just off what is now Wilshire and Fairfax in downtown Los Angeles. But the biggest flying bird ever discovered lived about six million years ago in Argentina. This bird, Argentavis magnficens, was the mightiest of the teratorns, with a mass of 80 kg and a wingspan of six metres. This creature was about the right size to have carried a few hobbits in its talons, or even a wizened wizard on its back -- it makes a perfect model for the heroic eagles such as Gwaihir or Thorondor who always seem to turn up just in time to save the day. Roosting in the pinnacles of what is now the Andes, the eyries of Argentavis would have been homes fitting for the Eagles of the Lords of the West.

Although many kinds of enormous, non-flying birds are known to have lived -- one, Phorusrhacos, stood 2.5 metres tall and had a truly fearsome beak in a skull the size of that of a horse -- it is not known whether any birds existed larger than Argentavis that could have flown.

The problem of flight is not so much one of staying aloft -- it is getting up there to begin with. The crucial parameter is something called 'wing loading' -- which describes the mass of the creature as a function of its wing area. A large creature can take off, provided it has a big enough area of wing to support its weight -- and provided that it can reach a certain minimum take-off speed. This take-off speed depends on the wing loading, and it turns out that Argentavis would have had to have been moving at around 40 kilometers per hour (25 miles per hour) before it would have generated enough lift to get airborne. This is why large birds, such as swans and geese, flap vigorously and have a long run-up to get themselves into the air. However, it is doubtful whether a bird with a six-meter wingspan could have flapped its wings while standing on the ground: albatrosses, which have very long wings and relatively short legs, cannot do this very well. In any case, it is worth remembering that the only birds to achieve 40 km/h on the flat are specialized, flightless runners such as ostriches and rheas: the limbs of teratorns are stout and stocky and not built for fast sprinting.

But airspeed is relative: all that matters is an airflow of sufficient speed is moving across the wings, irrespective of whether the bird is running pell-mell towards it. All a large bird needs to do to take off is to run against a stiff breeze, or -- if the wind is strong enough -- simply face the wind and open its wings, and it can take off from a standing start. Condors take off in precisely this way. In a paper on the flight capabilities of Argentavis, Sergio F. Vizcaíno and Richard A. Fariña suggest that the bird would have roosted in the Andes and hunted on the pampas to the east. Stronger easterly breezes blew across the region then than is the case today, because the Andes were lower -- these breezes would have allowed the birds to have taken off from the Pampas with little difficulty, once they had landed to scavenge or hunt: much larger birds than could manage this feat in the gentler Argentine breezes of today, explaining why condors still exist, but large teratorns do not.

Of course, problems of aerodynamics matter less for large birds living in trees, cliffs or mountain peaks. All they need to do is fall off their perches, open their wings, and gravity will do the rest. It is not for nothing that large birds such as eagles, condors and albatrosses live in mountains or on high, rocky islands.

As birds grow larger, their wings must expand in area at a greater rate than the mass they are required to lift. This is why you see small birds -- but not large ones -- flying successfully with small, stubby wings. But large wings impose an additional penalty -- they are hard to flap up and down. Large birds flap more slowly than smaller ones, and this imposes certain restrictions on how they live. Condors and albatrosses do not flit from peak to peak entirely by their own muscular exertion. Instead, they gain lift from rising currents of air called thermals, from air moving up mountain slopes, and -- in the case of albatrosses -- from air deflected from waves. Thermal soaring -- in which large birds such as vultures cruise from thermal to thermal, sometimes for hundreds of kilometers without a single flap -- is probably how Argentavis got around.

Another consideration that affects large birds is wing shape, or what engineers call 'aspect ratio' -- the ratio of wingspan to the mean chord, or breadth of the wing. Wings with a low aspect ratio are deep and relatively stubby, whereas wings with a high aspect-ratio are long and narrow. When air moves across a wing, it tends to break up and become turbulent at the trailing edge, creating drag. This effect is minimized in narrow, high-aspect-ratio wings, because the air passes all the way across the wing before becoming turbulent. Fliers with high-aspect-ratio wings -- such as albatrosses, or sailplanes -- can fly very economically, exploiting the merest updraft of air with minimum loss of height, and can also circle very tightly at low speed. Vultures, on the other hand, have deep, low-aspect-ratio wings, but get around the problem of drag by spreading their trailing-edge flight feathers, creating trailing-edge slots. A bird with a low aspect ratio will sacrifice lift for speed -- an advantage for a soarer winging between thermals, and for a bird of prey that wishes to lose height very quickly, such as when it stoops on its prey from a great height.

