Geoscientist
The magazine of The Geological Society of London
Furthest towards dusk
- the quest for Ithaca
Volume 16 • No 9 • September 2006
www.geolsoc.org.uk
Quest for Ithaca
This article by Professor John Underhill of Edinburgh University has been reproduced
from Geoscientist magazine Vol. 16 No. 9 (September 2006) with the kind
permission of the editor, Dr. Ted Nield. Geoscientist is the monthly colour magazine
of The Geological Society of London.
The photograph depicts John Underhill (left) and Ted Nield (right)
discussing the geological setting of Strabo’s Channel on location in
Thinia, Kefalonia. It was taken on August 23 2006 when Ted Nield took
advantage of his holiday on nearby Zacynthos to visit and assess the
principal geological localities for himself. The image was photographed
from a vantage point at Petrikata on the eastern side of the Thinia
Valley overlooking the proposed site of the buried marine channel. The
rocks in the middle and far distance are the easterly-dipping
Cretaceous and Paleogene limestones that form the western slopes of
the Thinia Valley. Photograph © Robert Bittlestone.
The book describing this project is called Odysseus Unbound: The Search for
Homer’s Ithaca. For further information about the book, including a personal view
from Ted Nield, see the final page of this extract or visit the project website at
http://www.odysseus-unbound.org
contents
Geoscientist
The magazine of The Geological Society of London
The
Geological
Society
Front cover:
Westward view over the Ionian Sea from the
Peloponnese, Greece. Where, over the winedark and broad-backed, lay Odysseus’ rocky
homeland? Feature p4. Photo © Ted Nield
Geoscientist is published by
The Geological Society Publishing House,
Unit 7, Brassmill Enterprise Centre,
Brassmill Lane, Bath BA1 3JN
Tel: 01225 445046 Fax: 01225 442836
Email: [email protected]
Editorial Team
Editor-in-Chief: Professor Tony Harris
Editor: Dr Ted Nield
Features Editor: Dr Robin Cocks
Editorial Adviser: Dr Joe McCall
Contributing Editor: Dr Sue Bowler
Production Editor: Carol Liddle
Reviews Editor: Dr Martin Degg
Council Representative: Dr Nick Rogers
Fellowship Queries
The Geological Society,
Burlington House, Piccadilly,
London W1J 0BG
Tel:+44 (0)20 7434 9944
Fax: +44 (0)20 7439 8975
Email: [email protected]
Society Web Site: www.geolsoc.org.uk
Library
Tel: +44 (0)20 7432 0999
Fax: +44 (0)20 7439 3470
Email: [email protected].
Advertising
D P Media, Suite 201, QC30,
30 Queen Charlotte Street,
Bristol BS1 4HJ
Tel 0117 904 1283,
Fax 0117 904 0085
Email: [email protected]
F E AT U R E S
4
Quest for Ithaca
by John Underhill
18
A school trip to Iceland
By Nancy Belford and Lisa Bennett
REGULARS
6
People
8
Geonews
11
Society at large
22
Opinion
26
Calendar
31
Crossword
Geoscientists in the news and on the
move
News items relevant to the interests of
working geologists
What your Society is doing at home
and abroad, in London and the
regions
Reviews
A forward plan of Society
activities
Win a special publication of
your choice
Printed by
City Print (Milton Keynes) Ltd,
17 Denbigh Hall, Bletchley,
Milton Keynes MK3 7QT
Tel: 01908 377085
Fax 01908 649335
Email: [email protected]
©The Geological Society of London is a Registered
Charity, number 210161
ISSN 0961-5628
Volume 16 • No. 9 • 3
feature
Background picture: Bay at Ormos Mirtou, NW Kefalonia is a direct consequence
of its geological setting, in soft Miocene marl within the core of a major
SW-verging asymmetric syncline in the immediate footwall to an important
thrust. This structure was termed the Kalon Thrust by Underhill
(1989) and is one of a number of such faults that transect
the Hellenide (pre-Apulian) foreland in the island.
Fig. 1: Troy, Mycenae and the Ionian islands. Boxed area - region shown Fig. 2.
Note bathymetric profile of Hellenic Trench subduction zone, curving south of Crete
and up the western coast of Greece. KTF = Kefalonia Transform Fault. Source:
NASA World Wind.
Quest for
Where was Odysseus’ homeland? John Underhill* explores the geological,
geomorphological and geophysical evidence for relocating Homer’s Ithaca.
