The Colossus of Rhodes

“The site of the Colossus has never been determined with any accuracy, nor has the pose been described for us by reliable eyewitnesses,” writes Lawrence Durrell. “Argument over possible sites and poses is likely to go on until the next earthquake, in which presumably the whole island of Rhodes will sink into the sea and leave behind it legends as tenuous as those which make up the myth of Atlantis.”

In the meantime, we can divide our questions about the Colossus of Rhodes into three categories. There are those — albeit far too few of them — that are answered with reasonable definiteness by the ancient texts that have come down to us, such as the dates when its construction began and ended and the impetus behind it. There are those questions we can never realistically hope to answer, such as why the Rhodians waited ten full years to begin work on a statue meant to commemorate their great victory over Demetrius, using their booty from that very same battle. And then there are those questions we cannot answer using direct historical evidence, but for which speculation, combined with our more general knowledge of ancient Greek culture, technology, and building practices, can provide some probable solutions. Having largely exhausted what we know with certainty about the Colossus of Rhodes in our first chapter, it’s to these sorts of speculations that we’ll have to turn now as we attempt to fill in the rest of the where, what, and how of it to whatever extent we can.

The first question to address, then, is just where the Colossus would have been built, given that the harbor-bestriding monster of popular legend is a no-go. Three separate sites have garnered a measure of scholarly support over the years: the tip of one of the moles in Mandraki Harbor (as opposed to both of them at the same time); the center of Rhodes Town, where a Temple of Helios is known to have stood, near where the Palace of the Grand Master — a legacy of the Knights Hospitaller — stands today; and the acropolis above the city. Robert B. Kebric, the most recent historian to write about the Colossus at any length, made what strikes me as a very persuasive argument for the last in 2019, just eighteen months before his death. I am very much in his debt for what follows.

We learned in the last chapter that the city of Rhodes was possessed of a very impressive acropolis on the top of the hill that is known today as Monte Smith. Yet I neglected to mention one of the site’s most interesting aspects: Monte Smith is one of the surprisingly few places on earth, and the only place known to the ancient Greeks, where one can watch the sun rise out of the sea in the east, wheel across the sky over the course of the day, and sink back into the sea to the west, always with a completely unobstructed view. This quality would have invested the site with sacred importance under any circumstances, but that importance must have been doubled or trebled on Rhodes, the self-styled island of the sun, which enjoyed Helios himself as its patron deity. What better place to build a monument to the god of the sun than the one spot where people could watch his daily journey unimpeded by any obstructions? (In more recent times, a similar question has been asked and answered by communications companies; the top of Monte Smith is dotted today with the television and cellular towers that link Rhodes to the rest of the world.)

There is also a more practical consideration to be added to these religious and cultural advantages. The material of Monte Smith is solid limestone, an ideal natural foundation for a large monument such as the Colossus. The stability it provided would have been especially important on an island such as Rhodes, which is rocked regularly by earthquakes, not to mention the occasional violent thunderstorm with its dozens of lightning strikes. Yes, the Colossus would eventually be felled by an unusually powerful earthquake, but the fact that it managed to withstand 56 years worth of smaller ones is itself a testament to the talent and forethought of its builders.

The other locations proposed for the Colossus of Rhodes have none of the geological advantages of Monte Smith, being made of softer, more porous earth. Preparing ground such as this to support a structure of the size and weight of the Colossus would have been a massive engineering project in itself, one quite probably well beyond the Rhodians’ capabilities; the human-made moles of Mandraki Harbor would, needless to say, be an especially hopeless proposition in this regard. The ground of the acropolis, by contrast, would have required no real preparation at all beyond a careful leveling. The ancient Rhodians were surely adept enough at structural engineering to understand the site’s advantages in this respect — and, if by chance they weren’t, they enjoyed close links with the Egypt of Ptolemy I, who may very well have sent advisors to assist with the project. The Egyptians’ ability to build on limestone to a monumental scale was as legendary in the third century BC as it is in our own; the Pyramids of Giza, those most famous of all ancient monuments, had themselves been built circa 2500 BC on and of limestone, the only foundation that could have supported their weight.

So, the acropolis had much to recommend it as a site for the Colossus. Yet it must be acknowledged that, despite much looking, archaeologists have never uncovered any indubitable traces of it there.