We do not know if larger birds than Argentavis ever flew, because the maximum size of birds is determined by lifestyle, environment and metabolism as much as by aerodynamics. However, it is doubtful that any birds grew as large as the largest pterosaurs, such as Quetzalcoatlus, which lived in Texas in the Late Cretaceous Period, around 80 million years ago. The smallest of these creatures, distant reptilian relatives of both birds and dinosaurs, were no bigger than a sparrow. The largest, however, had wingspans of the order of ten or eleven meters -- as much as a small plane, and twice as great as Argentavis. However, their skeletons were constructed with extraordinary lightness. Birds are light for their size because many of their bones are hollow, and some are fused into rigid, airframe-like structures. Pterosaurs took this tendency to the extreme, and the result was enormous wings, very low wing loading, and, possibly, high aspect ratio -- although this last is controversial, because paleontologists disagree about the wing shape of the largest pterosaurs. Large pterosaurs would probably have been expert soarers, taking advantage of the tiniest air currents, manoeuvering on the weakest, narrowest thermals by circling within their own wingspans, and being able to take off by opening their huge, hang-glider-like wings into the mildest breezes. They would have flapped only when cruising close to the ground or the sea, when air reflected off the surface would have given them an extra boost. In this respect, the depiction of the huge, Nazgûl steeds in the Peter Jackson film adaptations of The Lord of the Rings, flapping very slowly and in general only when they are close to the ground, are very accurate indeed.

So why are there no large pterosaurs today? Or if not pterosaurs, why have no animals evolved to fill the niche of giant, living sailplanes? The answer could simply be environmental. In the Cretaceous period, the polar regions were not glaciated as they are today: warm, equatorial seawater percolated right round the globe, so that contrasts of temperature were far less than they are nowadays. This would have meant that winds were, in general, light -- perfect for a sailplane, but not for anything more robust. Things changed when Antarctica moved across the South Pole, preventing seawater access to keep it warm; and when the northern continents moved to isolate the Arctic Ocean from warmer seas. The result was an increase in the contrast in temperature between poles and tropics, which created much windier weather than the large pterosaurs -- adapted for flying slowly in mild conditions -- could have tolerated. Accustomed as they were to nothing stronger than keen, fitful gusts, the ensuing tempests would have blown the pterosaurs out of the skies. Continental drift, then, not the arrows of keen-sighted elves, would have done for the pterosaurs in the end, even if they had not perished at the end of the Cretaceous Period 65 million years ago, quite suddenly, along with the dinosaurs.

Where does this leave the Balrogs? Let's look at Balrog aerodynamics from the very few clues Tolkien left. Tolkien said that a Balrog was of approximately humanoid shape, and between one and two times the height of a man. Let us compromise and say it was about one and a half times as high, or about three meters tall. Without any other clues about the Balrog's build, we have to take Tolkien at his word and assume that if he likens it to a human in shape, it might have been like a human in build, and had no unusual bird- or pterosaur-like weight-saving features such as hollowed bones. Using a standard ideal height-weight chart as a (very) rough guide, and owning that such extrapolations run the risk of straying into complete fairyland, a humanoid this tall would have a mass of around 140 kg -- getting on for twice the mass of Argentavis.

However, as described in The Bridge of Khazad-Dûm, the Balrog had a wingspan that stretched from 'wall to wall'. This is as close as Tolkien gets to a description of Balrog wingspan -- but we have no idea of the dimensions of the hall concerned, except that it was large. We might hazard a guess that the hall was -- say -- 30 meters across, but this estimate is, frankly, as good as any other. In any case, the span was likely to have been very much larger than the Balrog's actual height, effectively dwarfing it. So, unless they looked like Dumbo's ears, these wings would have had to have been quite narrow.

