The mystery of the four-handed Apollo

This is something that I have been meaning to write a post about ever since I tweeted about it well over a month ago (work and a holiday got in the way). I’ve been working on an article to do with what happened to the cities of Greece in Late Antiquity and was reading the thought-provoking book by Hedrik Dey, The Afterlife of the Roman City. There I came across an intriguing reference to an image of the god Apollo with four hands, by Libanius. The fourth century AD Greek writer is describing his home city of Antioch, in what is now southern Turkey near the east coast of the Mediterranean, and to explain the arrangement of its principal roads he says:

“From four arches which are joined to each other in the form of a rectangle, four pairs of stoas proceed as from an omphalos, stetched out toward each quarter of the heavens, as in a statue of the four handed Apollo”

Dey was discussing this passage of Libanius’ fairly lengthy description of the city in order to try to reconstruct the Roman period layout of Antioch and, as such, gives the Apollo reference no thought but as soon as I read it I was immediately intrigued – and rather distracted from what I was supposed to be working on – by the puzzle of what on earth Libanius might have meant here.

The famous Apollo Belvedere in the Vatican Museum

Anybody who’s ever even picked up a book on Greek or Roman Art, or visited any museum with even a modest collection of ancient sculpture has almost certainly come face to face with a statue of Apollo, Greek god of, among other things, music, oracular predictions and healing. Representations of Apollo are some of the most common of all ancient religious sculpture, particularly for the Roman period which is when the bulk of surviving ancient sculpture dates to. Apollo is always shown as an idealized youthful man, usually naked except for a cloak about the shoulders and typically with one his two favourite attributes, the lyre or, as in the case of the famous Apollo Belvedere, his bow and arrow (well the Apollo Belvedere has lost his bow but it’s clear from his pose that he was originally holding one). Of all the statues of Apollo I have ever seen – or for that matter, for all of the statues of any Greek or Roman gods, men or women – I have never, however, seen one that had four arms.

Intrigued I did a bit of quick internet searching to see if I could find anybody commenting on what Libanius might have been talking about. I also put the question to my followers on Twitter. The combined results of my own search and my request to the Twittersphere (thanks @pythika and @lacuscurtius) were a couple of interesting references in some quite old scholarship, which I’ll come back to in a minute. Llewellyn Morgan (@llewelyn_morgan) quipped that with four arms Apollo could have held both his lyre and his weapon of choice at the same time! Whether he’d have been able to play the instrument and fire off arrows without getting his arms in a tangle or doing himself some damage is another matter….

Since then, in between continuing working on my article and other things I’ve done a little bit more digging and have given this a bit more thought but I have as many questions now as I did when I started and am posting this partly just to share my fascination for the mystery, partly in the hope that somebody out there might be able to clear some of it up for me.

Statuette of four-armed Hindu god Shiva c. 1300 AD), Walters Art Museum, Baltimore. Source Wikipedia.

My initial thought on hearing about this four-armed god was of the representations of deities with multiple limbs known from Hindu art. It’s well known that Hellenistic sculpture had made inroads into Asia by the period in question, in particular having a marked, though still not fully understood, influence on the art of Gandhara, in what is now northwest Pakistan/east Afghanistan, between the 1st to 5th centuries AD (see the excellent project at Oxford on Gandharan Art led by Peter Stewart). I know very little about Gandharan art but I happened to have recently read an article that mentioned that at least one Gandharan goddess was sometimes shown with four arms and was wondering if somehow the influence might have gone in two directions and that somewhere on the eastern fringes of the Roman Empire by Libanius’ day Greek-speaking people had started imagining their gods as looking like Vishnu or Shiva or had imported some eastern cult and called the exotic god Apollo to make him more familiar.

However, I was perhaps letting my imagination run away with me there. For one thing Peter Stewart (@PeterCNStewart) tweeted that Gandharan gods weren’t usually shown with four arms; and as I understand it the representations of Hindu gods I had in mind were all much, much later – Gandhara was a Buddhist and not a Hindu region – and much further removed geographically from the Greek and Roman world. Perhaps an even bigger problem, however, is that Libanius seems to be describing four arms set in a crossroads arrangement, rather than two arms on each sides of a figure as in the Indian images I had in mind. So what are the other possibilities?

In my online searches I found quite a few scholars citing the Libanius passage but all because they were interested in the city of Antioch and none of them seemed to find the Apollo reference interesting or odd. I did, however, find some references that seem to point in a very different direction as the possible origin of the mysterious image; not to the orient but far west of Antioch to the heartland of old Greece, to Lakonia, the region of the Peloponnese dominated in antiquity by that most anomalous of Greek city-states, Sparta.

