The Luce Canon

Comments on ‘Celebrating Homer’s landscapes – Troy and Ithaca revisited’ by Professor J V Luce (1998)

Professor J V Luce in 1998 wrote a well argued and splendidly illustrated popular book describing

Homer’s landscapes as he found them at Hisarlik and on Ithaki. He was well qualified for this task as

he had taken many touring parties to both sites.

Controversy surrounds the location of both Troy and Ithaca. Homer demonstrated his remarkable

knowledge of the geography of the Greek-speaking world in the famous Catalogue of Ships in the

Iliad Book 2. So how do we explain that he seemed to be unaware of the detailed geography and

topography of Troy and Ithaca, the two places which feature most prominently in the Iliad and

Odyssey? Apparently he did not know that Troy had no towering acropolis and no geothermal spring.

And he tells us that Ithaca lay low and to the west of a group of islands adjacent to Cephalonia. Yet

Ithaki, today’s Ithaca, is not low-lying, and lies to the east, not west, of Cephalonia.

In 2005 Robert Bittlestone published his splendid book, ‘Odysseus Unbound’ which showed, from Homer’s geography and landscape descriptions, that Ithaca was probably now found on Paliki, the western arm of Cephalonia which was once a separate island. Luce was not impressed. He defended his preference for Ithaki as ancient Ithaca in a short article on the Odysseus Unbound web site. His main point was that at least three important features of Homer’s landscape were much more impressive on Ithaki than on Paliki. Unfortunately, the argument that two or three such features seem more impressive on Ithaki carries no weight against the collective discoveries by Bittlestone and Diggle of almost all Homer’s Ithacan landscape features on Paliki. Bittlestone referred to Luce’s arguments as the Classical hypothesis. He pointed out (p193) that Ithaca on Ithaki ‘is not only a non-starter in regard to its external geography (modern Ithaki is not the furthest to the west) but that it also fails to test the internal topography…’

Problems and contradictions

Luce’s book will inevitably have impressed his many readers. Since I now claim to have found the

real Homeric Trojan landscape at Bergama, it may be helpful here to point out, and without wishing in

any way to be unkind, what I think are some of the main weaknesses concerning his identifications of

Homeric features at Hisarlik.

1. The wrong and right approach

Luce believes implicitly that Hisarlik was Homer’s Troy, and that the Mendere and Dombrek were the Skamander and Simois. These articles of faith become the ‘facts’ upon which his book is founded. He therefore has to fit Homer’s descriptions of the landscape of the Trojan War into the environs of Hisarlik. This cannot be done without saying that many of Homer’s descriptions, not convincingly found at Hisarlik, were poetic imaginings. Luce, like everyone else, does not consider the possibility that some of geographical signposts pointing to Troy at Hisarlik may have been later interpolations. A better approach would have been to use all Homer’s landscape descriptions to reconstruct a theoretical Homeric Trojan plain, and then see which river plain in the catchment of Mt. Ida offers the best fit. But if he had done this, his successful and informative book would not have been written.

2. Hisarlik was a coastal fortress

By 1982 it was generally accepted that, in the Trojan ear, Hisarlik was a coastal fortress. Had Schliemann known this, he would never have dug there. He would never have claimed that Troy was a coastal city, for Homer’s Troy was not visible from the Greek ships. There was a large plain between Homer’s Troy and the sea, upon which the Trojan War was fought. Also, in front of the ships was a ‘rising in the plain’, upon which the Trojans camped one night overlooking the ships on the beach. To overcome these objections, Korfmann followed others in claiming the Greek camp must have been at Besik Bay. Yet Luce rejects the Besik Bay option. Instead, he places the Greek camp inside the now vanished bay. But both these alternative locations, like Hisarlik itself, fail to match Homer’s descriptions of the Trojan landscape.

3. No room for the Games of Patroclos

The site Luce chooses for the Greek camp on the Bay of Hestiaia solves some problems but creates others. Inside a bay that is subject to river floods and silting up is a poor site for beaching ships that might have to escape at short notice. The Troy Bay of the Iliad was filled from one headland to the other with the ships of the Achaeans, which does not match the site suggested by Luce. Luce’s camp would require a sideways assault by the Trojans, while a frontal attack is implied by the text. At Homer’s camp there is plenty of space facing the plain for the games of Patroklos, which included a chariot race out into the plain and back again. Luce fails to mention these games, perhaps because at his site they were not possible.

