THE REUSE OF MARBLE IN THE EASTERN

MEDITERRANEAN IN MEDIEVAL TIMES

By
Godfrey Goodwin
i
Precious
building
materials have
always
been used
over
and over
again
-
the
spoils
of the architectural
war. A section of the Colosseum is now the Palazzo
Farnese and the Church of the Wisdom of God in the
respectable Surrey
town
of
Kingswood
harbours
capitals
from
Ephesus,
the
Studion,
and the
Myrelaion
in
Istanbul,
besides
a
quantity
of
Byzantine
marble. Precious marbles were trans
ported
far and wide
by
sea
and,
although
it is not
surprising
that
porphyry
from
the Red Sea coast of
Egypt
is used in the
Pantheon,
it is
interesting
that Giallo
Antico from
Algeria
or Tunisia and Pavonazzetto from
Phrygia
which
decorated,
for
example,
the Basilica Julia in Rome have been found at
provincial
Colchester.1
The
quarries
of Proconessus
supplied
a
low-grade
but harmonious
grey
and
white marble in
great quantities
to both the
Byzantines
and the Ottomans. The
effect of these marbles
on the character of Ottoman
architecture,
matched
by
the reuse of others not
procurable
fresh
cut,
is
important
not because of the
influence of the
past
in terms of
polychrome beauty
but because of its influence
on the
proportions
of columns
on
portico
and colonnade.2
Like
Ephesus, Constantinople
was a
depot
for marbles and there are still
marble
shops
in the former Phanariot section of the
city today.
Without such
wholesalers,
Hagia Sophia
could never have been built so
rapidly.3
Our know
ledge
of the marbles in the church comes not
only
from
Procopius
but from the
detailed authorized account of Paul the
Silentiary,
Poet Laureate and Head of
the Civil Service. His
accuracy
is vindicated
by
former doubts over a statement
that black rivers crossed the
pavement
of the
nave,
which were resolved when
the
prayer rugs
were
removed to reveal that it was
divided
by
dark ribs of marble
which
might
at least be called brooks.
Only
the
provenance
of the
huge porphyry
columns of the exedrae is in
doubt.
They may
have come
from Rome
and, indeed,
been stored for some
grandiose
use
because of their
unique
value, although
the bronze bands
probably
mark where several have been broken
or are
composite. Porphyry,
which is the
hardest of all
marbles,
is notorious for
wearing
out
cutting
tools, yet
it cracks
easily.4
The columns of the
courtyard completely disappeared, probably
before the
Ottoman
conquest
and
so at the time when the Venetians
stripped
the
principal,
west
faqade
of the
church,
leaving
one fractured
panel
of Proconessian marble
hanging
still as
evidence of its former
grandeur.
This
piece
was
probably
one of a
series,
like the 36 floor slabs in the centre of the West
Gallery,
that were cut in
GOODWIN, Godfrey, The Reuse of Marble in the Eastern Mediterranean in Medieval
Times , Journal of the Royal Asiatic Society of Great Britain and Ireland, No 1 (1977), p.
17-30
18 REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN
PLATE I
Venice. St. Mark's. The Tetrarchs and marble
panels
from
Constantinople.
slices from a
single
block
through
which the veins ran true and
opened
up
to
pro
duce a
continual
pattern
of mirror
images.5
It was
only
this west face of the
Great Church that could be seen
properly
because the other flanks
were masked
by
the
palace
and
secretariat,
library,
and other
subsidiary buildings.
Its marbles
went to enhance the basilica of St. Mark
along
with materials from the immense
church of St.
Polyeuctus,
which had been built between A.D. 524 and 527
by
Anica
Juliana,
the sister of Justinian. As with
Hagia Sophia
this was a
very
short
time in which to erect a stone and brick
building
almost
as
large
as the
Suley
maniye mosque.
The so-called
Syrian pedestals
which stand outside St. Mark's
on the Piazzetta side were
matched
by
identical
pedestals
discovered
during
the
excavation of St.
Polyeuctus
at
?ehzadeba?i.6
It was at this time that Dr. Firath of the Istanbul
Archaeological
Museum
retrieved the lost half foot of one of the Tetrarchs
on the Piazzetta corner of
St. Mark's while
working
in the enormous
buried rotunda which forms
part
of
the remains of the
palace
of Romanus
Lecapenus,
below the Church of the
Myrelaion (Bodrum Cami).
The Proconessian and other marble of St. Mark's and other churches in Venice
in
part,
at
least,
was not mint
cut,
but looted from dead and
living
monuments
of
Constantinople.
Those used
on
the Piazzetta wall
were as
haphazardly slapped
up
as other marbles
hung
on kiosks of
Topkapisaray
later.
REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN
19
PLATE II
Marmara Island
(Proconessus).
The Maible
Quarry.
II
The Proconessian
quarries
cover the southern
extremity
of the modern Mar
mara
Island between Istanbul and the Dardanelles
opposite
Erdek. The centre of
life is the
battered, formerly
Greek
village
of
Palatya,
which is set into a well
protected bay
and is one of several
villages
or hamlets in the area. It can be
reached
by jeep
over hill tracks and more
easily by
sea,
although
this is
usually
rough
even when the water off the northern tourist coast is dead calm. The site
is
large,
for the marble has been
quarried
almost at random for thousands of
years.
The
present
track leads
up
the hillside
past giant,
overgrown
blocks which
20 REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN
PLATE III
^^^^^^^^^^^^^^^H^^^H^^
~f
*
'\{itfiffiMt^BBKKB^^^KUmmUBBm
Marmara Island
(Proconessus).