Homer’s Iliad and Odyssey are two of the world’s oldest texts. The Iliad describes the events of
the Trojan War, (12th Century BCE, Mycenaean era), while the Odyssey tells the story of the
subsequent return of Odysseus to his palace on an island called “Ithaca”. Both poems are
thought to have been handed down via oral tradition over several centuries before being
written down during the Archaic era (8th-6th Centuries BCE).
For many years scholars believed that these poems described fictional locations. However, in
the second half of the 19th Century, German businessman and archaeologist Heinrich
Schliemann (1822–1890) proposed that the Troy described in the Iliad had been a real place.
He identified its location in NW Turkey, and excavations subsequently revealed a major Bronze
Age site (Fig. 1).
© John Underhill
In the Iliad, the Achaeans set up camp near where they had beached their ships at the mouth
of the river Scamander (modern Karamenderes). Homer described Troy as standing on a hill
across the river plain where the battles of the Trojan War took place. The site of the
ancient city today is some 15km from the coast, but the ancient mouth of the
Scamander 3000 years ago was about 5km further inland, pouring into a bay
that has since filled with alluvial sediment.
In February 2003 geologist John Kraft, working with a team
including classicist John Luce, published the results of nearly
a quarter of a century’s investigations into the geology of
the Troad region. They compared the present landscape and coastal features with those described in
the Iliad and other classical sources, notably
Strabo’s Geography1. They concluded that there
was strong geographical consistency between
Homer’s description of the city of Troy and
the available geological evidence obtained
through drilling and radiometric dating of
the sedimentary bay-fill2. Schliemann also
successfully investigated Agamemnon’s
palace at Mycenae and ‘mighty walled Tiryns’
nearby. However, although he visited the
*Prof. John R Underhill: Grant Institute of Earth Science, School of
Geosciences, The University of Edinburgh (e: [email protected])
Fig. 2: Ionian Islands off western Greece. Contrary to Homer’s ‘furthest to sea
towards dusk’, today’s Ithaki is shielded from the open sea by Kefalonia and lies to
its east, towards dawn. Homer also refers to Zacynthos, Same (Kefalonia) and a
hitherto unknown island called Doulichion. Source: NASA World Wind.
Fig. 3: Digital Elevation Model of Kefalonia and Ithaki. Note contrast between low elevation
of Paliki peninsula and higher, rugged terrain of the modern Ithaki. Source: DEM data
obtained from the NASA Shuttle Radar Topography Mission (SRTM-90) and processed
by Global Mapper.
Ithaca
island of Ithaki on the west coast of Greece in 1868, he found
no convincing evidence to support the claim that it was the
homeland of Odysseus.
The geographical description of Ithaca in the Odyssey is very
puzzling. When Odysseus makes himself known to King
Alcinoös on the island of Scherie (thought to be Corfu) he
introduces his homeland with lines that scholars have pondered
over many centuries:
I am Odysseus, Laertes’ son, world-famed
For stratagems: my name has reached the heavens.
Bright Ithaca is my home: it has a mountain,
Leaf-quivering Neriton, far visible.
Around are many islands, close to each other,
Doulichion and Same and wooded Zacynthos.
Ithaca itself lies low, furthest to sea
Towards dusk; the rest, apart, face dawn and sun.
Odyssey 9, 19-26 (trans. James Diggle)
The natural interpretation of ‘towards dusk’ is west-facing, and
dawn clearly implies east. So Odysseus’ Ithaca was a low-lying
island, furthest out to sea west of Greece, with three other
islands nearby: Doulichion, Same and Zacynthos.
However, a glance at the map (Fig. 2) makes it clear that the
island of Ithaki is not west-facing, nor is it farthest out to sea;
while a digital elevation model and perspective satellite views
(Figs. 3, 4) confirm that it is mountainous rather than low-lying.
Furthermore, although Zacynthos exists today, and almost all
experts regard Homer’s Same as Kefalonia (there is a town with
ancient ruins called Sami), the island of ‘Doulichion’ has
remained a mystery for 3000 years.