This lack of physical evidence for the Colossus on the acropolis is frustrating, but it is nowhere close to sufficient to prove a negative about the statue’s existence there long ago. If the fallen Colossus was cut up for its raw materials at some point after Pliny wrote about it — something historians believe did occur, for reasons which we’ll explore in the next chapter — the pedestal on which it stood would presumably have been removed as well. And because the natural limestone of Monte Smith was virtually all the foundation the Colossus required, no trace of the Wonder of the World that had once stood here might be left even a few decades after it was removed, much less centuries.

Against the unfortunate but eminently explainable lack of physical evidence of the Colossus on Monte Smith must be balanced the site’s elevation, its unique relationship to the sun, and the practical advantages of its limestone shelf. Robert B. Kebric had no doubts at all in 2019 about its most likely location: “By themselves, each of these three extraordinary factors might be offered as the major [emphasis original] reason for why the Rhodians would decide to build the Colossus of Rhodes where they did; together, they are a remarkable triad of physical realities centered only at this one place on the island.” So, we’ll continue to follow his lead in picturing the figure of Helios towering over the city and harbors of Rhodes from a perch atop the acropolis.

The what and how of the Colossus are, alas, even less easily addressed than the where of it. The one thing that everyone can agree on is that its surface was made of bronze, the perfect shiny stuff for a statue of a sun god on an island famed for its sunshine. Beyond that, the difficulties are legion, not least because Philo, the one ancient writer we have who deigns to describe how the Colossus was made, writes things that cannot all be simultaneously true. We’ll return to these problems momentarily, but first I should briefly describe how bronze-casting, the process most often assumed to have been employed in the creation of the Colossus, actually works. What follows is a description of the “direct lost-wax” process that has been in use for thousands of years. I want to emphasize that it is a grossly simplified description of what could be an extremely complex process in reality; anyone tempted to believe that ancient peoples were in any way our intellectual inferiors should study the art of bronze-casting to be quickly disabused of the notion. Still, the simplified version should suffice for our purposes.

Let’s imagine that we want to create an object in bronze of human size or smaller. First, we sculpt a full-sized version of the figure we wish to cast out of clay. We then coat our clay model with a thin layer of beeswax, wait for it to harden, and cover the entirety with more clay. After making some strategically placed holes in the bottom of the piece, we heat it in a kiln to cause the wax to melt and run out, leaving behind a hollow space — a negative image of the original model — between the two layers of clay. Into this space we pour molten bronze. After it has been allowed to cool and harden, we can chip away the outer layer of clay to reveal the bronze sculpture beneath, to which we can make any adjustments, corrections, and enhancements that prove to be necessary using a hammer and chisel, and then polish its surface to the desired patina — a process known as “chasing.”

But what if we wish to make a bronze sculpture twenty times the height of a human being? Clearly some other techniques would have to come into play, but there exists no scholarly consensus as to what they might have been. Indeed, only one scholar in the course of the past century has even tried to provide a thoroughgoing explanation of how the Colossus was made. On December 3, 1953, one Herbert Maryon, a 79-year-old British authority on sculpting and metalworking in ancient and modern times, presented before the Society of Antiquaries in London the most comprehensive answer ever to the problem of the Colossus of Rhodes. His brave proposal, which appeared in The Journal of Hellenic Studies in 1956, rested upon his own immense reserves of practical knowledge, combined with a close reading of Philo.

Maryon imagined that the artist in charge of the project — a man named Chares if Strabo and Pliny are to be believed — started by sculpting a scale model of Helios in plaster, of human size or slightly larger. He then made a “chassis,” a sort of miniature scaffolding of wood, to surround it.

This he forms of straight, squared bars, perfectly truly planed and with every angle a right angle. Considerable care is devoted to this structure to ensure that all its outer surface is true and square. The chassis is then placed round the plaster figure, special care being taken to ensure that its sides are placed exactly vertically. Then it is fastened permanently to the model. It will be remembered that in the case of the Colossus the enlarged figure would have many times the dimensions of the original model, so any mistakes in the setting of the small chassis might have serious results.