4. Priam’s Kingdom lay between Lesbos and Phrygia

Homer tells us in lines 24.543-546 that Priam’s kingdom lies between Lesbos and ‘Phrygia in the uplands’. Luce is aware of these lines, but his translation of them purports to show that Homer places Phrygia and Troy to the north of Lesbos. This is necessary if Hisarlik is to be marketed as Troy. But, apart from the forced translation, Luce ignores what Homer tells us about the northwest region of the Troad. This was occupied by other tribes, principally the Dardanians, thus distinguishing it from Priam’s kingdom and Phrygia. A more natural translation of these lines places Troy between Lesbos and Phrygia, whose capital, Gordion, lies some 500km almost due east of Lesbos, in the uplands well beyond the Bakir Çayi valley and Bergama.

5. Luce states as fact that Homer’s Hellespont is not the open sea

Others, including Prof. Janko in his commentary on the Iliad Books 13-16, disagree with him.

6. Homer’s landscape descriptions are authentic

Luce claims Homer’s landscape descriptions are authentic, but that a few features are displaced to enhance the narrative. Put simply, the features he finds are the ones that Homer has described accurately. Those he fails to find are features which Homer described badly, or which he simply

invented for poetic effect. In schoolboy terms this seems equivalent to ‘Heads I win, tails you lose.’

7. The defensive wall at the ships

The defensive wall of Book 7 and Book 12, as proposed by Nestor, was, Luce tells us, imagined by Homer. So Luce finds no trace of it at his chosen site for the Greek camp. Yet why the poet would suddenly invent such a key part of the action is not explained.

8. Homer’s ‘Throsmos’, or ‘rising (swelling) in the plain’

The Trojans camped on this higher part of the plain overlooking the Greek camp before they stormed it the next day. Luce claims this was a

gently sloping hillside on the eastern side of the ridge protecting the bay. Yet Homer’s ‘throsmos’ lay between the ships and the Skamander (8.560), which does not fit the landscape at Hisarlik.

9. The Watersmeet near Troy or the Camp

Luce claims the watersmeet, where the Simois joined the Skamander, was near Troy. But Hera and Athene (5.773-5) came to the watersmeet to help the Greeks by their ships. Luce refers to lines 5.773-4, where they come to the watersmeet, but not to the next line, 5.775, where they both walk from there to the Greek camp. So from Homer we learn the watersmeet is near the camp, and from Luce we learn it is near Troy. Luce ignores two problems. Firstly, Homer’s watersmeet was near the ships. Secondly, near Hisarlik at the time of the Trojan War, because of the Bay of Hestiaia, a ‘watersmeet’ of the two rivers did not exist. Instead, both the Mendere and Dombrek discharged directly into the bay.

10. Width of Skamander

Luce tells us of the trees and vegetation along the river bank when Achilles fought the river, and that the river in the lower plain in places is 2-300 feet wide. Yet he does not mention that Homer (21.240-5) says a fully grown elm tree had fallen and spanned the river. Is this because such a tree would not normally be more than about 120-150ft high, so the Mendere is too wide for the poet’s elm tree?

11. The Wall of Herakles

Luce identifies a circular tomb mound with a flattened top as Homer’s Wall of Herakles. We are left to wonder why Homer, of all people, should regard such a feature as having the appearance of a wall.

12. Kallikolone

Luce appears content with the traditional identification of Kallikolone as a hill some 8 kilometres east of Hisarlik in the Dombrek/Simois valley. Yet the battle watched from this lookout supposedly took place on the opposite side of Hisarlik. He makes no attempt to explain why Homer should have chosen such an unnecessarily remote viewing platform for the gods supporting the Trojans. Also, the appearance of this hill, with no distinctive brow, does not warrant the name ‘beautiful hill’ given it by Homer. So neither the location nor the appearance of this hill at

Hisarlik match Homer’s descriptions.

13. Tomb of Ilus

No site for such a tomb mound, roughly half way across the plain, can be found near a ford at Hisarlik.