One solution to this obvious contradiction is that perhaps Homer
simply didn’t know his east from his west, his dusk from his
Fig. 4: Perspective views, Ithaki and Paliki. In Odyssey, ‘Ithaca itself lies low’, but the island
of Ithaki is mountainous along most of its terrain, with sheer cliffs. In contrast, most of
Paliki, the western peninsula of Kefalonia, is low-lying. Strabo wrote ‘Where the island is
narrowest it forms an isthmus so low-lying that it is often submerged from sea to sea. Both
Paleis and Cranioi are on the gulf near the narrows.’ Source: NASA World Wind.
dawn nor the difference between low-lying and mountainous
islands. Homer may simply have been mistaken because perhaps
he lived in Asia Minor (W. Turkey), composed the poems
several hundred years after the events of the Odyssey, had
heard vague stories of a distant Ithaki, and made some errors
accordingly. Nevertheless, a geographically incompetent Homer
left many classicists and some archaeologists feeling uneasy.
Most notable among them were Wilhelm Dörpfeld (18531940), who spent the latter stages of his career searching for
Odysseus’ palace further north on Lefkas, and A E H
Goekoop (1859-1914) who in 19083 advocated a site in southeast Kefalonia. In addition, Kefalonians Volterras (1903)4 and
Tsimaratos (1954)5 proposed alternative locations in western
Kefalonia as the site of Odysseus’ palace, albeit without any
supporting geoscientific evidence. However, the notion that
Homer simply got it wrong now seems questionable in view
of the emerging evidence at Troy, where the poet’s description
was precise.
Three years ago, management consultant Robert Bittlestone
asked: “What if Homer’s description of Ithaca was also
precise, but subsequent geological changes have altered the
layout of these islands?”He sought geological expertise in the
investigation of this question and contacted me because of
my continuing work on the geology of the Ionian Islands
since completing a PhD in the 1980s which focused on
this region.
Quest for Ithaca continues on page 14
Volume 16 • No. 9 • 5
Quest for Ithaca continued from page 5
Fig. 7: View to the south across the isthmus of Thinia which runs roughly NE-SW for c. 6km and is
1500m wide in the north to about 250m in south. Elevation of valley floor ranges from sea level
to a maximum of 180m. On its west Agrilias reaches an elevation of 456m, while the summit of
Mount Imerovigli is at 993m. Source: NASA World Wind.
Knowing the islands well, having read Homer at school and with a
keen interest in Greek history, I was intrigued. I set about testing
whether there could be a plausible geological explanation that would
not only satisfy Homer’s description of Ithaca, but would also enable
the lost island “Doulichion” to be identified. I was not the only
expert involved: Bittlestone also contacted James Diggle, Professor
of Greek and Latin at Cambridge University, who re-translated
the original texts to ensure that we had a firm foundation for the
geographical references that might provide vital new clues.
Fig. 6: Geological Map of Kefalonia - main geological structures and stratigraphy highlighting
main thrusts Ionian, Kalon, Aenos, Argostoli (AT), White Rocks (WRT) and Atheras (AF).
Boxed area shows the approximate extent of the geological map of the Thinia area, Fig. 10.
Source: Updated after Underhill (1989).
The main clue for an alternative location for ancient Ithaca came
from the work of the geographer Strabo (c. 63 BCE – 24 CE), who
also wrestled with the problem of these islands. In his Geography
he makes an unusual and very specific observation of Kefalonia:
‘Where the island is narrowest it forms an isthmus so low-lying
that it is often submerged from sea to sea’. Moreover he states that
these narrows occur on the gulf between the ancient cities of
Paleis and Cranioi, the locations of both of which are known on
Kolpos Livadhiou – the gulf separating the main part of Kefalonia
from the Paliki peninsula on its west (Fig. 5).
So in the summer of 2003 modern geoscience entered the picture.
Could a marine channel, subsequently described by Strabo as a
low-lying isthmus, once have separated the westernmost peninsula
of Kefalonia from the rest of the island during the late Bronze
Age? Because if it did, Paliki would then have been an island that
‘lies low, furthest to sea and towards dusk’.
The story of our initial investigations was published in October
2005 in the book Odysseus Unbound: The Search for Homer’s Ithaca 6.
In the remainder of this article I will summarise the geological
evidence presented there and update our research
since the book went to press.
Fig. 5 (background image): View south over Kolpos Livadhiou towards Argostoli (left of centre).
Strabo’s reference to the established sites of Paleis and Cranioi on this marine bay leads to the
conclusion that if it existed, the sea-covered isthmus he describes must have been located at the
head of the gulf in Thinia. © Robert Bittlestone.