Assuming, therefore, an extremely high, albatross-like aspect ratio of about 20, the wing area of a Balrog would have been 30 x 1.5 = 45 square meters, and its wing loading would have been 140/45 = 3.1 kg/m². This would make for a very light flier, with a proportionate wing loading far less than the very heavy Canada goose (loading 20 kg/m2) or even the relatively light peregrine falcon (9.1 kg/m².). Argentavis had an estimated wing loading of 11.4 kg/m²., which, as Vizcaino and Fariña note, is lower than expected for a soarer of its size, and even lower than expected for a marine soarer such as an albatross. Even compared with this, the wing loading for a Balrog seems incredibly small -- and decreasing the aspect ratio (while keeping wingspan the same) would have made the wing loading lower still.

From this it is simple to calculate the minimum airspeed that a Balrog would have needed to get airborne. This speed is given by dividing the wing loading measured not in kg/m². but in pascals (which is a measure of weight rather than mass) by a constant determined by wing shape, and taking the square root of the product. The constant is held to be 0.9kg/m³. for what Vizcaino and Fariña describe as 'well-designed' wings. If we assume that the wings of Balrogs were so designed, the minimum take-off speed is 5.84 meters per second, the equivalent of 21 km/h or about 13 miles per hour -- a moderate breeze on the Beaufort scale, enough to raise dust and move small branches. This is half the estimated take-off speed of Argentavis -- but remember that we have adopted a relatively high-aspect wing for a Balrog. If we were to reduce the aspect ratio while retaining the same 'wall-to-wall' span of 30m, we would increase the wing area still further and reduce minimum take-off speed. Adopting an eagle-like aspect-ratio of about 9, Balrog wing area becomes 99 m²., loading becomes a tiny 1.4 kg/m³., and minimum take-off speed is just 3.9 meters per second, which translates as 14 kilometers or just 8 miles per hour, a mild breeze strong enough to be felt on the face, and to rustle a few leaves.

In the light of these calculations, the wing loading of a Balrog was such that it would have been able to take off by extending its wings over the updraft caused by an energetic fire, which would have been just the thing for Thangorodrim, if not within the confines of Moria. However, aerodynamics would have constrained it to have behaved like a sailplane -- it could never have flapped these enormous, sail-like wings, which would have been precisely the wrong kind of wing for the rapid interception implied when Tolkien alludes to Balrogs in flight -- notably, the tempestuous, winged pursuit to rescue Morgoth. Say what you like about sailplanes, but I do not think that one can ever describe their motion as 'tempestuous': any tempest, or even any rapid motion, would have crumpled a flying Balrog like an umbrella in a hurricane.

But all these arguments are as nothing when you examine the great size of a Balrog's wings, at least given the meager hints offered by Tolkien -- wings this enormous would have been a Balrog's most prominent feature, so it is strange that Tolkien never mentioned them until the later drafts of The Lord of the Rings, and even then in such ambiguous terms. Or perhaps the wings were so huge that any self-respecting Balrog would have been -- well -- too embarrassed to have drawn attention to them? Yup, I think I'll have seen just about everything, when I see a Balrog fly.

There is a possible let-out clause, however -- that the wings of Balrogs were not really wings, but shadowy extensions of the body designed to make them seem even larger and more frightening -- in other words, like the cape of Count Dracula. And despite Tolkien's protestation at the existence of Disney (see Letters by J. R. R. Tolkien #13), I see a strong resemblance between Balrogs and the enormous, very scary winged creature depicted in the 'Night on the Bare Mountain' sequence in Fantasia (1940) and the 'Firebird' sequence in Fantasia 2000 -- pointing up the contrast between fire and shadow that is the essence of Balrog, as well as displaying wings which, while terrifyingly huge, were too big to have been aerodynamically feasible, at least for the purposes stated by Tolkien.

--Olog-Hai

Source: Vizcaíno, S. F. and Fariña, R. A. On the flight capabilities and distribution of the giant Miocene bird Argentavis magnificens (Teratornithidae), Lethaia 32, 271-278 (1999).