the trojan war dated by two solar eclipses - …maajournal.com/issues/2012/vol12-1/full6.pdf ·...

ABSTRACTThe Trojan War was very significant for the ancient Greeks and they dated historical

events according to the number of years after the fall of Troy. However, there was alreadyin antiquity no consensus as to the exact date of the war when compared with differentepochs. Even after the modern discovery of the ancient city, there has been disagreementamong different excavators as to which layer corresponds to the city mentioned in theIliad attributed to Homer.

In this paper an attempt is made to identify the strange obscuration of the sun that oc-curred during the final battle of the Iliad as a total solar eclipse close to the southern borderof the zone of totality. There exists only one solar eclipse that corresponds to the descriptionin the text and this is the total solar eclipse of June 11, in 1312 BC.

When I first presented this date in 1986, there was a difference of about 60 years com-pared with the most common archaeological dating at that time. My date is now fully sup-ported by the latest results from the German-American excavation that identifies the fallof Homer’s Troy with the destruction of the archaeological layer Troy VIh, dated to about1300 BC.

Further independent support is provided by another solar eclipse that dates the reignof the Hittite king Muwatalli II. This king wrote a letter to king Alaksandu in Wilusa, iden-tified as the Hittite name for Ilios, the most frequently used name for Troy in the Iliad.Alexander was another name for Paris who abducted Helen, the crime that resulted in thewar. Muwatalli II was king 1315-1297 BC, according to the chronology for the Hittite King-dom based on a solar eclipse in 1335 BC, during the tenth year of King Mursili II (1345-1315 BC), the father of Muwatalli II.

THE TROJAN WAR DATED BY TWO SOLAR ECLIPSES

Göran Henriksson

Department of Physics and Astronomy, Uppsala University

S-751 20 Uppsala, Sweden

Corresponding author:[email protected]: 17/11/2011Accepted: 4/12/2011

KEYWORDS: Iliad, Homer, Trojan War, Total Solar Eclipse, Achilles, Patroclus, TroyVI, Wilusa, Muwatalli II and Mursili II.

Mediterranean Archaeology and Archaeometry, Vol. 12, No 1, pp. 63-76 Copyright © 2012 MAA

Printed in Greece. All rights reserved.

INTRODUCTION

New progress has been made in the re-search on Troy by the German-American ex-cavations that began in 1988. The importantnew results from the first ten years of this in-vestigation have been presented by the dif-ferent specialists in Troia, Traum undWirklichkeit written in 2001 for an exhibitionin Berlin, with the director of the excavation,Manfred Korfmann as the main editor. Afterhis investigation of the site's many buriedlayers of occupation from about 3000 BC toAD 1200, Korfmann concluded that Troy wasa prosperous and powerful centre of com-merce in the late Bronze Age, the centuriesaround the time of the legendary Trojan War.Its geographical position made it possible tocontrol shipping and collect duties on allgoods passing through the Dardanelles. Themost important new discovery is the lowercity, where the ordinary citizens lived.

The Iliad, according to ancient Greek tra-dition, was used as a history book in ancientGreece. Most modern scholars place the au-thor Homer in the 8th century BC and hisbirthplace in either Chios or Smyrna, both inIonia on the western coast of Asia Minor. Itmay be possible that he actually visited theimpressive ruins of Troy and it is certain thathe had heard of its glorious history. There aremany indications that Homer’s work is acompilation of an old oral tradition (Parry1971), which dates back to the Bronze Ageand is a mixture of fact and fantasy. The factthat the Iliad is written in hexameter is a clearindication that it had been preserved as anoral tradition.

Although the Iliad is written in Greek, it isa Greek with many anachronisms. For exam-ple, the name Achaeans was no longer usedduring the time of Homer. Some of thenames used may be of Luwian origin, but thepreserved texts in this language have unfor-tunately not been interpreted so far. Greatprogress has now been made, however, inthe identification of the place names men-

tioned in the Hittite texts written before 1200BC (Starke 2001).

THE BACKGROUND TO THIS PAPER

In November 1985, a private person con-tacted the Astronomical Observatory in Up-psala and said that he had heard a popularlecture on Troy given during the 1940s by theSwedish astronomy professor Knut Lund-mark, an early cosmologist with a greatknowledge of the history of astronomy. In hislecture Lundmark had dated the Trojan Warby a total solar eclipse that, according to hisinterpretation, had taken place during a bat-tle. He identified this eclipse as the total solareclipse that took place in the middle of theday in May, 1145 BC. The interested listenerto this lecture visited Troy during the sum-mer of 1985 and promised the guide therethat he would find out and let him know theexact date for this solar eclipse, which he hadforgotten. He called the Astronomical Obser-vatory in Uppsala and was referred to me, asI had recently finished a computer programfor the calculation of ancient solar eclipses.

At first I was sceptical because Lundmarkused Canon der Finsternisse (Oppolzer 1887),which is no longer considered reliable. Onthe other hand, I was interested in this op-portunity to test my new computer programand asked him to tell me how the solareclipse was described in the text. However,he never called me again and after a week Idecided to read the Iliad myself and look fora description of an unusual darkness in con-nection with the sun and the moon.

After three days of exciting reading Ifound what I was looking for in book 17,lines 366-377:

“So they fought like blazing fire, nor wouldyou say that sun or moon still remained intact,for with darkness were shrouded in the fight allthe chief men who stood around the slain son ofMenoetius. But the rest of the Trojans and thewell-greaved Achaeans fought unimpededunder clear air, and over them was spreadthe piercing brightness of the sun, and on allthe earth and the mountains was no cloud

64 GÖRAN HENRIKSSON

seen; (my italics) and they fought restingthemselves at times, avoiding one another’sshafts laden with groans, and standing farapart. But those in the middle suffered woesbecause of the darkness and the war, andwere worn down with the pitiless bronze, allthey who were chief men.” (Wyatt 1999).

