When Schliemann began his excavations in 1870, he found a large mound projecting out into the plain, and it was this that he excavated. However, was there a much bigger city on the lower ground just outside Troy? Professor Manfred Korfmann of the University of Tübingen who has been re-excavating there since 1988 is sure that there has been found enough of a very much larger city to make it more than ten times as big as the traditional size. Not all are convinced, and when a major exhibition was held in Germany in 2001-2 on Troy: Dream and Reality, there was considerable dissension. The non-believers, headed by the historian Frank Kolb, also of the University of Tübingen showed up – via newspaper interviews. The controversy had the usual result: the exhibition was a huge success, attracting over 850,000 visitors. But what is the archaeological evidence – and the archaeological problems – of locating the Bronze Age city of Troy?
Troy has been dug in three major expeditions. The first and most famous was that carried out by Schliemann and his architect Dörpfeld from 1870 to 1894. Then from 1932 to 1938 an American team led by Carl W. Blegen of the University of Cincinnati added to and illuminated Schliemann’s work. Subsequently from 1988 onwards it has been excavated by a German team led by Manfred Korfmann of the University of Tübingen, joined by Professor Brian Rose who is Blegen’s successor at the University of Cincinnati; Korfmann concentrates on the Bronze Age, while Brian Rose looks at the Graeco-Roman town when Troy became a big tourist attraction for the Romans.
Troy appears today as a mound projecting out of the surrounding plain where the Trojan wars were fought. However more detailed examination shows this is somewhat misleading. Basically, Troy is a low plateau of limestone, on one side of which is a mound that forms the Troy as known to Schliemann. There are, in fact, three levels: the plain, the plateau, and the mound to one side: is the plateau the site of a ‘lower town’? Certainly it has long been recognised that the Lower Town has always been occupied by a Graeco-Roman city, for it is covered by sherds of Greek and Roman pottery. Was it also the site of a Bronze Age city?
Prehistoric sherds lying around, together with Greek and Roman pottery, indicate not only extensive settlement activities but the limits of that settlement. The recent work began with an extensive magnetometer survey of the whole area of the Lower City. This revealed in startling clarity the layout of the Greek and Roman Town with its regular alignment of insulae. However, in places there were just a few rather fainter hints of a layout on a different alignment. Was this the alignment of an earlier ‘Mycenaean’ town? The recent campaigns have been aimed at investigating these faint traces.
There are problems. These are mainly to do with the later Roman town. For one thing, the Greeks and Romans left an awful lot of rubbish, so that there is at least a one to two metres of buildup over the underlying limestone, and the extensive Graeco-Roman deposits need to be removed to reveal the underlying Mycenaean. The even bigger problem is that many of the Roman houses were built with Roman thoroughness and their foundations chopped through the underlying deposits; as a result, it is quite improbable that a complete plan of a ‘Mycenaean’ building will be discovered. Settlements were built again and again here for about 700 years, starting from the first systematic city layout some time before 200 BC, up to 500 AD. These represent about 25 generations. And there were about 700 years of decay and a new start in Byzantine times, about 1200 AD. There was an exploitation of former stonework and especially foundations. Thus whenever remains of Bronze Age houses were found, they were dug out, and the stones used again.
The evidence for the Bronze Age town comes in three main features – the houses, the defences, and the wells. The most extensive house remains come from a large area excavation of a rambling Roman house just inside the protected area: roughly half the middle city is specifically protected under Turkish law, the other half is not. Here two walls at right angles were located underlying the later walls at a different alignment, though on the same alignment as the putative remains revealed by the magnetometer survey. They were accompanied by late Bronze Age pottery and a range of radiocarbon dates of the 13th century BC. Unfortunately the later Roman foundations had cut away the rest of the house.
There was, however, one very intriguing earlier discovery. Running across the opposite corner of the area, on a different alignment, was a row of postholes, cut down into the rock, and fronted by a palisade slot. This was accompanied by a few sherds of early Bronze Age pottery, while radiocarbon dates centering on 2,600 BC were obtained; this makes it equivalent to Troy I or II, the thriving Early Bronze Age city to which Schliemann’s famous treasure should be dated. Is it possible that even at this early date, the lower city was already occupied?
The most complete houses of the late Bronze Age were discovered just outside the walls of Troy VI, where the remains had been covered by later Greek and Roman dump and thereby preserved. Behind them the Troy VI citadel wall rose up, so they were standing in their ‘shadow’, and protected by them. Furthermore the buildings that overlay them were official, long-standing buildings, like temples, the council house and a small theatre, and these were not rebuilt so frequently.
The largest area of the Mycenaean lower town was that excavated underneath these temples, revealing a central courtyard surrounded by rooms, in two of which there were the remains of a series of large pithoi or storage jars, evidence for the existence of a house of some standing. The entrance area had an impressive threshold of stone. But there were many other houses around, identified by sections of walls, floors, and storage facilities.
The second major evidence for the Late Bronze Age town lies in the discovery of a defensive u-shaped ditch. This shows up as a meandering line on the magnetometer survey, though the meanders appear to accompany the contours. Nevertheless this ditch goes up and down quite remarkably. Excavation in a number of different places have revealed this as a rock cut ditch going down well over a metre into the underlying limestone. They have even discovered the position of a gateway about 10 metres wide, where the ditch ends, and a remarkable square cut terminal, matched by a similar square cut terminal where the ditch resumes. The critics would like to see this as a water course, but if so it is difficult to understand this obvious break in the sequence and the variation in height. Up to now the existence of the surrounding ditch is proved for a length of 700 metres!