14
Geoscientist
Geological setting
Kefalonia lies at the external (foreland)
edge of the Hellenide fold-and-thrust
system created in response to Cenozoic
continental collision following closure of
the Tethyan Ocean. The island is dissected
by a number of major thrust faults (Fig. 6),
the most notable of which are (W-E): the
Ionian, Kalon, Aenos, Argostoli, White
Rocks, Agio Sotira and Atheras thrusts7.
Although it can be shown that the locus of
deformation migrated westwards during
the Neogene and Quaternary, it is evident
that movement has persisted on some of
the faults to the present day by virtue of
the island’s highly seismogenic position
in the NW corner of the Hellenic Trench
subduction zone where it meets the NNE
trending Kefalonia Transform Fault.
The key area for investigating the hypothesis
is Thinia, immediately west of the Aenos
Thrust in NW Kefalonia. The area forms a
distinctive NE-SW trending steep-sided
valley c. 6km long, between Kolpos Agias
Kiriakis and Kolpos Livadhiou in NW
Kefalonia (Figs. 7, 8). Because the valley
floor today rises to c. 180m, it is clearly
demanding to suggest that it might have
been at sea level as recently as the Bronze
Age (late Holocene). As a result, I anticipated
that Bittlestone’s hypothesis would be easy
to test - and disprove. However, rebuttal
has not proved at all straightforward.
None of the results of geological and
geomorphological fieldwork performed so
far rules out the hypothesis that a marine
connection as described by Strabo could
have existed at that time.
Before taking a closer look at the geology
and the tests carried out to date, it is
important to stress that while compressiondriven deformation is evident in many
field localities (Figs. 9a-c) and from field
observations in the immediate aftermath
of the 1953 earthquake8,9, evidence for
Holocene uplift appears limited within
a
b
Fig. 8: (a) Panorama of the Thinia isthmus looking eastwards from Agrilias towards Kefalonia. To the left (north) is
Kolpos Agia Kiriaki; the mountain opposite is Imerovigli, 993m; to the right (south) is Kolpos Livadhiou and at its
north-eastern shore, the harbour of Agio Sotira. The location of the viewpoint is indicated in Fig. 10. (b) View north
towards Ormos Agia Kiriaki from Petrikata illustrating valley profile. Source: © Robert Bittlestone and John Underhill.
Kefalonia in general and at Thinia in particular. There is a distinctive wave-cut
platform at c.6m in the northeast of the area (Fig. 9e), which is interpreted as a
raised beach; there are distinctive wave-cut notches in a rock at Poros in the
southeast (Fig. 9d); there is a vegetated terrace and bay-fill characterising Agio
Sotira; and a raised harbour at the northern end of Kolpos Livadhiou (Fig. 9f).
Consequently it is clear that Holocene uplift amounts to only a few metres or
tens of metres at the most - insufficient to account for the present elevation of
the Thinia valley.
Another explanation must therefore be sought if we are to interpret Strabo’s
marine channel as historically accurate. That realisation has led to a field
programme consisting of geological mapping, geomorphological investigations
and latterly, of non-invasive geophysical experiments, conducted under work
permits issued by the Athens-based Institute of Geology and Mineral
Exploration (IGME).
Recent field studies in the area show that Thinia may be separated into two
distinct parts: a western area in which a largely stratigraphically conformable,
(but folded and locally thrusted) succession of Cretaceous and Paleogene limestones10 overlain by Miocene marly and clastic sediments11, dips gently east-
Volume 16 • No. 9 • 15
d
a
a
b
e
c
f
Fig. 9: Field evidence for contractional deformation in NW Kefalonia. (a) Westward verging asymmetric folds with an axial planar stylolitic cleavage in Paleogene limestone in the
immediate footwall to the Kalon Thrust, Agia Efimia; (b) Backthrust that emplaced Paleogene limestones onto a raised beach, itself c.20m in elevation. (c) Detail of backthrusts
affecting the angular unconformity between westerly dipping Paleogene limestones (below) and horizontal raised beach sediments (above); (d) Offshore islet at Poros marked by
wave-cut notches indicative of successive and abrupt relative sea-level falls interpreted as co-seismic uplift events. (e) Angular unconformity separating easterly dipping Miocene
conglomerates (below) from horizontally-bedded raised beach conglomerates (above). The latter form a distinctive wave-cut platform also at c. 6m. (f) A well-defined late Holocene
estuary bay-fill developed at the northern end of Kolpos Livadhiou. A similar but much smaller flat-lying vegetated area also characterises the southern Thinia coastline at Agio
Sotira. © John Underhill.
ward; and an eastern area (Fig. 10) in which Cretaceous
and Paleogene limestones dip steeply westward7.