My identification of the darkness men-tioned there with the total solar eclipse onJune 11, 1312 BC, was presented to the post-graduate seminar at the Department of Clas-sical Archaeology and Ancient History atUppsala University in the spring of 1986.This description of a remarkable darknessduring daytime, without any obscuringclouds seen over the mountains or on all theearth, and in which the sun and the moon nolonger were shining normally, can in myopinion be explained most simply and natu-rally as a non-technical description of a totalsolar eclipse in which the moon completelycovers the solar disc. Moreover, the informa-tion that the piercing brightness of the sunwas spread over the Achaeans who fought inthe outer parts of the battlefield can easily beexplained if the outer parts of the battlefieldlay just outside the zone of totality.

As chariots were used in the battle, thecentral parts of the battlefield must havebeen situated in the dry area to the south orsouth-east of the lower city of Troy, and theouter parts of the battle must have takenplace further to the south.

The slain son of Menoetius was Patroclus,the best friend of Achilles, the greatest war-rior among the Achaeans. To frighten theTrojans, Patroclus used Achilles’ armour, and

Hector, the greatest Trojan hero, thought thathe had defeated and killed Achilles himself.

I have discussed the original Greek textwith Dr Johan Flemberg, Department ofClassical Archaeology and Ancient History,Uppsala University. In his opinion the men-tion of the sun and the moon and the strongemphasis on the contrast between the dark-ness in the middle and the brightness outsideit makes it conceivable that this passage con-tains the memory of a solar eclipse. The wordfor darkness in the passage quoted above isaer (mist, haze), which is not unique, whereasthe episode itself is. According to Edward,commenting on lines 370-377 (1991: 98):“This amplification of the mist theme isunique; nowhere else is the darkness local-ized like this, and much of the language is in-novative and vivid”. Edward has foundseveral unique features in these lines, andthis strengthens the opinion that it is a gen-uine description of an unusual phenomenonand not just a poetic expression.

We are therefore looking for a total solareclipse during the Bronze Age with thesouthern limit of the total zone situated somekilometres to the south of Troy.

TOTAL SOLAR ECLIPSES IN TROY

In Table 1 all solar eclipses, between 1000-2000 BC, with a magnitude in Troy of morethan 0.990 of the sun’s diameter, can befound. The calculations were performed in1986 with the original formulas by CarlSchoch (1931). (All dates in the Gregoriancalendar).

THE TROJAN WAR DATED BY TWO SOLAR ECLIPSES 65

Table 1All solar eclipses in Troy, with magnitude greater than 0.990, between 1000-2000 BC. The calculatedtime has an uncertainty of about one minute, but the seconds are presented to make it possible for otherauthors to compare this calculation with results from different computer programs.

Year Date Local mean Magnitude of the eclipse Comments

BC solar time in Troy 4' (7.4 km)

in Troy south of Troy

1014 14/11 17t 13m 01s 0.998 0.999 Partial1312 11/6 12t 35m 24s 1.002 1.000 Total, southern limit1341 26/12 10t 23m 50s 1.009 1.010 Total1406 30/6 14t 43m 11s 0.994 0.996 Partial

The coordinates for Troy used in this cal-culations have been measured from a map:

longitude = 26° 14' 21" E and latitude = 39°56' 54" N.

Only two of the four solar eclipses weretotal in Troy. The only interesting eclipse hereis the one in 1312 BC, June 11, at 12.35.24local mean solar time in Troy, whose south-ern limit of the zone of totality is situatedonly 4' (7.4 km) to the south of Troy (Fig. 1).The fight over Patroclus' dead body musthave taken place some kilometres south orsouth-east of Troy where it was possible touse chariots as described in the Iliad.

A new calibrationmade by the authorin 2011 of the lunarsecular accelerationin longitude, basedon 33 total solareclipses back to 3653BC and the use of co-ordinates for Troyfrom Google Earth,longitude = 26° 14'20" E and latitude =39° 57' 27" N, shiftedthe limit of the totalzone just about 400 mfurther south, andthe time of maximumeclipse became12.36.08 local meansolar time in Troy.

The most difficultproblem in the calcu-lations of ancientsolar eclipses is thecorrection for thebraking of the rota-tion of the earth. Ascan be seen in Fig. 3the rotation of theearth has been delayed by about 10 hourssince the fourteenth century BC. If this effectis not taken into account the calculated zoneof totality will be shifted by 150o westwardsand there seems to have been no solar eclipse

at all for instance in Troy. To get useful re-sults the clock error in the calculationsshould be just a few minutes.

The German astronomer Carl Schochsolved this problem in 1926 and his theorywas used in my computer program in com-bination with modern astronomical param-eters until July of 2011 when I made the newcalibration. Unfortunately Schoch died in1929, but his important work was publishedafter his death by P. V. Neugebauer (Schoch1931). The apparent rotation of the earth isthe fundamental clock in these formulas, andtherefore corrections for the braking of its ro-

tation must be in-cluded in theformulas for the mo-tion of the sun, themoon and the planets.These corrections areproportional to themean motion of thesecelestial bodies andare easy to apply.

I started to developmy computer pro-gram in 1977, but itdid not work suffi-ciently well until1985, after many im-provements in the cal-culation methods andthe introduction ofabout 450 periodicterms in the perturba-tion formulas for theorbit of the moon(Hansen 1857).