Next a vastly larger wooden scaffolding was constructed, stretching all the way up to the planned height of the finished statue. This scaffold was an exact proportional duplicate of its smaller companion; one inch (2.5 centimeters) on the small scaffold might correspond to two feet (61 centimeters) on its full-size equivalent. The model inside the small scaffold would serve to keep Chares oriented as he worked on the full-size Colossus; a system of plumb lines and set-squares attached to both scaffolds would ensure that he stayed precisely on track.

Within the large scaffold was first raised a framework of iron to serve as the skeleton of the god. To this Chares and his assistants fit stucco panels, shaping them to match the same parts of the model statue. These were then removed and taken down to a foundry, where other craftsmen beat out their duplicates in bronze plates. Finally, said plates were hauled back up to be permanently affixed to the statue’s iron skeleton. The Colossus was not cast at all, in other words, but hammered laboriously into shape, panel by panel. As it rose higher, a ramped mound of earth was also raised around it to allow the workers easy access. And so the bronzed god slowly climbed toward the heavens, at a pace of perhaps ten feet (three meters) per year. Maryon provides a vivid image of the scene in his article.

Suppose that we are standing at the top of an immense mound of earth, up which we have climbed by a spiral pathway. There is an extensive view all round: the little town of Rhodes lying at our feet with its harbours, and the rocky coastline stretching away in the distance. Across the sea, some dozen miles [19 kilometers] away, is the coast of Asia Minor. Close before us rises a rectangular wooden scaffolding, like the framework of a building. Within it we see a screen of bronze which, as a close look tells us, is shaped like part of a man. The lower part of his body and his legs, we find, may be seen in the great pit which, framed by the scaffolding, penetrates the centre of the mound. A platform spans the gap, and we can look down into a great cavern within the body with sides formed by the bronze plates. Tall columns of stone rise from the bottom of the cavern, and from them radiate numberless iron struts which support the bronze walls. Nearby on the level top of the mound the original model for the Colossus stands on a bench within its chassis.

Work is in progress. [A stucco] panel some 4 feet [1.2 meters] long [rises] a few feet higher than the finished portion of the figure. The master sculptor is at work, employing a large riffle with which he works over the surface, modifying it to his liking. When he is satisfied with the work he will give an order, and his assistants will remove the panel from its position on the figure and bear it away to where, at some distance from the foot of the mound, the group of workshops is situated. Nearby [is] a foundry in which a large bronze plate is removed from its mould. When the plate had cooled enough it would be carried into the principal workshop. The craftsmen would take [the] sheet of bronze, and with hammers beat it to shape. Direct hammering would be continued until the master craftsman decided that the work was now far enough advanced for chasing to begin.

When the chasing was completed the panel and the stucco model of which it was a copy would be stood up side by side successively in a number of different lightings, both indoor and out, and the correct modelling of the form checked before it was finally passed by the sculptor. The work of fitting the plate to its neighbors on either side and into its position on the Colossus followed. Finally, it would be riveted in position.

Within a year of the publication of Maryon’s article, an historian and linguist named D.E.L. Haynes — the same whom we met in Chapter 1, when we learned of his argument for the traditional dating of Philo’s text about the Colossus — wrote a riposte which claimed Maryon’s thesis to be invalidated in its entirety by a mistake in the translation of Philo which Maryon had employed. Where Maryon had read that each successive layer of the Colossus was “filled up,” noted Haynes correctly, Philo had in fact written that each layer was “cast on top.” “Whatever we do, let us at least try to understand what Philo actually said,” concluded Haynes a little snippily. Ever since, most historians have more or less dismissed Maryon’s lengthy labor of love as being the unfortunate product of a botched translation — garbage in, garbage out.

Yet to do so strikes me as doing Maryon a great disservice. For taking Philo completely literally, as Haynes seemingly wishes us to do, is highly problematic, as Maryon well recognizes. Translation issues aside, Maryon does consider whether each stage of the statue might have been cast in place, and concludes that to do so would be “extremely inconvenient and impractical,” entailing as it would building a whole new foundry at each level; such an installation, capable of heating metal to 2000 degrees Fahrenheit (1100 degrees Celsius) and then pouring it safely, was at the cutting edge of ancient technology, and tended to be neither particularly cheap nor particularly mobile. Further, Philo’s vague statement in the corrected translation that each layer was “cast on top” of the one before it doesn’t make much sense even if we grant him his peripatetic foundries. Exactly how were these successive layers joined together to make a complete god? In his rejection of Maryon’s thesis, Haynes states blithely that “since the molten metal which was to form the new part would presumably have come into direct contact with the existing part, fusion would probably have resulted.” But, as Maryon doubtless could have told him, bronze-casting simply doesn’t work that way; pouring molten metal on top of its solid counterpart doesn’t result in an instantaneous “fusion” of the two.