Although its trace is largely obscured by rockfall debris,
where observable the boundary between the two dip
provinces is marked by a reverse (thrust) fault emplacing
the eastern, westerly-dipping Cretaceous-Paleogene limestones onto easterly-dipping Miocene sediments. The steep
(45->60°) westerly dips that characterise the eastern side
of the valley result from their forming the western limb of
a major hangingwall anticline lying above the Aenos Thrust,
which formed as a natural consequence of its emplacement.
Many bedding planes show evidence (e.g. slickensides) for
surficial, down-dip (flexural) slip having occurred along them.
The general easterly dip seen in the footwall to the Aenos
Thrust is complicated by the presence of at least one
subsidiary thrust fault, referred to here as the Agio Sotira
Thrust, which repeats Paleogene limestones and Miocene
sediments in the southern part of the valley. Folding and
16
Geoscientist
thrusting also occurs further to the west. A major thrust
fault (the Atheras Thrust) runs across the northeastern
side of the Paliki peninsula through the village of
that name and is marked by a spring line. A major
hangingwall anticline occurs immediately to the east of
the thrust plane itself and defines the eastern slopes of
a palaeo-estuary area at the northern end of Kolpos
Livadhiou.
The occurrence of sculpted embayments in the topographic
profile is very striking and is analogous to mass-wasting
processes that have been seen to operate on steep
hillslopes12,13. Significant talus fans containing large
blocks of Paleogene limestone are found immediately
down-dip from the scallop-like hillside valleys (Fig. 10).
The occurrence and derivation of the material combine
to suggest that gravitational instability set up by the
steep westerly dip of bedding is a major driver in creating
the present topography.
Significantly, this phenomenon
also characterises other valleys
and coastal cliffs on the island,
most notably on the north-eastern
flank of Ormos Mirtou and along
the eastern side of Kolpos Agias
Kiriakis itself. Consequently, the
present valley shape appears
to be controlled not only by
the solid geology but also by
recent processes of rockfall and
landsliding. The clear implication
is that prior to geomorphic
degradation, the valley itself was
once much deeper.
Four distinctive narrow, steep-sided
outwash channels occur within the
Thinia area, all of which extend to
the coast. The northern channels
merge into one, which displays an
ever-shallowing gradient and less
pronounced incision as it leaves
the Paleogene limestone outcrop
and approaches Kolpos Agias
Kiriakis. The two channels draining southward both primarily lie
within the most easily eroded
fine-grained Miocene sediments.
The eastern tributary appears to
have cut a waterfall (now dry)
where it crossed Paleogene limestones before entering the sea at
Agio Sotira.
A notable flat-lying area lies
towards the south of the Thinia
valley immediately to the west
of the small settlement of
Katochorion. The strata lie unconformably above easterly-dipping
Paleogene and Miocene strata and
the Agio Sotira Thrust and appear
to onlap onto both the landslipped
rock debris to the east, the
Paleogene limestone bedrock on
its western flank and the headward
parts of the drainage channels
described above. The strata consist
of horizontal, laminated, finegrained, poorly consolidated
clastic lacustrine sediments.
The stratigraphic relationships
between these various geomorphic
elements enable us to piece
together the following story of
how the Thinia area evolved in
time and space. The valley walls
experienced periodic landslips that
led to its becoming infilled. New
drainage pathways were set up as
the rapidly evolving, landslipmodified topography led eventually
to the formation of four main outwash channels. Further landslides
below the Petrikata-Kondogourata
Fig. 10: The Geology of Thinia. Digital Elevation Model image of the Thinia valley with draped geological and geomorphological
features. This view from the SW highlights the landslip coverage and scalloped embayments that characterise the eastern hill
slopes of the Thinia valley from which debris was derived. Key - Green: Cretaceous; Dark orange: Paleogene; Light orange:
Miocene; Yellow: Raised beach; Light blue: Lacustrine; Dark blue: Landslipped; Grey: Outwash channels. The viewpoint for Fig. 8a
is indicated. Source: Survey Map (Underhill) elevated by DEM data from the Hellenic Military Geographical Service and processed
by OziExplorer 3D.
Fig. 11: Geological sections illustrating end-member geological models. (A) depicts the subsurface projection of bedrock.