After June 1985,the agreements be-tween my computa-tions and the ancientsources have been ex-

cellent. These sources include five solareclipses observed by Copernicus, notationsin medieval monastery chronicles, observa-tions by ancient Greek astronomers such asThales and Hipparchos, Late Babylonian


Fig. 1. The map shows the reconstructed shore-lines during Troy VI-VII, ca 1300 BC, drawn

after Korfmann (2001).The southern limit of the zone of totality forthe solar eclipse on June 11, 1312 BC, at 12.36local mean solar time in Troy. The upper line,7.4 km south of Troy, corresponds to the posi-tion of this limit according to the original for-mulas by Schoch (1931) and the lower line, 7.8km south of Troy, is the position according to

the new calibration by Henriksson in 2011.

cunei-form texts, Hittite annals by King Mur-sili II and interpretations of depictions oftotal solar eclipses on Swedish Rock-carvingsfrom the Bronze Age (Henriksson 1999,2005).

The most important and conclusive test ofmy computer program was the identificationof two total solar eclipses in Babylon in 1859BC and 1558 BC that fit very well with thedetails in the written sources and also withthe interval between the eclipses of 301/300years (Henriksson 2006). Six solar eclipseshave been used to determine an absolutechronology for the pharaohs from the 18th,19th and 20th dynasties in Egypt, and twosolar eclipses date the exodus from Egypt bythe People of Israel (Henriksson 2007). TheXia, Shang and Western Zhou dynasties havebeen dated by identification of importantChinese total solar eclipses preserved in laterhistorical chronicles (Henriksson 2008).

The sky on Achilles’ new shield

The famous “Shield of Achilles” is theshield that Achilles used in his fight withHector described in a passage in Book 18,lines 478-608 of the Iliad. The description ofthe shield falls between the fight over Patro-clus’ body and Achilles’ re-entry into battle.Achilles had lost his armour after lending itto his friend and companion Patroclus whowas killed by Hector and his weapons takenas spoils. Thetis, Achilles' mother, asked thegod Hephaestos to make new armor for herson. Homer gives a detailed description ofthe imagery which decorates the new shield,which may be considered as a physical en-capsulation of the entire world. It starts fromthe shield's centre and moves outwards incircles. The central motif is the earth, sky andsea, the sun, the moon and the constellations,ILIAD, BOOK 18: 484-489:

“On it he fashioned the earth, on it theheavens, on it the sea, and the unweariedsun, and the moon at the full, and on it theconstellations with which heaven is crowned– the Pleiades and the Hyades and mighty

Orion and the Bear, that men call also theWain, that circles ever in its place, andwatches Orion, and alone has no part in thebaths of Ocean.” (Wyatt 1999). (The Wain isalso called the Big Dipper.)

It may not be just a coincidence that all thecelestial objects mentioned in these lineswere visible in the sky during the total phaseof the solar eclipse on June 11, in 1312 BC, at12h 36m, during the fight over Patroclus’body. This part of the sky is the winter sky,not visible during the summer except duringthe total phase of a solar eclipse (Fig. 2).

The memory of this unique situation mayhave been considered as a proof of the inter-ference from the gods in the battle and mayhave been the inspiration to the central motifof the new shield that Hephaistos made forAchilles.

This independent evidence supports thehypothesis that the total solar eclipse wasthe explanation for the darkness described sodramatically in book 17.


Fig. 2. The sky above Troy during the total solareclipse on June 11, in 1312 BC, at 12h 36m localmean solar time, during the fight over the body

of Patroclus. This date corresponds toSkirophorion 1, in 1312 BC, according to the

much later Attic calendar.

CALCULATIONS OF ANCIENT SOLARECLIPSES

Unfortunately all modern computer pro-grams for the calculation of ancient solareclipses are based on the so-calledEphemeris Time (ET) (Stephenson and Mor-rison 1984). It is defined by the mean motionof the planets, but a conversion table to cal-

culate the Universal Time (UT) necessary inthe calculation of ancient solar eclipses mustbe established. This method has not beensuccessful for calculation of early solareclipses. John Steele (2000) admits that it iscompletely useless for epochs before 700 BC.However, there exist problems also withsome critical solar eclipses before 500 AD(Morrison and Stephenson 2002) (Fig. 3).


Fig. 3. The parabolic time-shift, ΔT=ET-UT, due to the deceleration of the rotation of the Earth as afunction of time with the coefficient 36.28 from Schoch(1931) and 31.0 from Stephenson and Morri-son (1984). The unit for the coefficients is seconds/(century)2 and the time is reckoned from 1800.0.

The symbols correspond to identified solar eclipses, mostly total or annular. Most of the identifiedeclipses can be found in Henriksson (1999, 2005, 2006, 2007, 2008, 2009 and 2010).

Today the results of the determinations ofthe lunar secular acceleration from ancientsolar eclipses can be tested by comparisonwith the Lunar Laser Range (LLR)-measure-ments. Before the observed lunar siderealsecular acceleration can be used in the calcu-lations of the LLR-measurements, by theThird Law of Kepler, it must be corrected forthe precession of the geodesic predicted inthe General Theory of Relativity by AlbertEinstein (1916). Such a test was in practice

not possible until during the 1970s when as-tronauts from the Apollo project placed laserreflectors on the surface of the Moon. Mod-ern calculations give the value for the preces-sion of the geodesic as -2 x 1.92"/cy2 =-3.84"/cy2 (Nordtvedt 1996).