An early reference and the first one that Google Books showed up was in an early 19th century book by Karl Ottfried Müller, The History and Antiquities of the Doric Race (the Spartan’s were Dorian Greeks), the first edition of which was published in 1824. Müller wrote of an “Apollo with four hands and four ears at Amyclae”, a small town within Spartan territory. In his footnote for his source Müller gave Hesychios (a 5th or 6th C AD compiler of a sort of dictionary of unusual Greek words) and Zenobius (a rhetorician of the time of Hadrian, early 2nd century AD), who was citing an older source, a Spartan scholar of the 3rd Century BC called Sosibius. It turns out by coincidence that Karl Ottfried Müller also wrote a pioneering monograph on the history and topography of Antioch (published in 1839) which drew heavily on Libanius’ description of the city so by that time he must have also come across Libanius’ Apollo reference. He doesn’t mention it in his earlier book and I haven’t been able to see a copy of the later volume so I’m not sure whether he ever made a connection between Libanius and these other sources.

Hesychios, Sosibius and Zenobius are all relatively obscure authors who I haven’t had much dealing with to put it mildly. It is a testimony to Müller’s scholarship, and indeed to all Classical philologists working in that age, that he was able to gather such abstruse material for his work. Nowadays an extremely comprehensive and searchable database of Greek texts is available online – the Thesaurus Linguae Graecae. There in a matter of minutes I was able to find a second Hesychios reference to the four-handed Apollo that Müller doesn’t seem to have known about. Müller presumably had to spend years pouring over manuscripts in a library.

The three passages are all extremely brief – Sosibius tells us that the Spartans had set up an Apollo with four arms and ears at Amyclae, Hesychios that this Apollo had the epithet “kouridios” (wedded) and, in the new reference I found, that there was a word “kunaktas” which was used for leather straps made of hides from bulls that had been sacrificed to the four-handed Apollo given to athletes as prizes. Presumably the athletes had competed at a contest in the god’s honour.

The dates of the sources are perhaps more interesting. We can be sure that the four-armed Apollo of Amyclae dated back at least to the 3rd century BC (time of Sosibius) and was still known about, at least to specialized scholars, in the 6th century AD (time of Hesychios) so well after Libanius’ time.

But a number of intriguing questions remain: Why did the people of Amyklae have an Apollo with four arms? What might the image have looked like? And could this be the image that inspired Libanius’ comment? To address the last question first, it is known from Libanius’ writings that he spent time in Greece and had definitely visited Sparta so he may well have seen the statue at Amyclae even though he doesn’t actually say so. Angeliki Petropoulou in a recent article about the Apollo of Amyclae takes it for granted that this was the statue Libanius meant (she also incidentally also picked up on my “new” Hesyhcios reference and added an extra piece of evidence from Sparta – a 2nd century AD inscription that mentions a four-handed Apollo).

At this point it is useful to bring in another ancient author, the considerably less obscure 2nd century travel writer Pausanias who’s featured in my work quite a bit. Pausanias came from Asia Minor. He travelled around Greece in the 2nd century AD and wrote a description of the country that he saw, focusing mainly on old monuments and sacred places. One of the places he visited was Amyclae, an insignificant village by then, and he describes seeing there a cult place of Apollo complete with statue. One 19th century author, Lloyd Watkiss, in an article I came across suggests this statue might have been the four-armed Apollo.

The Apollo at Amyklai as imagined by Quatremère de Quincy in the early 19th century based on Pausanias’ description

The statue of Apollo that Pausanias describes does indeed sound very strange. It stood on a structure that Pausanias calls a throne, which was large enough for a man to walk underneath it and decorated with what sound like a series of relief sculptures of various myths. The mythical hero Hyakinthos – lover of Apollo killed by the jealous wind god Zephyros and then turned into the flower that bears his name – was believed to be buried beneath the platform where there was a bronze door through which offerings could be made to him. The statue itself Pausanias describes as about thirty cubits, or about 14m tall. Pausanias says that the statue cannot have been by the 6th Century BC artist  Bathycles who made the throne  because of the statue’s rather odd, and rather un-Greek sounding appearance. He describes it as essentially a bronze pillar but with a head, feet and hands carrying a spear and a bow.

Tthere has been much speculation about what this statue and the “throne” might have looked like. Excavations at Amyclae have revealed something of the sanctuary but haven’t solved the mystery. There are plenty of interesting reconstruction drawings of this statue online which imagine it to have been some kind of pillar deity dating to Archaic times before the Greeks had started to imagine their gods with the idealized human bodies we are so familiar with from their art. I wonder, however, if there might be a different explanation for the strange form of the statue that Pausanias reported seeing. It was of comparable size to the famous 5th century Chryselephantine (gold and ivory) statues of Zeus at Olympia and Athena in the Parthenon at Athens by Pheidias. It is known that Pheidias’ giant statues were constructed around a wooden core and there are stories that the Athenians in later periods of crisis stripped the Athena Parthenos of its gold. Is it not possible that the statue at Amyclae too had once been adorned with gold but around a bronze instead of a wooden core and that it was the loss of this material that had given it is distinctly incongruous appearance? This would explain both the odd appearance and why the people of Amyclae thought that Bathycles had made it.