(B) provides alternative view in which rockfall debris forms only surficial cover. Detailed test awaits outcome of further
geological and geophysical methods, and boreholes. Source: John Underhill, field observations.
area led to channel capture and the development of internal drainage with the formation
of Lake Katochori. Slumping and erosion of the landslipped barrier or sill eventually
led to drainage of the lake.
We constructed a series of cross-sections across the Thinia valley (Fig. 11) to address
the issue of what form the bedrock profile takes, and whether this confirms or refutes
the possibility of a marine connection once existing there. The most important crosssections were those constructed in the highest parts of the valley (>175m). The results
all imply that a significant palaeovalley could once have existed, was infilled by landslipped and slumped Holocene cover derived largely from the eastern hillslopes
-probably in a succession of landslips rather than a single valley-filling event.
Significantly, since irrespective of where it is constructed along the length of the
Thinia valley, the bedrock profile projects to a level at or below sea-level, the intriguing
possibility remains that this palaeovalley once formed a narrow marine channel
connecting Kolpos Agias Kiriakis with Kolpos Livadhiou. Until more evidence emerges
to refute the idea, the present Paliki peninsula could, therefore, once have been an island.
Volume 16 • No. 9 • 17
Quest for Ithaca continued from page 17
a
b
Fig. 12: (a) On 12 August 1953 an earthquake of magnitude 7.2 devastated
Kefalonia: this aerial photograph taken shortly after indicates the extent of coseismic
cliff collapse. (b) Large boulders of Paleogene limestone forming part of scree
deposits on the eastern side of the Thinia Valley. The boulders were derived from
higher hillslopes and are part of a scree canopy consisting of angular boulders of
all sizes. Source: British Pathe newsclip, with permission; © John Underhill.
Fig. 13: Overview of presumed channel course. Robert Bittlestone contemplating
predicted site of Strabo’s buried channel immediately west of Petrikata.
© John Underhill.
Dates and rates
While the age of the landslipped material remains uncertain
at present, it is possible to place some constraints on timing
from artefacts found below or within the disaggregated
sediments. Firstly, the discovery of a wall buried beneath
landslipped material demonstrates that human activity
predated burial. Second, the identification of numerous
angular fragments of brick, glazed pottery and even sizable
houses within the sedimentary cover itself suggests that
much of the land slippage took place in the last few
hundred years.
The geological and geomorphological evidence allows a
diagnosis of previously greater relief in the Thinia valley
28
Geoscientist
Fig. 14: Possible course of Strabo’s Channel. Indicated route lies wholly within Miocene
marl outcrop and corresponds to projected intersection of surface dips in Paleogene
bedrock with sea-level. Course of the resultant channel form is covered by loose breccia in
central areas of the Thinia valley, interpreted to have been derived mainly from the eastern
mountainside via a series of catastrophic co-seismic landslides. Source: Digital Globe
Quickbird satellite – false infra-red imagery.
that may have been large enough for a narrow marine
connection to have existed. However this possibility must
remain speculative until definitive dates constrain each
geomorphic process identified. Significant burial has
undoubtedly taken place over the past 3000 years, much of
it in historical times. It is hoped that cosmogenic isotope
and optical thermoluminescence analysis of samples taken
during the recent field season will help to constrain and
quantify the ages of the more significant rockfalls.
The process continues unabated, as on 12 August 1953,
when a magnitude 7.2 earthquake caused major damage and
loss of life. A documentary filmed a few days later included
aerial photography of major cliff collapse and landslip (Fig.
12), attesting to the importance and scale of catastrophic
slope failure in the immediate vicinity of Thinia. It may be
no coincidence that the main village on its landslipped
eastern margin is called Petrikata, or “fallen rocks”.
Our current priority is to quantify dates and rates. The most
significant site for placing an upper limit on deformation is
the ancient lake-bed that covers the drainage channels and
onlaps onto the landslips. If 14C dates for lignitic material
derived from these sediments prove to be >3500 years old,
this may mean that Paliki was permanently connected to
the rest of Kefalonia during the late Bronze Age period.
However, if the lake-bed dates prove to be significantly
younger, this will increase the likelihood that earlier landslides infilled the topographic profile and buried a former
marine channel.
Further tests
Geological, geomorphic and geophysical tests are now
in progress that will allow us to choose between three
competing hypotheses.
• Around 1200 BC the terrain at the isthmus was well
above sea level, as it is today.