The lunar secular acceleration -25.85"/cy2

determined from the LLR-measurements byWilliams et al. (2008) is only valid in theearth-moon inertial system. The followers ofStephenson, for example John Steele (2000),

have taken the value -25.85"/cy2 from the de-terminations by LLR as a confirmation oftheir own value, -26 ± 2"/cy2, determined byMorrison and Ward (1975). However, thisdetermination was considered by Williamset al. (2008) as “not accurate to the numbergiven” and useless because the value fromLLR must be corrected for Einstein’s preces-sion of the geodesic to get the sidereal lunarsecular acceleration used in the calculationsof the solar and lunar eclipses.

This means that Stephenson’s method tocalculate eclipses is based on a too approxi-mate value for the lunar secular accelerationand is without physical relevance. The greatdeviations from the observations are ex-plained as non-tidal effects of unknown ori-gin. A detailed discussion of the problemswith Stephenson’s method to calculateeclipses can be found in Henriksson (2009and 2010).

The real sidereal lunar secular accelerationdetermined by the LLR-measurements, in-cluding the Einstein effect, is: -25.85"/cy2 -3.84"/cy2 = -29.69"/cy2. This value is very closeto the value -29.68"/cy2 determined bySchoch (1931) from ancient solar eclipses.This seems to be a very good confirmation ofEinstein’s prediction. However, Schoch’s the-ory has an undefined third-order term in hisformula for the lunar longitude. It has prob-ably been used as a final correction to his the-ory to get a closer fit to the observed ancientsolar eclipses. This means that the lunar sec-

ular acceleration is time-dependent inSchoch’s theory, and the constant -29.68"/cy2

is only valid around the year 1800 AD. Themain purpose for Schoch was to get the bestset of formulas for calculation of ancientsolar eclipses. He was probably less inter-ested in the real acceleration of the Moon.

In July 2011, the present author found aconstant value for the sidereal secular accel-eration of the Moon without the third-orderterm. The new constant value corresponds tothe true secular acceleration of the Mooncaused by the tidal acceleration by the earth.

DIFFERENT ATTEMPTS TO DATE THETROJAN WAR

There were two generations between thefall of Troy and the Dorian invasion, accord-ing to Eratosthenes. His date, 1184 BC, isbased on the traditional year 1104 BC for theDorian invasion and two generations setequal to 80 years. The date for the Dorian in-vasion, however, is a reconstruction and can-not be used as a fixed point for an absolutechronology. If we instead accept my date1312 BC as the year of the fall of HomericTroy, then the Dorian invasion should havetaken place about 1232 BC, which fits verywell with the introduction of the ceramicstyle LHIIICb, or about 1230 BC according tothe classification by the archaeologist ArneFurumark (1941) of Uppsala University, seeTable 2.


Table 2The Trojan war dated by ancient sources and modern investigations

Year Author/reference Ceramic CommentBC period

1135 Ephoros from Cyme*1184 Eratosthenes* 1104+2x401209 Parian Marble# Chronicle from Paros 1250 Blegen (1964) VIIa Furumark IIIBb1300 Dörpfeld (1902) VI Furumark IIIBb - IIIAs1300 Starke (2001) VIh Furumark IIIBb - IIIAs1312 Henriksson (1986) Total solar eclipse1334 Douris from Samos*

* Blegen (1964), # Jacoby (1929)

The Parian Marble is preserved as twofragments, one of which gives the traditionaldates for the kings of Athens from the leg-endary Kekrops, beginning as far back as1581/80 BC, down to events of 355/4 BC, andthe other fragment covers the period 336/5 -299/8 BC. The date 1209/8 BC is determinedas “24 From when Troy was taken, 945 years,in the <2>2nd year that [Menesthe]us wasking of Athens, on the 7th day before the endof the month Th[argelio]n”, translated byGillian Newing (2001).

When Heinrich Schliemann started his in-vestigation of Troy in 1871, he was lookingmainly for the treasure of King Priam. In1878 he made a deep trench through themound and at the bottom of this he found acollection of valuable vessels and jewellerymade of gold and other objects made ofbronze that he identified as the treasure ofPriam (Schliemann 1880, 1884). This level isnow called Troy II and dated ca 2500 BC,which means that it belongs to an epoch 1200years before the Homeric Troy.

After the more careful excavation lead byWilhelm Dörpfeld (1902), nine differentmain layers could be identified. Dörpfeld ar-gued that Troy VI, now dated ca 1300 BC,corresponds to the Homeric Troy. Schlie-mann realized in 1890, the last year he lived,that he had been mistaken and admitted thatDörpfeld was right.

The University of Cincinnati sponsoredthe next large investigations, those of 1932-38, led by Carl Blegen. His conclusion wasthat the city walls of Troy VI had been dam-aged by an earthquake and not by theAchaeans and he identified the HomericTroy as level VIIa, with a date of ca 1250 BC.

Fritz Schachermeyr (1983) considers the“Trojan war” to be a combination of twowars: the fall of Troy VI, 1300-1280 BC, andthe fall of Troy VIIa, 1220-1200. He calls thepottery of type LHIIIC in Troy VIIb1 Barbar-ian Ware, and the pottery of Troy VIIb2 iscalled Buckel-pottery.

The author of this paper dated the fall ofthe Homeric Troy to 1312 BC through iden-

tification of the total solar eclipse mentionedin the Iliad, and this was presented at the De-partment of Classical Archaeology and An-cient History, Uppsala University, in 1986.However no attempt was made to publishthe result, as there existed no independentsupport at that time for such an early date ascompared with the archaeological date of ca1250 BC made by Blegen (1964).