14. Tomb of Myrine (Batieia)

Luce finds this at Pasa tepe on the edge of a hillside overlooking the plain. He explains that a chariot could drive around its perimeter – which he says agrees with phrase ‘open ground on every side’. Also a valley lies between this tepe and Hisarlik. Hence its location fails to match Homer’s description that it was steep and lay far out in the plain with open ground on every side. Homer’s ‘open space’ was large enough to accommodate the Trojan forces who assembled there ready to do battle against the Achaeans, not jus a single chariot.

15. Tomb of Old Aesyetes

Luce follows Strabo in finding Old Aesyetes’ tomb, the lookout of Polites, on a spur ending in a steep little brow overlooking the plain. It would have made a suitable lookout point. But no sign of any ancient tumulus remains there, so Luce can only offer us a speculative location. And he does not explain what puzzled Strabo – why Polites should have chosen a lookout here instead of watching, from a similar distance, from the safety of Troy on the ridge at Hisarlik.

16. ‘Beetling’ Troy

‘Beetling’ Troy, Luce claims, is a mistranslation, perhaps because no site for a magnificent natural acropolis as described by Homer is found in the lower Mendere valley. Luce does not tell us that Zeus said of Ilios that it was ‘the place among men most dear to my heart’. He thus avoids explaining to his readers why, given that there are so many other wonderful ancient city sites, Zeus should say this about a small rather undistinguished fort on the ridge at Hisarlik.

17. The Skaean Gate

Luce offers us no convincing site for the Skaean Gate of Ilios. This, according to Homer, opened from the acropolis directly on to the plain, so must have been well down the hillside on the western side of the fort. No evidence has been found at Hisarlik that such a gate existed there in Mycenaean times.

18. The two springs

Luce finds no place adjacent to Hisarlik where two springs at different temperatures could flow out of the ground and into the Mendere. Homer’s springs were near the Dardanian Gate, and thus on the side of the fortress facing both Dardania and the springs. At Hisarlik the spring near the foot of the hillside is on the opposite side of the site to the road leading across the plateau towards Dardania. Yet Luce claims that Homer was not referring to the spring found by Korfmann at the base of the hillside, but to the two Pinarbasi springs, some 6 miles away to the SE of


19. Homer’s death of Hector

Achilles chased Hector round the walls of Troy three times, then killed him close to the Dardanian Gate and the springs. Luce’s choice of the springs at Pinarbasi seems to imply that they ran round an ellipse some six miles long, to include both Hisarlik and these springs. This would require a chase of over 36 miles for the three circuits after a hard days fighting, with both men wearing heavy armour. Surely the reason why Hector took flight around the walls is that he hoped he might get the chance to dash back in through one of the gates to safety. To suggest that he left the

potential safety of the walls for a twelve mile cross country run to Pinarbasi and back would make a mockery of Homer’s poetic end to Hector’s life. This is neither credible nor Homeric.

20. Poseidon’s 50 mile journey from Samothrace to Troy via Aigai on the Chersonese

Luce asks us to believe that Poseidon, rushing as fast as he can to help the Achaeans by their ships, walked 40 miles to a notional Aigai on the Chersonese in 4 steps, then got into his chariot and went 30 miles across the sea in a direction at right angles to Troy, so he could arrive at a cavern midway between Tenedos and Imbros. He then went on foot some 15-20 miles across the sea to reach the Greek camp. Again, this is neither credible nor Homeric.

21. Well-wooded Samothrace

Luce’s photo of Samothrace shows it to have few if any large trees on its upper slopes. Mount Fengari, rising to 1600m, is an eroded granite extrusion, and high seasonal rainfall ensures that the mountain side does not retain enough topsoil to support extensive forests. Luce chooses to ignore Homer’s epithet ‘well-wooded’ when referring to Samothrace in this photo.

To conclude, I hope that, from these few comments, you will appreciate that while Luce made a heroic attempt to find the Trojan landscape at Hisarlik, the evidence provided by Homer, surely the best authority, makes his findings less than convincing. With Ithaca now found on Paliki and Troy found at Bergama, he may now be awarded top marks for both geography and topography. The integrity of Homer’s descriptions is at long last restored. Now that really is something to celebrate.