• There was a thin terrain connection, such as between
Lefkas and the mainland (Fig. 2).
• There was no terrain at that time above sea level and
Paliki was a ‘sea-girt’ island.
An outcome supporting the first hypothesis would mean
reconsidering just where Strabo was describing his
“narrows”and “wave-inundated isthmus”. However, any
suggestion that he was referring to a location further south
towards Argostoli can be ruled out, since in contrast with
Thinia, absolutely no geological or geomorphic evidence
exists there in support of a Holocene marine connection.
If either of the other hypotheses is valid then Paliki will
correspond to Homer’s description.
If the Thinia valley once extended down to sea level the
most likely reason for its original formation was that it was
created by subaerial erosion during the last glaciation
(c. 30–20,000 BP) when sea levels are thought to have
been more than 100m below the present day14,15. Sea
level rise over the following 10,000 years is a plausible
mechanism for its drowning and the creation of a marine
connection between Kolpos Agias Kiriakis and Kolpos
Livadhiou. Subsequent marine erosion and slope instability
caused by seismic events then instigated landslides and
the valley became filled.
As well as geological and geomorphic mapping, we have
conducted offshore sidescan sonar surveys, carried out in
2005 in collaboration with IGME. In June 2006, land based gravity profiling, resistivity and ground penetrating
radar (GPR) surveys were undertaken. The latter was
performed with state-of-the-art equipment provided by
GSSI of New Hampshire with the support of their
resident specialist, Dan Welch. Results are being analysed.
If these and other tests confirm that a buried marine
channel exists at sea level, its most likely course can now be
determined in the field (Fig. 13) and is indicated in Fig. 14.
Drilling and coring at strategic sites along the channel’s
putative course should also enable us to obtain essential
14C dates for the sedimentary fill of the palaeolake covering
the rockfall debris. If these boreholes indicate that rockfall material
extends down to sea level at the highest elevations, then this may
justify the use of more advanced techniques, such as seismic
reflection data and superconducting airborne gravity gradiometry.
These methods would enable a 3D subterranean model of the
entire Thinia isthmus to be constructed non-invasively. The results
will go a long way to determining whether the existence of
Strabo’s Channel is a historical reality that may yet provide us
with an elegant solution to a 3000 year old mystery.
…and what of Doulichion?
If Paliki was really Homer’s Ithaca, what then was the ancient
name of the island that is now called Ithaki? Interestingly, it
appears that today’s Ithaki has indeed been given the name of
Doulichion in the historical record, from Virgil through to the
Venetians6. Its main town, now called Vathy, was called Dolicha as
recently as 1675, when it was visited by Jacob Spon and Sir
George Wheler.
So if Strabo’s Channel is confirmed as a late Holocene geological
feature, then not only will we have found Homer’s “rocky isle”but
the ‘lost’ island of Doulichion will reappear on our maps too. And
once again, we shall be able to credit Homer with remarkable
geographical precision.
Acknowledgments
As well as the continuing collaboration between James Diggle, Robert Bittlestone
and myself, this project has involved over 50 other international contributors whose
names have already been acknowledged in Odysseus Unbound. I would additionally
like to thank the following people, all of whom have helped us to advance the
project since the book was published: Tony Hayward (BP) for his generous financial
support of the geophysical research; Constantine Perissoratis and colleagues
(Institute of Geology and Mineral Exploration); former Minister Petros Tatoulis,
Victoria Tsoukala, Andreas Sotiriou and their colleagues (Hellenic Ministry of
Culture); Vassilios Rouhotas, Mayor of Paliki; Anna-Maria Constandaki for her help in
local negotiations over access and drilling; Mike Kaplin, Roger Hipkin & John Dixon
for geoscientific field support and advice; Dan Welch (GSSI, New Hampshire);
Christopher Boddy, Kirsten Hunter, David Taylor & Neil Taylor (University of
Edinburgh) for field assistance; Eleni and Pantelis Analytis, for generously providing
accommodation; and the residents of Thinia and northern Paliki for their continuing
enthusiasm and active cooperation.
Project Web site: http://www.odysseus-unbound.org/
References cited:
1 Strabo c. 1. Geography.
2 Kraft, J. C., Rapp, G,. Kayan, I. and Luce, J.V. 2003. Harbor areas at ancient Troy: Sedimentology and
geomorphology complement Homer’s Iliad. Geology, 31, 163-166.