The situation is now completely differentafter the investigation begun in 1988 by ateam of German and American archaeolo-gists lead by Manfred Korfmann. The resultsfrom this investigation identify the HomericTroy once more as Troy VIh, with a date ofca 1300 BC (Starke 2001). In Korfmann (2004)there is a table called “Chronological Table,Troia-project, state April 2004” which repeatsthe earlier conclusion that the end of TroyVIh may have been an earthquake and a firecatastrophe that took place ca 1300 BC. Theend of Troy VIi=(VIIa) was another fire catas-trophe perhaps in a lost war 1190/80.

ARCHAEOLOGICAL EVIDENCE USEDTO IDENTIFY THE HOMERIC TROY

The impressive city walls of Troy VI wereconstructed over a period of several hundredyears and a development towards strongerwalls can be seen. A wider and higher wallsuccessively replaced the older and weakerparts. This work was almost finished, exceptfor about 35 m of the southwestern wall,which has half the thickness of the adjacentparts to the south and north. They do not joinbecause the newer thicker parts of the wallwere built outside the thinner, older wall.The thinner wall was built in the middle ofthe period Troy VI (about 1550 BC) and itwas probably intended to be replaced by anew stronger wall like the one that encirclednearly all the city (Fig. 4). The work to re-build the weakest part was never completedand the gaps between the adjoining parts ofthe new wall were only provisionally filledup with large stones.

A possible explanation for this interrup-


tion in the construction of the new wall is asudden unexpected attack on Troy duringthe last part of period VI called Troy VIh(LHIIIBb- IIIAs). There are also traces of de-struction by fire of some houses in this sub-stratum but not of a catastrophic fire asexpected in a city taken by assault.

Blegen (1964) discovered that the upperparts of the city walls of Troy VI had fallendown on the buildings closest to the insideof the southern part of the wall. His conclu-sion was that this destruction could not havebeen caused by attacking enemy troops, butwas the result of an earthquake that tookplace about 1300 BC, and the pottery wasLHIIIBb.

After the destruction of Troy VIh by theearthquake, the destroyed houses and partsof the city wall were repaired during TroyVIIa. During this period the same pottery(LHIIIBb - IIIAs) was used and the same peo-ple continued to live in the city. Dörpfeldsuggested that Troy VIIa should be calledTroy VIi, but Blegen wanted to preserve thedesignation VIIa, as it was widely used bymany archaeologists.

The end of Troy VIIa, about 1250 BC, wasconnected with a great fire; several unburiedhuman bodies were found in the streets. Thejaw of one corpse had been cut by a sword.The destruction of Troy VIIa was identifiedby Blegen as the result of the assault by theAchaeans as told in the Odyssey, written in a

more fanciful style than in the Iliad.During the following period, Troy VIIb, a

completely different kind of less sophisti-cated handmade pottery was found, similarto that found in excavations in modern Bul-garia and very unlike the pottery from theearlier periods, which was made on the pot-ter’s wheel.

Schachermeyr (1950) has related the con-struction of the Trojan horse to the earth-quake that destroyed Troy VIh and helpedthe Achaeans to take Troy. They believed thatPoseidon, the Earth shaker, had given themtheir victory and offered him the woodenhorse because his symbol was a horse.

ABSOLUTE CHRONOLOGY FOR THE HITTITE KINGS BASED ON A SOLAR ECLIPSE

When Edwin Forrer (1926) investigatedthe annals of the Hittite king Mursili II, hefound that a very sinister solar omen wasmentioned in the beginning of his 10th yearof rule. It happened right at the outset of amilitary expedition against the country ofHayasha in north-western Armenia. Forrerrecognized this as a solar eclipse and con-tacted Carl Schoch, who the same year haddeveloped new improved formulas for thecalculation of ancient solar eclipses. Schochidentified this solar omen as the powerfulpartial solar eclipse on February 28 (March13 Julian calendar) in 1335 BC, with magni-tude 0.89 at the capital city Hattusa.

The absolute years of rule for the Hittitekings (Astour 1989), calibrated by the solareclipse in the 10th year of Mursili II, correlatesperfectly with the Egyptian chronology viathe battle at Kadesh 1299 BC, in the fifth yearof Ramses II (1304-1238 BC) and the lastyears of rule of the Hittite King Muwatalli II(1315-1297 BC). The first year of Ramses II,1304 BC, is independently determined by thedate for the new moon during his coronationfestival (Parker 1957).

The historical résumé in the next para-graph is mainly based on a paper by Frank


Fig. 4. Part of the mighty eastern wall of Troy VI(Photo G. Henriksson 2006)

Starke (2001) with Hittite dates corrected byadding 25 years due to the identification ofthe tenth year of Mursili II as 1335 BC by thesolar eclipse. A discussion on the correlationbetween the Hittite chronology by Schoch(1931), my calibration of the Old Babylonianchronology from identification of two totalsolar eclipses in Babylon and my calibrationof the years of rule for Egyptian pharaohs ofthe 18th, 19th and 20th dynasties from identifi-cation of eight solar eclipses, can be found inHenriksson (2006 and 2007).

Troy/Ilios identified in Hittite

cuneiform texts

In the western part of Asia Minor, to theeast of the Dardanelles, the Luwian lan-guage, related to the Indo-European Hittitelanguage, was spoken at the end of theBronze Age. The control of the narrow straitof the Dardanelles had long been of great im-portance, and already around 3000 BC a for-tified city was built on a hilltop on itssouth-eastern side, which later became fa-mous as Troy or Ilios in the Iliad.

