The Antikythera Mechanism

A paper on the Mechanism appears in tomorrow’s Nature. In brief what it is these days is a rather unim­press­ive look­ing lump of heav­ily cor­roded metal. I have a photo of it some­where, but it’s a very bad blurry photo which doesn’t do justice to its lumpy unim­press­ive­ness. Fortunately Wikipedia has this much bet­ter photo.

Antikythera Mechanism
The Antikythera Mechanism in the National Museum, Athens. Photo from Wikipedia.

The reason why it’s news is that there’s been a lot of painstak­ing work to try and see bey­ond the cor­ro­sion, and its proven spec­tac­u­larly suc­cess­ful. The mech­an­ism has been examined using X-ray tomo­graphy, which is where X-rays are used to build up a cross-section of a sub­ject slice by slice without phys­ic­ally pulling the sub­ject apart. The res­ults are con­firm­ing that Greek tech­no­logy could be stag­ger­ingly soph­ist­ic­ated.

The mech­an­ism was found in the early 1900s. It was found when some sail­ors sheltered from a storm off the island of Antikythera. They went sponge diving, but one of the divers returned to the ship and told the cap­tain of the corpses he’d found on the sea bed. The other divers went to con­firm his story and found the wreck of a Roman ves­sel which sunk in the first cen­tury BC. The corpses were bronze and marble statues, some of which can be found in the National Museum of Greece today. Along with these finds were recovered some cogs con­gealed by rust into lumps.

Antikythera DiagramThe prob­lem is what would cogs be needed for? The device was throughly invest­ig­ated by Derek de Solla Price and his con­clu­sion was start­ling. The device may have looked like clock­work, but it was no clock. Instead he argued it was an ana­logue com­puter, used for cal­cu­lat­ing the pos­i­tion of the sun, moon and plan­ets against the celes­tial sphere.

The let­ter in Nature for Nov 30 2006, announces the find­ings of the X-ray team and appears to con­firm this astro­nom­ical hypo­thesis. The X-rays have enabled more of the mech­an­ism to be read, an inscrip­tion on the back door appears to include astro­nom­ical peri­ods, includ­ing the Saros cycle of 223 syn­odic months, a syn­odic month being the period from one new moon to the next. The inscrip­tion on the front door is frag­ment­ary but with lines like “brings towards the Sun up to — and con­junc­tion” it would take a power­ful ima­gin­a­tion to con­clude it wasn’t some form of cal­cu­la­tion device.

Antikythera Reconstruction?As a piece of inter­dis­cip­lin­ary work it’s all really impress­ive. The record­ing of the inscrip­tions should be more than enough to give epi­graph­ists to argue over, and a full pub­lic­a­tion of that will fol­low. There will also be a data-set online at www​.anti​kythera​-mech​an​ism​.gr, which at the time of writ­ing is inac­cess­ible, pre­sum­ably due to the quant­ity of traffic its get­ting. The con­clu­sions drawn from the inscrip­tions and ima­ging of the gears also appears to be emin­ently reas­on­able. There is a danger, when you know what the right astro­nom­ical answer is, that you inter­pret the his­tor­ical data to fit the answer. So far I haven’t seen that here. Gaps in the inscrip­tion are there, and ques­tion­able read­ings where the glyphs may say one thing or pos­sibly another are pro­fes­sion­ally acknow­ledged. The skill of the work is hard to under­es­tim­ate. It’s not simply a 2000 year old jig­saw. It’s like put­ting together a 2000 year old jig­saw where are the pieces that haven’t been chewed by the dog of time are miss­ing, and no pic­ture on the box to say how the thing should look.

It does open up some more ques­tions. One is if the Greeks were cap­able of pro­du­cing such soph­ist­ic­ated items then why wasn’t there a sci­entific revolu­tion earlier? It’s like Greek his­tory now has its own Needham ques­tion. Another reas­on­able ques­tion is why did these devices dis­ap­pear so com­pletely from the archae­olo­gical record? In his news item François Charette argues that just as Greek astro­nomy was pre­served by the Arabs, so too was Greek tech­no­logy in the form of Astrolabes.

The Astrolabe, the later des­cend­ant of the Antikythera Mechanism? Photo by list­entoreason.

This is a story where poten­tially the dis­cov­er­ies are as excit­ing as the sci­ent­ists say they are. However as Rob Rice noted:

It is neither facile nor unin­struct­ive to remark that the
Antikythera mech­an­ism dropped and sank–twice. The first time was around 76 B.C., when the intric­ate astro­nom­ical com­puter was lost with the rest of a treasure-ship’s cargo. The second time came after Derek de Solla Price ana­lyzed and pub­lished its con­struc­tion and nature dec­ades after its recov­ery. Since his Gears from the Greeks in 1975, little atten­tion has been paid to our most excit­ing relic of advanced ancient technology.

Hopefully the new evid­ence this team has uncovered will help this pro­ject avoid the same fate.

The project’s web­site
In search of lost time by Jo Marchant at Nature (free access), an excel­lent write up of why this is so inter­est­ing.
X-tek’s page on their work for the pro­ject.
The Wikipedia entry for the Antikythera Mechanism.
The press release.


When he's not tired, fixing his car or caught in train delays, Alun Salt works part-time for the Annals of Botany weblog. His PhD was in ancient science at the University of Leicester, but he doesn't know Richard III.

18 Responses

  1. Darmok says:

    Now that is remark­able. One won­ders just how many sig­ni­fic­ant accom­plish­ments by vari­ous cul­tures have been lost, or how many times know­ledge has been forgotten.

  2. Darmok says:

    I don’t know if this is typ­ical for your blog, but I just thought I’d let you know that this entry is the top post on WordPress’s home page. I grabbed a screen­shot for you in case you miss it.