Consider as well Philo’s claim that 12.5 tons of bronze went into the Colossus. This figure looks impressive at first glance, but when one does the math, as Maryon did, one finds that it equates to a likely thickness across the statue’s enormous surface area of somewhere between .06 and .1 inches (1.5 and 2.5 millimeters) — about the thickness of the average modern coin. The typical example of cast ancient bronze, by contrast, has a thickness of about one inch (2.5 centimeters). And it is equally impossible to take literally Philo’s claim that this 12.5 tons of bronze “might have exhausted the mines.” While it was and is a considerable quantity of the metal, to be sure, such a figure represents a tiny sliver of the entire annual bronze trade of the third century BC.

And then what to make of Philo’s description of the Colossus as hollow, such that it needed to be “held steady with stones that had been put inside?” (The same claim would later be echoed by Pliny in his description of the fallen statue; he writes of “vast caverns yawning in the interior.”) If its pieces were cast using the traditional method, the clay of the model of each piece would remain inside. And if said clay was somehow removed, a cast-bronze statue of the size and thinness of this one would never be able to support its own weight; it would crumple and fall to earth like a skyscraper made out of cardboard.

All of which is to say that Philo’s text is riddled with logical inconsistencies, as even our textual literalist D.E.L. Haynes is forced to acknowledge in the end. “Are we to reject [the rest of Philo’s description of the Colossus’s construction] simply because a single figure [i.e., the total weight of the bronze employed], mentioned by Philo once and not supported by any other evidence, cannot be reconciled with it?” he asks. “Since figures are notoriously liable to corruption, it seems more reasonable to reject the figure.” And so we have it straight from the horse’s mouth: Haynes too is picking and choosing which parts of Philo’s text to accept as true. His argument with Maryon is ultimately nothing more than a difference of opinion about which parts those should be, based on reasoning that is conjectural at best.

It seems to me that this whole debate is driven by an understandable but misguided desire to make Philo’s text into something it really isn’t. Historians like Haynes and to some extent even Maryon have wished to see him as a sort of ancient investigative journalist, dutifully reporting facts picked up from credible sources. It’s much more likely, however, that Philo was combining some measure of casual Rhodian scuttlebutt with a huge measure of conjecture of his own. Although he was an engineer himself, and thus filled with an engineer’s desire to get to the bottom of how things were done in a technological sense, he was by all indications no authority on bronze-casting or sculpting; none of the other texts of his that have come down to us touch on either of those subjects. His text on the Colossus probably constitutes little more than his best guess of how it was made — a guess grounded in a limited level of knowledge about the actual processes of bronze-casting. In these respects, he was ironically little different from D.E.L. Haynes, the man who would later wish to cite him as an arbiter of historical truth.

I would therefore argue that Herbert Maryon’s thesis deserves rehabilitation. For, unlike those of Haynes or Philo, his speculations were grounded in well over half a century of personal, intimate study of ancient bronze-casting, including much practical experimentation with materials and techniques. Of all those who have studied the Colossus of Rhodes, he may have been the best equipped to separate the possible from the impossible, the reasonable from the unreasonable, in the fraught debate over the methods of its construction. Certainly none of his detractors have ever offered up what he did: a sober, realistic, complete, and believable account of how a statue as large as the Colossus could have been made during the third century BC. In the absence of alternative explanations, I’m willing to give him the benefit of the doubt and accept his thesis of a statue that consisted of an iron frame surfaced with panels of beaten rather than cast bronze, with stones and whatever else was handy thrown into the hollow space inside the structure in order to weigh it down. Such a thesis isn’t as sexy as that of a cast-bronze statue — this in itself may account for much of the resistance to it — but it fits the known facts of the case much better, without requiring any technologies which the ancient Rhodians are not known to have possessed.