3 Goekoop, A.E.H. 1908. Ithaque La Grande. Beck & Barth, Athens.
4 Volterras, G. 1903. Kritiki Meleti peri Omerikis Ithakis (A critical study of Homeric Ithaca), Athens.
5 Tsimaratos, E. 1998. Poia I Omeriki Ithaki? (Which is Homeric Ithaca?). Etaireias Meletes Ellenikes
Historias, Athens.
6 Bittlestone, R. with Diggle, J. and Underhill, J.R. 2005. Odysseus Unbound: The Search for Homer’s
Ithaca. Cambridge University Press. 618 pages, 340 colour illustrations. ISBN: 0521853575.
7 Underhill, J.R. 1989. Late Cenozoic deformation of the Hellenide foreland, western Greece. Bulletin
of the Geological Society of America, 101, 613-634.
8 Pirazzoli, P.A., Stiros, S.C., Laborel, J., Laborel-Deugen, F., Arnold, M., Papageorgiou, S. & Morhange,
C. 1994. Late Holocene Shoreline Changes related to Paleoseismic events in the Ionian Islands,
Greece. The Holocene, 4, 397-405.
9 Stiros, S.C., Pirazzoli, P.A., Laborel, J & Laborel-Deugen, F. 1994. The 1953 Earthquake in Cephalonia
(Western Hellenic Arc): Coastal Uplift and Halotectonic Faulting. Geophysical Journal International,
117, 834-849.
10 Accordi, G., Carbone, F. & Pignatti, J. 1998. Depositional History of a Paleogene carbonate ramp
(Western Kefalonia, Ionian Islands, Greece). Geologica Romana, 34, 131-205.
11 Bizon, J. 1967. Contribution à la connaissance des foraminifers planctonique d’Epire et des îles
ioniennes (grèce occidentale) depuis le Paleogene jusqu’au Pliocene. PhD thesis, Paris, France.
12 Willett, S.D., Slingerland, R., and Hovius, N. 2001. Uplift, Shortening and Steady State Topography in
Active Mountain Belts, Amer. J. Sci., 301, p. 455-485.
13 Densmore, A.L., Ellis, M.A. and Anderson, R.S. 1998. Landsliding and the evolution of normal faultbounded mountains, Journal of Geophysical Research, 103, 15203-15219.
14 Pirazzoli, P.A. 1996. Sea-Level Changes: The last 20,000 years. Wiley, Chichester.
15 Runnels, C. N. and van Andel, T. 2003. ‘The Early Stone Age of the Nomos of Preveza: Landscape
and Settlement’. In Landscape Archaeology in Southern Epirus, Greece, I (edited by J. Wiseman and
K. Zachos). American School of Classical Studies at Athens, Athens.
Volume 16 • No. 8 • 29
Odysseus Unbound: The Search for Homer’s Ithaca
• Odysseus
Unbound
is
a
ForeWord 'Big Ten' Exceptional
Book for 2005
• Joint winner in The Living Past
category of University Presses
The book describing this project is called:
Odysseus Unbound: The Search for Homer's Ithaca
Cambridge University Press, October 2005.
Robert Bittlestone, with James Diggle and John Underhill
618 pages, 340 colour illustrations. ISBN: 0521853575
Website: http://www.odysseus-unbound.org
“This curious, spellbinding book is a masterpiece of writing for the general public. The geological
argument in particular is first-class and leaves me in no doubt about the possibility of the theory
being proposed.”
Professor Tjeerd van Andel, Honorary Professor in Earth History, Quaternary Science and Geoarchaeology, University of Cambridge.
“This book is a gem. Its reconstruction of prehistoric Ithaca has a convincingly Homeric 'look and
feel' to it. Reading the Odyssey is unlikely ever to be the same again.”
Professor Gregory Nagy, Francis Jones Professor of Classical Greek Literature, Harvard University
and Director of the Center for Hellenic Studies, Washington DC.
“Odysseus Unbound presents a highly readable
personal account of what can happen when an
enthusiast with a compelling synthetic vision glimpses
a solution no specialist has seen and uses his
considerable resources of energy and curiosity to bring
renowned experts like Professors Underhill (Geology,
Edinburgh University) and Diggle (Classics, Cambridge
University) to focus on solving a puzzle that has
mystified scholars for centuries. Robert Bittlestone may
one day emerge as Homeric studies' Alfred Wegener of
the Internet age.”
Dr Ted Nield, Editor, Geoscientist magazine