The third and most mysterious evidence for earlier activity comes from a couple of wells that are linked at the bottom by an underground water channel. This comes out at the side of the hill at a point where the water is still running and where the Romans constructed a series of fish ponds. What is the date of this feature? A crude form of dating can be obtained from the shape of the wells. Three of them are round and one is large and rectangular, and the usual rule of thumb at Troy is that wells with circular shafts are Graeco-Roman, while wells with rectangular shafts are late Bronze Age.
In order to try to date the system of shafts and tunnels, the excavators turned to one of the latest scientific techniques that enables one to date stalactites and stalagmites. There are numerous stalagtites formed on the roof of the channel and some were submitted for uranium/thorium dating, which gave a starting date of around 2,900 BC and ending in Roman times. This is much too early for the late Bronze Age city but would fit in nicely with the early Bronze Age material of Troy I and II. That means that the first builders of Troy made parts of this water system work, and that this was in use still in ‘Mycenaean’ times, and even later on, up to Roman times. The energy needed and the knowledge involved supports the fact that the people of the Troy I/II settlement considered it worthwhile to have access to water up on the ridge. This can only mean that a considerable settlement, a lower town already existed in these early days.
The Graeco-Roman town
While the search for the ‘Mycenaean’ city has been continuing on the lower ground, Brian Rose and his colleagues from the University of Cincinnati have been investigating the Graeco-Roman city of the time when Troy became a tourist attraction. The earliest known tourist at Troy was ironically the Persian King, Xerxes the Great, who in 480, before his attempted conquest of Athens, called in to pay tribute to the site of the earlier clash of Asiatics and Greeks. Subsequently a great temple to Athena was built on the top of a mound, the foundations of which almost certainly removed all traces of the ‘Mycenaean palace’. However since most of the temple was removed by Schliemann, the possible underlying palace remains hypothetical. But once Augustus and Virgil decided that the Romans were descended from Aeneas, and that Aeneas was a Trojan fleeing the sack of Troy, Troy turned into a major tourist centre: it became the very source of the Roman Empire, and as such it flourished mightily in the Roman Era.
Two particular aspects are currently being studied. One is at the North East Bastion, perhaps the greatest of all surviving bastions of ‘Mycenaean’ Troy, where a steep stone staircase running up the outside was excavated by Schliemann, who did not make much sense of it. Recently Brian Rose has been able to work out the story. The Iliad records how Ajax dragged Cassandra from the cult statue of Athena, where she had sought protection, and murdered her, thus violating the sacred area. This crime had to be atoned for. Thus the townfolk of Locris in central Greece, from which Ajax came, every year sent over a bevy of virgins to tidy up the temple. They had to do so while avoiding the gaze of the cult statue of Athena. Hence the external staircase where they could climb down to the well, fill their buckets with water and make their way round the back of the temple, and tidy everything up without being seen by Athena.
The other major recent excavations have been at the south western corner of the town, where a sanctuary probably of Cybele has been excavated. Here there were two linked sanctuaries, each containing a prominent well from which the pilgrims could purify themselves. After purification, they then left the sanctuary and climbed up to a dining room at a higher level where they could dine on the animals they had sacrificed. It was the spoil from the construction of this dining place that preserved the remains of the earlier ‘Mycenaean’ house.
The sanctuary, as indeed much of Troy, suffered a major disaster in 85 BC, when it was destroyed by the rebel Roman general Fimbria, but it was rebuilt bigger and better by Augustus in the early first century AD, and his altar base now dominates the site. He also constructed a viewing platform with tiers of seats from which the spectacles could be viewed, but this has been removed to reveal a well-preserved length of the wall of Troy VI underneath.
In conclusion, how do we assess Bronze Age Troy today? To anyone coming from the East, from the Hittite Empire or beyond, Troy would have seemed a somewhat ordinary town, medium to small by Hittite or Near Eastern standards. To someone coming from the north, however, from Europe, it would have seemed absolutely fabulous – something bigger and more sophisticated than anything to be seen in contemporary Europe. Perhaps we should see Troy as the Hittites might have seen it, as part of a tolerated group of satellites – probably speaking Luwian – along the Aegean Coast. The best known of these apart from Troy is Milawanda – the Greek Miletus, where Professor Wolf-Dietrich Niemeier, the director of the German Institute in Athens, is uncovering extensive remains of a Mycenaean Town and its earlier Minoan predecessor. Ephesus was probably Apusa, and there must have been at least one town in what they call the Saha river basin, perhaps either Pergamum or Sardis. A programme of field walking in this whole area could prove very illuminating.
But we should not be too surprised that no complete Bronze Age house has yet been discovered in the lower city. Brian Rose who is also investigating the Greco-Roman remains points out that he has not discovered a complete Hellenistic or Roman house either. They were all robbed away in the late Roman period when the builders realised that robbing older houses was easier than quarrying fresh stone. What he also points out, however, is that whenever he digs in the Roman town, he always finds Mycenaean pottery mixed in the debris, mostly cooking vessels, table wares, and water jugs – in other words, debris from households, specifically kitchens.
Troy still generates enormous excitement because of Homer. There is now a general consensus that the Iliad and the Odyssey were composed by a single person presumably called Homer in the 8th or 7th century BC, even though they were not written down until a couple of centuries later. Nevertheless they reflect a Bronze Age of four or five centuries earlier. The recent work confirms that a large town which the Hittites probably called Wilusa existed on the site that the Greeks and Romans identified as being Homeric Troy. It was probably rather larger than has hitherto been allowed.
This article is an extract from the full article published in World Archaeology Issue 1. Click here to subscribe