All it did require was care, money, and a willingness to dedicate skilled and unskilled laborers to the project for a very long period of time. “Modern people cannot easily appreciate what mass labour can achieve,” write the historians John and Elizabeth Romer, very accurately, in their own study of the Colossus. Yet the archaeological remnants of the ancient world provide even the casual modern tourist with plenty of monumental evidence of just how much mass labor really was able to achieve using fairly rudimentary building techniques, from the Pyramids of Giza to the Colosseum in Rome. Had the Colossus of Rhodes enjoyed a different fate, it too might have joined that bucket list.

But let us move on to a less controversial subject: that of the completed statue’s height. As we saw in Chapter 1, our three principal ancient sources are in relative agreement here: Strabo and Pliny claim the Colossus was 106 feet (32 meters) tall, Philo that it was 120 feet (36.5 meters) in height. Most modern scholars are content to average these figures, arriving at a nice round number of about 110 feet (33.5 meters).

But what does such a number really represent in terms of subjective space? A frame of reference is always useful in translating a measurement into an imaginative vision. Fortunately, we have a convenient one to hand, in the form of a modern monument that was built as a conscious evocation of the Colossus of Rhodes: the 1886-vintage Statue of Liberty in New York Harbor, whose pedestal goes so far as to sport a sonnet proclaiming it to be “The New Colossus.” Ignoring its 154-foot [47-meter] pedestal and measuring only the statue itself, we arrive at a height of 151 feet (46 meters). But if we measure Lady Liberty only from her toes to the top of her head, setting aside the torch she holds aloft, we find that she is just 111 feet (34 meters) tall — i.e., roughly the same height as her inspiration. This, then, is the scale on which our imaginations need to work. I hope the comparison serves to reinforce what an amazing achievement the Colossus was in its day.

The comparison also proves useful in another way. In his remarks on the fallen Colossus, Pliny tells us that its thumb was so thick around that “few men can clasp [it] in their arms,” while “its fingers are larger than most statues.” The Statue of Liberty’s thumbs are not individually articulated due to its pose — one hand is holding a tablet, the other holding the torch aloft — but the middle joint of its index finger is 3.5 feet (1.1 meters) in circumference, which is indeed a stretch for any but the longest-armed human huggers. The same finger is just over 8 feet (2.5 meters) long — i.e., “larger than most [ancient] statues,” which were generally built to human scale or just slightly bigger. Thus we can feel fairly confident that Pliny’s descriptions match his numbers.

Like the Statue of Liberty, the Colossus of Rhodes stood on a pedestal; this Philo explicitly states. The most likely scenario here is a three-tiered base, in order to create the best visual effect for spectators at ground level. Philo claims the pedestal to have been made of white marble, but this much-sought-after material would need to be imported at enormous expense, for Rhodes had no white-marble quarries of its own. Robert B. Kebric has suggested that only the third tier of the pedestal might have used white marble, the other two Rhodes’s own less expensive gray-blue marble, and this would definitely seem a reasonable compromise. Regardless, only the visible surface of the pedestal would have been made of marble. Its internal structure would have been limestone, not just for reasons of cost but because the heavier rock would have provided a much more stable base for the statue.

And how tall was the pedestal? Again, we have only informed speculation to go on, but such is not without value. Kebric has noted that the Colossus was surely intended to be viewed to excellent effect from the sea. In order to achieve this, its builders would want to place it so that its feet stood high enough above the surrounding buildings and landscape of the acropolis that the statue could be seen from head to toe by arriving mariners. Based upon this consideration, Kebric proposes a pedestal of about 50 feet (15 meters) in height at its third and tallest tier.

At this point, then, an admittedly highly speculative but possible or even probable version of the Colossus of Rhodes is starting to emerge for us. The 110-foot (33.5 meter) statue stands on its 50-foot (15-meter) pedestal on the acropolis of Rhodes, itself 270 feet (82 meters) above sea level. Thus the top of the god’s head is fully 430 feet (131 meters) above the waves of the city’s harbors. It must have been a truly awe-inspiring sight for any visitor to Rhodes — a sight absolutely unique in the ancient world, a sight to put even the likes of our modern Statue of Liberty to shame. (After all, the Colossus didn’t have skyscrapers and container ships to compete with.) Only one question remains: what did the Colossus actually look like?