  3. Candy Minx says:

    This is so excit­ing I can hardly sleep. Some of us already knew by study­ing Mythology that pre­l­it­er­ate cul­tures knew of lunar and orbital astro­nomy. This is so excit­ing to have this tool recre­ated. Love the pic­tures you have here Alun!

    This is like, as excit­ing as Christmas.


  4. Alun says:

    I was away in Cardiff yes­ter­day, attend­ing a talk on the latest find­ings at Bryn Celli Ddu, so I missed the excite­ment. I saw I was at No. 1 this morn­ing, which I think is the first time its happened. Whenever I get a large num­ber of vis­it­ors I tend to won­der what I’ve done wrong.

    There’ll be fol­low ups on this and the Bryn Celli Ddu talk some time in the future, but I don’t know when yet. It depends on whether I can get back down to Cardiff again in the near future.

    I’m out most of today too, which means I prob­ably won’t be able to point to some of the other good blog entries on the topic like this one.

  5. Sai says:

    Wow, this is a great find.

  6. You know, in the medi­eval period the Europeans had a device known as an “armil­lary sphere”. I had always wondered if the medi­eval notion of the heav­ens was inher­ited from books, or from a rep­res­ent­a­tion device. It looks like this tech­no­logy could have been passed on from the Romans.

  7. marialectrix says:

    Probably the “Greek com­puter” thing about the Antikythera device got dropped because Von Daniken’s Chariots of the Gods crowd glommed onto it so hard. I mean, if it’s a com­puter, it must be an _alien_ com­puter! And so on.

    It’s very unfor­tu­nate, because little Greek mini-steam engines and stuff like this are much more inter­est­ing than aliens.

  8. thaed says:

    This has been get­ting a lot of press. It star­ted back in September.

    I’ve often thought that this was lost by the Greeks because it wasn’t even widely known at the time and the concept wasn’t well pre­served in documents.

  9. I would like to sug­gest that an instru­ment that is, or was, used to track the pos­i­tions of the Sun, Moon and plan­ets through the zodiac, as well as the dates of eclipses, would be one that was primar­ily used by astrologers.

    It is there­fore incor­rect to refer to this dis­cov­ery primar­ily as an astro­nom­ical one.

    The same would apply if one were to find an ancient sex­tant, which would be primar­ily referred to in a mari­time or nav­ig­a­tional con­text, rather than an astro­nom­ical one.

    I feel that credit should be given to the most applic­able context.

  10. Alexandra Smith says:

    I think the reason why devices like this aren’t seen in the archae­olo­gical record is two­fold. The first is that most of them were prob­ably melted down for reuse once they broke (or became oth­er­wise dis­carded) and the other is that they simply haven’t been recog­nised for what they are. Contextless cogs will be assumed to date from peri­ods known to have cogs, and cor­roded lumps of bronze may not be invest­ig­ated as closely as they could be — after all, it was only by chance that they dis­covered the mech­an­ism, after it mys­ter­i­ously split apart some years after being recovered. Mike Edmunds (the senior researcher on the pro­ject) hopes that this work will prompt museum cur­at­ors to look at the cor­roded lumps of bronze in their col­lec­tions a little more closely, just in case they are gears.

  11. tara says:

    this make you think about how prim­it­ive they actu­ally were it is probly the best find ever (my own opin­ion but yours may be different)

  1. November 30, 2006

    […] Justin Mullins at New Scientist dis­cusses this remark­able device and what it might have been used for. Also see Alun’s fine post at his blog Archaeoastronomy dis­cuss­ing the relic, the find­ings, and the implications. […]

  2. November 30, 2006

    […] Archeoastronomy points to a stun­ning piece of research on the “the Antikythera Mechanism”, a cor­roded lump of metal­lic stuff found on a Roman wreck which sank around 76 B.C. Recovered early in the last cen­tury, it was found to con­tain gears — gears! — which cre­ated an endur­ing mys­tery which has been con­clus­ively solved a cen­tury later with X-ray tomo­graphy. I’ll let Alun Salt tell you tne answer, and point you to the rel­ev­ant note in Nature. […]

  3. December 1, 2006

    […] Alun Salt, “The Antikythera Mechanism“, Archaeoastronomy blog, Nov. 29, 2006. […]

  4. December 1, 2006

    […] This week’s issue of Nature also sev­eral off-topic, but very inter­est­ing papers: Freeth et al have recon­truc­ted the gear func­tion of the Antikythera mech­an­ism, an ancient astro­nom­ical cal­cu­lator dis­covered on a Greek island in 1901. The device dates back to the second cen­tury BCE, and is the earli­est known geared mech­an­ism. Its dis­cov­ery turns on its head the notion that tech­no­logy advances with time. The remain­ing frag­ments of the mech­an­ism are covered in astro­nom­ical inscrip­tions and con­tain more than 30 inter­lock­ing, mov­ing parts. The Antikythera mech­an­ism was used to cal­cu­late the move­ments of the sun and moon, and pos­sibly the plan­ets. It could also be used to pre­dict the dates of eclipses. Read more about the Antikythera mech­an­ism over at Archaeoastronomy. […]

  5. December 2, 2006

    […] The Antikythera Mechanism « Archaeoastronomy have always found this device fas­cin­at­ing, imply­ing as it does a pre­vi­ously unreal­ized depth to Greek (and prob­ably other culture’s) tech­no­lo­gical soph­ist­ic­a­tion — it’s like the Baghdad bat­tery (tags: Greek Antikythera Mechanism com­puter mys­tery ancient) […]

  6. January 23, 2007

    […] Salt, “The Antikythera Mechanism,” […]