Homer used the place name Troy 50 timesand Ilios 106 times in the Iliad as names of thecity and the country. Starke (2001) has iden-tified this pair of names with the Luwianplace names Truwisa and Wilusa respectively.In fact, these identifications and those dis-cussed below were already suggested by thebrilliant Swiss linguistic Edwin Forrer dur-ing the 1920s, but they were not accepted bythe other authorities of that time (Forrer1931).

Archaeologists identify nine main levelsof the city, built one upon the other with atotal of 46 substrata. The main candidate forthe Homeric city is Troy VIh or Troy VIIa.However, it is almost impossible to identifydetails mentioned in an ancient poetic de-scription of historic events with actual ar-chaeological finds from nine superimposedcity layers, partly destroyed when consecu-tive buildings were constructed upon earlierones and later partly removed by early exca-

vations made by non-specialists, mainlylooking for treasures. Even today the site isonly partly excavated. One exception is themassive city walls of hard limestone fromTroy VI, which are the most dominant con-struction among the ruins.

It is therefore of the greatest importancethat some historical information about an-cient Troy can be found in recent studies ofthe Hittite cuneiform texts. Many placenames mentioned in the Hittite annals cannow be identified with later Greek placenames, and the story told by Homer can to agreater extent be related to historical eventsbefore and after the time of the Iliad.

The Hittite Kingdom, called Hattusa afterits capital, was very well organized, and thedifferent kings preserved important eventsduring their years of rule in archives in theform of annals. Important treaties withfriendly and hostile neighbouring countries,diplomatic correspondence and even familyproblems in the royal families are preservedin these annals.

Wilusĳa is included among the states be-longing to the Land of Arzawa, in the earliestform written as Arzawĳa. Its correspondingLuwian name was Wilusa and it was men-tioned together with the country ofTaruwisa/Truwisa, identified with the Greekname for Troy. Today the local Turkish nameof Troy is Truva.

Frank Starke, according to Joachim Latacz(2001), convincingly solved the problem withthe identification of the name Wilusa in 1996after years of work by him and other schol-ars.

Rudolf Wachter (2001) describes the devel-opment of the earliest Greek alphabet. In thatalphabet there existed a letter with the soundof W, the digamma, but this letter did not existamong the East Ionian letters used byHomer. Thus the Greeks of Homer’s timewere not able to represent the sound w witha letter; the Luwian Wilusa became Ilios inGreek.

The Hittite kings led many military cam-paigns against small neighbouring countries,


mainly along its eastern border. The mostpowerful of them was the northernMesopotamian State of Mitanni, but theirmain enemy were the Egyptians in northernSyria to the south.

The first time Wilusa is mentioned in theHittite annals was in connection with a mili-tary campaign by King Tudhalĳa I (1445-1425 BC) against his western neighbours theArzawa states, which included Wilusa; thedate corresponds to Troy VIg. The capital ofArzawa was Abasa (Greek Ephesos). Thenext king of Arzawa, Tarhuntaradu, wasvery powerful and he wrote two letters inHittite to the Egyptian king Amenhotep III(1416-1379 BC). These letters have beenfound in Egypt among the Amarna letters.At that time stratum VIh had begun to ap-pear in Troy, and Tudhalĳa II (1400-1380 BC)was the Hittite king. The next Hittite kingwas the powerful Suppiluliuma I (1380-1345BC), and in the beginning of his rule,Kukkunni became king of Wilusa. He is thefirst king of Wilusa mentioned in the Hittitesources.

The land of Millawanda (Hittite name)was a southern neighbour to Arzawa alongthe western coast of Asia Minor. It has nowbeen identified as the Greek place nameMiletos, which was a Minoan colony duringthe 16th century BC and became a Mycenaeancolony ca 1400 BC. The archaeological findsbelong to Late Helladic (LH) IIIA2. The Hit-tite sources tell us that it was dominated bythe land Ahhĳawa, whose main area was sit-uated on the islands in the Aegean Sea, prob-ably including Crete. Ahhĳawa appears forthe first time in the Hittite sources ca 1425BC, LH IIIA1, with the oldest form of thename Ahhĳa, close to the original Luwianname that can be identified as the Achaeansmentioned in the Iliad under the commandof King Agamemnon from Mycenae.

Millawanda was destroyed by the Hittiteking Mursili II in his third year, 1342 BC. Thisdestruction is clearly discernable as a firehorizon, and above it the archaeologicalfinds are classified as LH IIIB1. After the de-

struction of Millawanda, Mursili II con-quered Arzawa, and Abasa fell without afight. King Uhhazidi and the royal familyfled to the territory of Ahhĳawa “over the seato the islands”, according to the annals byMursili II. After that, Arzawa lost its sover-eignty and the federation of states was dis-solved. The former members of thefederation were forced to enter a new feder-ation with Hattusa ca 1340 BC. The western-most among the former Arzawa States,Wilusa, had friendly relations with Hattusaand was not forced to be a member of thefederation.

If the final battle of Troy took place duringthe total solar eclipse in 1312 BC and the Tro-jan War really lasted for ten years, the warstarted around 1322 BC, at the end of thereign of the Hittite King Mursili II. As Wilusawas not a member of the federation, therewas no formal reason for Mursili II to helpthe city against the attacking army from Ah-hĳawa.