Here even informed speculation can take us very little distance at all. We have no detailed written descriptions from ancient times of the Colossus’s appearance, and no known visual representations of it either. Still, one thing at least is certain: the sculptor would have had to balance aesthetics with structural integrity. An outstretched pose like that of the Statue of Liberty would probably not have been manageable. Benefiting from thousands of years of progress in metallurgy, the latter is made of iron and pure copper rather than iron and bronze, and employs a sophisticated internal frame that allows it to move and flex with the changing winds and temperatures; the statue’s torch, for example, can sway as much as five inches (13 centimeters) in high winds. Taking into consideration the need for stability as well as the beaten rather than cast bronze I echo Herbert Maryon in proposing to have been the principal material of its surface, the real Colossus of Rhodes, if we could go back in time and visit it in its heyday, might well appear disappointing when viewed close up by we who have been exposed to so many centuries worth of fanciful depictions. “To remain upright,” suggests historian Reynold Higgins in his study of the Colossus, “[the] statue would have to be very simple, approximately columnar in shape and in attitude not unlike an archaic Greek kouros figure.” It would, in other words, look more like the rigid, somewhat awkward Greek statuary from the era of Homer than it would the dynamic, graceful pieces that were being sculpted on a regular basis by the third century BC.

But would it necessarily have to? Once again, Herbert Maryon dared to propose something quite different. A sketch which appears in his 1956 article shows a Colossus that is shading its eyes with its upraised right hand while it trails a cloak from its left arm which cleverly conceals a vital additional support column. Later studies were quick to dismiss Maryon’s suggestion because it was based on a relief, discovered at Rhodes in 1932, which was believed at the time to be exactly the ancient visual depiction of the Colossus which historians had so long dreamed of; alas, it has since been reevaluated, and is now believed to depict merely an ordinary human athlete. But, just as one mistake in translation shouldn’t invalidate Maryon’s entire thoughtful proposal of how the Colossus might have been built, surely a mistake like this one doesn’t render worthless his demonstration that many structural sleights of hand are possible when an ingenious artist sets his mind to it.

It’s possible as well that Helios may have been riding in his chariot — a very common motif for this god — or even sitting on a throne; both would have been more inherently structurally sound than a free-standing figure. But whatever the details of his appearance, and however disappointingly crude he might even have looked when viewed close-up, the god surely looked spectacular when viewed from the sea, with his bronze skin shining in the famously brilliant Rhodian sunshine.

There is just one more aspect of the Colossus as a physical object that begs to be mentioned. In 2019, Robert B. Kebric also made the intriguing suggestion that it might have served a practical as well as an aesthetic and political purpose: that it might have functioned as a lighthouse. Kebric draws parallels between Rhodes and Alexandria, the latter being the one Greek city of the third century BC that arguably outshone even the former as a hub of commerce and culture. Alexandria’s own, probably slightly later lighthouse, built on a low islet in the harbor and crowned by its own statue of a god (precisely which one is uncertain), was such a breathtaking sight that it eventually joined the Colossus as an acknowledged Wonder of the World. Did the two serve a similar practical purpose in addition to serving as mascots and advertisements for their respective cities? Did the one perchance even inspire the other?

In one respect at least, Rhodes was actually better equipped to support a monumental lighthouse than was Alexandria: unlike the desert land of Egypt, the wooded island of Rhodes has plenty of timber, meaning that supplying enough firewood to keep its beacon burning wouldn’t be a major issue. Kebric proposes a number of ways that the Colossus might have been made to serve this additional purpose even if we reluctantly accept that a gleaming torch in the god’s upraised hand, Statue of Liberty-style, was probably not a part of its design. For example, a fire tower might have been built next to the Colossus, with a system of pulleys that would allow workers to kindle a fire in a pod at ground level, then raise it up to the top. A network of mirrors on the tower might then have reflected the light of the flames onto the statue’s bronze paneling, both lighting it up for the world to enjoy at night and amplifying the light itself for the benefit of ships. Perhaps other fires burned on the pedestal down below, with mirrors of their own to reflect their light up to the statue, whence out to sea. A perpetually shining god of the sun to mark the island of the sun — if nothing else, the image is a fitting one.

Did you enjoy this chapter? If so, please think about pitching in to help me make many more like it. Pledge any amount you like on Patreon.

(A full listing of print and online sources used will follow the final article in this series.)