More relevant for our purposes, however, is whether Pausanias’ Apollo at Amyclae was the four-handed Apollo reported by our other sources. Most scholars have tended to disagree with Watkiss and assume that this was a different representation of the god and in fairness to Watkiss even he makes his suggestion tentatively and with reservations. Petropoulou doesn’t even consider the possibility. Lewis Richard Farnell in his 1907 Cults of the Greek City States argues that there must have been two images of Apollo at Amyclae and that Pausanias didn’t see the four handed one. The main objection to the interpretation that he did is that it would be odd for Pausanias not to have mentioned the statue had four arms if it did simply because that would have been so unusual. On the other hand Pausanias does report that the statue he saw held a bow and a spear and it is hard to see how this Apollo could have been imagined to fire his bow if he was also holding another weapon so perhaps the statue did have four arms after all. I did stumble across one online restoration of the statue as a pillar god and with four arms but now, frustratingly I’ve not been able to find it again!

Whether the different sources for these strange Apollos at Amyclae concern one or more statue they at least serve to remind us that Greek religious imagery was often far stranger than the images of athletic looking nude gods that we’re now so familiar with. The fact that this is what the vast majority of surviving ancient statues look like is arguably largely a reflection of tastes in Roman times when most of these statues – “copies of Greek originals”, as museum labels usually put it – were made.

But all this started with Libanius’ description of Antioch and it is to Libanius that we should now return because even if this was the Apollo that he was thinking of the puzzle that we are still faced with is why exactly he chose to mention it.

For a start, the street arrangement he is referring to sounds simply like a crossroads, two avenues meeting at right angles. Why would he have needed to use a reference to a statue of Apollo to make that image clear to his readers? A cross roads isn’t that unusual is it? Of course we don’t know for sure how the arms of the statue at Amyclae – if that was the statue Libanius was thinking of – were arranged so perhaps the reference did serve to make something clear about the street arrangement that is now just lost to us.

Another possibility is that Libanius, a pagan at a time when Christianity was strengthening its hold over the Roman Empire and well on its way to becoming the official religion, simply liked using a pagan image to illustrate his point. He refers to the cross-roads where the roads met as an “omphalos”, a word strongly associated with Delphi, Apollo’s most important sanctuary in Greece, where the “omphalos” stone was held to exist and mark the very centre of the world. Perhaps the reference to the statue was simply a way of building on the Apollo metaphor. Libanius actually delivered his description of Antioch at an oration at the local Olympic Games in 360 AD. (There were lots of festivals in Asia Minor in Roman times that emulated the old Olympic games in Greece). This festival was held just outside Antioch so most of his original audience would have either lived there or would have visited the city and so seen what the streets looked like for themselves. It is therefore possible that more than simply describing the city Libanius was trying to suggest its very layout was imbued with the power of the old pagan gods.

The biggest puzzle, to come back to why I got interested in this in the first place, however, is how Libanius’ audience and readers could possibly have known what he was talking about. As I said at the beginning, no image of a four-armed Apollo seems to have survived from antiquity, at least not to my knowledge. And yet, Libanius’ reference would only work as a way of explaining the street layout at Antioch if his readers had either seen the image or at least heard about. It seems unlikely that Amyclae in the fourth century AD was somewhere that large numbers of either his audience at the local Olympics (who presumably mainly came from Asia Minor or the Near East) or his readers (members of the broader Roman elite) would have visited and it seems unlikely that they would have all read the same obscure sources that the scholars Zenobius and Hesychios consulted.

So could they perhaps have known the Amyclae statue through copies?

Bronze Statuette of Apollo, Roman period. Metropolitan Museum, New York.

Many images of ancient statues did circulate in small-scale copies as marble or bronze statuettes and figurines. For what it is worth Libanius’ Greek refers to a rather than the statue of a four-handed Apollo ἐν Ἀπόλλωνος τετράχειρος ἀγάλματι.” which does seem to suggest the image was not unique. But if copies of such as a statue were widespread it is surprising that not one of them seems to have survived. It is, I suppose, just possible that such objects were common enough to have been widely known but existed in far lower numbers than the other types of figurines that have survived in their hundreds. The other possibility is that Libanius was making a deliberately obscure reference in order to show off his erudition, or to shame his, no doubt at least partially Christian, audience at their lack of knowledge of pagan cults. Neither of these explanations, however, seems completely satisfying.

As I said at the outset the passage has led me to more questions than answers: What exactly did Libanius’ four-handed Apollo look like? Was he really thinking of the statue at Amyclae mentioned by Hesychios, Sosibius and Zenobius? Was this statue at Amyclae the same as the pillar statue that Pausanias saw there? To my mind the most intriguing question of all is how widely known were images of four-handed Apollos in the Roman Empire?

From the fact that most of the scholarship I’ve cited is very old it might seem that this is an arcane sort of problem of little relevance to the current interests of ancient historians or Classical archaeologists. I think, however, that the case brings into sharp focus just how much we still don’t understand about the ancient world which, sobering though that is, is just as important as what we do know. I find the idea that the image of a four-armed Apollo was once familiar throughout the eastern Mediterranean an alluring one. That what it looked like and what it meant have both been lost to us reminds us how much there is still to discover about the Greek and Roman worlds.

And since I’m still lacking answers to much of this if anybody has any further thoughts or comments I’d be glad to hear from you.