On the other hand it is more reasonable tosuppose that the ten years duration of thewar is just “a nice number” used to enhancethe magnitude of the struggle and the suffer-ing of the heroes. The Iliad relates only whathappened during the last 51 days of thefighting, whereas the war may have lastedjust one summer or at most a few years. Aftertheir bad experience in the war against thepowerful King Mursili II, the leaders of Ah-hĳawa may have waited until the king diedin 1315 BC before they started an attack onthe small and independent but rich countryof Wilusa.

It is known from Hittite texts that in 1315BC, Prince Pĳamaradu, a grandson of the de-feated King Uhhazidi of Arzawa, was sup-ported by the king of Ahhĳawa in militaryoperations from his islands to gain influencein Mira, one of the former Arzawa states.After that, when Pĳamaradu attacked thecoast of Seha, the neighbouring country toWilusa in the south, the king in Wilusa, Alak-sandu, was ordered by the Hittite king tosend military help. However, Pĳamaradu,


supported by the king of Ahhĳawa, was vic-torious and conquered the island Lazba(Greek Lesbos) belonging to Seha andcaused great military losses for Seha andWilusa. King Manabatarhunta from Seha fi-nally asked for military help from the Hittitearmy to restore order in Wilusa after the tres-pass of Pĳamaradu. But King Muwatalli IIdid not send military help until later, whenthe situation in Wilusa threatened the stabil-ity of the whole Hittite federation of states.

After the intervention in Wilusa by theHittite army to stabilize the rule of the coun-try, Wilusa lost its sovereignty and becamean integrated part of the Hittite Kingdom.

THE TROJAN WAR DATED BY A HITTITE LETTER

According to the Iliad, King Priam of Troyhad two sons, Hector and Paris (Alexander).Priam and Hector were killed at the end ofthe war, but Paris survived and became kingin Troy after the war.

There exist six copies in the archives inHattusa of a cuneiform letter from the HittiteKing Muwatalli II (1315-1297 BC) to KingAlaksandu in Wilusa. It contains an agree-ment that regulated the status of the Wilu-sian kingdom after the intervention by theHittite troops. King Muwatalli II dictated theagreement and it limits the sovereignty of thecountry because it was not capable of de-fending its borders.

Alaksandu is Luwian for the Greek nameAlexander; an early Mycenaean feminineform Alexandra is recorded.

The Iliad describes only 51 days during thelast year of the war, and one of these dayswas June 11, in 1312 BC, if my identificationof a total solar eclipse in book 17 is correct.This date is in perfect agreement with theonly possible interval, 1315-1297 BC, for theletter from King Muwatalli II to King Alak-sandu.

The fight over Patroclus’ body took placeon June 11, in 1312 BC, which corresponds toSkirophorion 1, the first day of the last monththat year in the Attic calendar, used by theGreek authors. The Trojan War ended accord-

ing to Eratosthenes on Skirophorion 23, butEphoros and the Parian Marble give the samedate one month earlier, on Thargelion 23.There is an uncertainty of one month at theend of the year because of intercalation.

The story of the attack on Troy/Ilios by theAchaeans in the Iliad may be the poetic ac-count of a successful attack by the king ofAhhĳawa and his allied forces on Wilusa toget compensation for their loss of influencein Milliwanda and Arzawa.

CONCLUSIONS

If the passage quoted above from book 17of the Iliad describes the total solar eclipse inTroy in 1312 BC, the battle at Troy may beconnected to historical events in the Hittitearchives. These, in turn, can be related to anabsolute chronology by the solar eclipse in1335 BC, in the tenth year of Mursili II.Therefore we know that there was an attackat Wilusa/Ilios/Troy by Ahhĳawa/Achaeansshortly after the death of Mursili II in 1315BC. In the Hittite archives a prince Pĳama-radu from Arzawa was the commander ofthe forces from Ahhĳawa that took the islandLazba, identified as Lesbos, and in the Iliadthe Achaean hero Achilles and his men con-quered Lesbos before they attacked Troy,both taking place at the same time.

ACKNOWLEDGEMENTS

I want to express my gratitude to Dr JohanFlemberg, Department of Classical Archae-ology and Ancient History at Uppsala Uni-versity, for his help in 1986 in finding EdwinForrer’s handwritten German translationfrom Hittite of the annals by Mursili II in thegreat collection of books in Carolina Rediviva,the University Library in Uppsala. He alsohelped me with the reading of the originalGreek text in book 17. Finally, I want to thankmy research college associate professor MaryBlomberg, also of the Department of Classi-cal Archaeology and Ancient History in Up-psala, for useful discussions and thecorrection of my English.


REFERENCES

Astour M. C. (1989), Hittite History and Absolute Chronology of the Bronze Age, §10-12. PaulÅströms förlag. Partille.

Blegen, C. W. (1964), Troy and the Trojans. New York.Dörpfeld, W. (1902), Troja und Ilion: Ergebnisse der Ausgrabungen in den vorhistorischen und his-

torischen Schichten von Ilion, 1870-94. Athen.Edward, M. W. (1991), The Iliad: a commentary, vol. 5: books 17-20. Cambridge. Einstein, A. (1916), Die Grundlage der allgemeinen Relativitätstheorie, Annalen der Physik

49.Forrer, E. (1926/29), Forschungen. Vols. I.1 and II.1, 1-3. Berlin.Forrer, E. (1931), Apollon, Vulcanus, und die Kyklopen in den Boghazkoï-Texten, Revue hittite

et asianique 5, pp.141-144.Furumark, A. (1941), The Mycenaean Pottery. Analysis and Classification. Stockholm. Hansen, P. A. (1857), Tables de la lune. Londres.Henriksson, G. (1999), Prehistoric constellations on Swedish Rock-carvings, In Actes de l Vème

conférence de la SEAC, Gdańsk, 5-8 septembre 1997 (Światowit supplement series H:Anthropology, 2), ed. A. Le Beuf and M. Ziólkowski, 155-173. Warsaw.

Henriksson, G. (2005), Solar eclipses, supernova and Encke's comet on Swedish rock Carv-ings, in Proceedings of the Fifth Oxford International Conference on Archaeo- astronomy,Santa Fe, August 1996, ed. Fountain, J. W. and Sinclair, R. M., Carolina AcademicPress, Durham, North Carolina, pp. 475-485.

Henriksson, G. (2006), A new Chronology of the Old Babylonian Kingdom and Ur I-II basedon identification of solar and lunar eclipses, in Proceedings of the SEAC 2002 Confer-ence in Tartu.

Henriksson, G. (2007), Chronology for the Egyptian Pharaohs of the Amarna period and theIsraeli leaders Moses and Joshua by correlation with eight solar eclipses, in Pro-ceedings of the SEAC 2004 Conference in Kecskemet. BAR International Series 1647,133-148.

Henriksson, G. (2008), A new attempt to date the Xia, Shang and Zhou dynasties by solareclipses, Proceedings of the Oxford 8 and SEAC 2007 Conference in Klaipeda, Astronomyand Cosmology in Folk Traditions and Cultural Heritage, Archaeologica Baltica 10, 105-109. Klaipeda.

Henriksson, G. (2009), A new test of Einstein’s theory of relativity by ancient solar eclipsesCosmology across Cultures, Astronomical Society of the Pacific Conference Seriesvol. 409, 166–171. Ed. J. A. Rubiño-Martín, J. A. Belmonte, F. Prada and A. Alberdi.

Henriksson, G. (2010), Einstein’s Theory of Relativity confirmed by Ancient Solar Eclipses,Journal of Cosmology, 2010, Vol. 9, 2259-2270.

Jacoby, F. (1929), Die Fragmente der griechischen Historiker 2B. Korfmann, M. (2001), A guide to TROIA, Istanbul.Korfmann, M. (2004), Die Arbeiten in Troia/Wilusa 2003 - Work at Troia/Wilusa, Studia Troica

14, 3–31.Latacz, J. (2001), Homers Troia/Ilios: Erfindung oder bewahrte Erinnerung? Ein überblick,

in Troia, Traum und Wirklichkeit, Konrad Theiss Verlag GmbH. Stuttgart.Morrison, L. V. and Ward, C. G. (1975), An analysis of the transits of Mercury 1677-1973.

Mon. Not. R. Astr. Soc. 173, 183-206. Morrison, L. V. and Stephenson, F. R. (2002), Ancient Eclipses and the Earth's Rotation, in

Highlights of Astronomy, vol. 12, IAU, editor Rickman, H.


Newing, G. (2001), The Parian Marble: Translation, University of Oxford, Ashmolean Museum.Nordtvedt, K. (1996), From Newton’s Moon to Einstein’s Moon, Physics Today, May 1996,

Vol. 49:5, p. 29. Oppolzer, T. R. von (1887), Canon der Finsternisse.Wien. Parker, R.A., (1957) The lunar dates of Tuthmosis III and Ramesses II, Journal of Near Eastern

Studies 16, 39-43.Parry, Milman. (1971), The making of Homeric verse; the collected papers of Milman Parry, ed.

Adam Parry, Clarendon Press. Oxford.Schachermeyr, F. (1950), Poseidon und die Entstehung des griechischen Götterglaubens. Bern.Schachermeyr, F. (1983), Die griechische Rückerinnerung im Lichte neuer Forschungen. Wien.Schliemann, H. (1880), Ilios the City and the Country of the Trojans. London.Schliemann, H. (1884), Troja: Results of the Latest Researches and Discoveries on the Site of Homer’s

Troy. London.Schoch, C. (1931), Die säkulare Accelaration des Mondes und der Sonne, Astronomische Ab-

handlungen, Ergänzungshefte zu den Astronomischen Nachrichten, Band 8, B2. Kiel. Starke, F. (2001), Troia im Machtfüge des zweiten Jahrtausends vor Christus, Die Geschichte

des Landes Wilusa, in Troia, Traum und Wirklichkeit, Konrad Theiss Verlag GmbH.Stuttgart.

Steele, J. M. (2000), Observations and Predictions of Eclipse Times by Early AstronomersArchimedes. Vol 4, pp. 37-41. Dordrecht (2000).

Stephenson, F. R. (1982), Historical Eclipses, Scientific American, October, p 154-164.Stephenson, F. R. and Morrison, L. V. (1984), Long-term changes in the rotation of the Earth:

700 B.C. to A.D. 1980. Philosophical Transactions of the Royal Society. Vol. A 313, pp.47-70.

Wachter, R. (2001), Die Troia-Geschichte wird schriftlich, Homers Ilias wird zum Buch, inTroia, Traum und Wirklichkeit, Konrad Theiss Verlag GmbH. Stuttgart.

Williams, J. G., Boggs, D. H. and Folkner, W. M. (2008), DE421 Lunar orbit, Physical Libra-tions, and Surface Coordinates, IOM 335-JW,DB,WF-20080314-001, March 14, 2008.

Wyatt, W. F. (1999), Iliad, by Homer, revised translation, LOEB Classical Library, LCL 171,Harvard University Press, Cambridge, Massachusetts, London, England.


the trojan war dated by two solar eclipses - …maajournal.com/issues/2012/vol12-1/full6.pdf ·...

Documents