Plato back to back with Socrates.
relate to those at
Palmyra
and
Baalbek,
where three
2nd-century
monsters
measure
62
x 11 ft. and
weigh
600
tons;7
but these were
quarried locally:
the
blocks at Marmara had to be
shipped
in stouter boats than the takkas which
ply
the route from
Palatya
to Istanbul
today.
15
years ago,
the owner of a
small
quarry opened
a new section in an area
unworked since Hellenic times. He
came
upon
an unfinished column and
a tran
som with
Sophocles
carved
on one side and Pl&to on the other. Thus
they
were
worked to an
advanced
stage
in the
quarry
where conditions were
sufficiently
bad to reduce
a man's
average working
life to ten
years:
hence the
use
of slave or
REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN
21
PLATE IV
Marmara Island
(Proconessus). Quarrying
tools.
criminal labour
by
Christians and Muslims alike
?
which is
why
the
population
was Greek until recent times.
After the Ottoman
conquest,
columns as such were no
longer shaped
until the
18th
century,
for the
quantity
of
Byzantine
columns available obviated the need
to
preserve
such a
skill,
although
the
segmental
facets of the enormous
piers
of
the
mosque
of Sultan Ahmet
were,
in a
sense,
an
exception.
The
quarry
was
employed cutting
blocks to be sliced into thin revetments or
pavement
or to be
carved into the standard stalactite or
lozenge capitals
of the Ottomans. The tech
niques
have altered little down to the
present day. Formerly
small holes
were
bored into the
quarry
face with the aid of water and sand until the block broke
from the bedrock. The historic tools included
picks,
some of which were chisel
ended,
jumpers,
which
nowadays
are driven into the stone to make holes for
blasting charges,
masons'
points,
and
lifting dogs
or scissors. Instead of
being
dragged,
the
rough-cut
blocks now
go
down to the
quay by lorry
and thence to
the next
bay
where there is a
factory
with one
engine
to drive the four
cutting
machines,
each armed with 50 saws. These cut a
block into 50 slabs in six
days.
Sand is still filtered into the
groove
to eat into the marble and water to cool the
blade
just
as when the work was done
by
hand. Unlike the Greeks and
Byzantines,
the Ottoman masons
appear
to have finished their work on the
building
site
just
22 REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN
PLATE V
Edirne.
Courtyard
of the
Mosque
of Selim II.
as the
Saljuqs
did before them.8 I have watched a
young
mason start on a
roughly
chalked block and
copy
a stalactite
capital
in less than a week in the Tabhane
court of the
Siileymaniye complex.
Because Marmara Island was a
quarantine
station,
we have numerous refer
ences to the
quarries throughout
Ottoman times and this
paper
is but
a
traveller's
tale to add to a
long
list. Ahmet Refik
quotes
an order in council for marble that
has been
shipped
to
Eregli
to be taken to Edirne for use in the
great mosque
of
Selim II there9 and
a more elaborate order for the
mosque
of the Atik Valide at
Oskudar in 1570. This order to the
Qddti
of Iznik and
Sapunca
states that some
buildings
contain marble in his district. The intendent of the
mosque
had been
REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN 23
PLATE VI
Istanbul. The
courtyard
of the
Mosque
of Ahmet
I, requiring
28 marble
columns,
including
Proconessian and
Egyptian granite.
obstructed
by
local
people
and it was therefore decreed that no-one
might
inter
fere or
frustrate him from
extracting marble;
and the
qadThad
to see that trans
port
was
paid
to deliver the material to the location of the
mosque.
He was to
avoid
negligence
but also
damaging property
while the marble
was
extracted.10
This order is
likely
to refer to the columns of
Byzantine
and
even Roman build
ings
needed for the madrasa and
hospital
colonnades of the
complex.
An order for stone from Karamursel would
imply
that materials were
gathered
at the
depot
there from
Anatolia,
but in 1604 Ahmet I ordered that marble was
to be
brought directly
from Marmara Island for his father's tomb and since the
ships
sent were late further
ships
were to be sent there.11 A further order states
that the marble still had not reached the
architect,
Ahmet
Dalgic, (or
"the
Diver").
Ahmet I had continual
difficulty
in
getting
his orders
obeyed
as his
troubles over Iznik tile
supplies
also make
plain.12
When
George Sandys
visited the island at the same
period
he stated that it
was
"formerly
called Proconessus... celebrated for the
quarries
of white marble:
and thereof now called Marmara where
a number of
poor
Christian slaves do hew
stones for the
magnificent mosque
which is now
building
at
Constantinople by
this sultan."13 In 1723 La
Mottraye reports
that "marble is extracted in
large
quantities
from
Pallatia, Marmara,
for
mosques employing
200 vessels".14
24 REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN
Lechevalier, however, says
that where there were once
sizeable towns one now
sees
nothing
but
a few miserable Greek
villages
and
a
few
monasteries,15
while
Texier,
40
years later,
states that one
exploits
blocks of little
size, slabs,
and
tombstones for
Constantinople.
One detaches the block from the mountain with
chisels after which one lifts it from its bed with iron
wedges.16 Today
the annual
output
is
12,000
cubic metres and its
cheapness
is due to low labour costs and
primitive living
conditions.17
Ill
Their
stripping by
Ottoman builders and
peasants explains
the lack of
Byzan
tine monuments in
Anatolia,
so
that
a small church near
Nigde
has a
spurious
importance.
Before the
coming
of the
Saljuqs,
cities like
Kayseri
and
Konya
had
cathedrals and
many
churches.
Inevitably they
made
way
stone
by
stone for
Saljuq hospitals, colleges,
and
mosques.
The
Saljuqs,
in common with the
Armenians,
whom
they employed among
others as
builders,
had
an
appreciation
of
some facets of
architecture, yet
were blind to others. While the Armenians
disregarded symmetry
to such
a
degree
that one
wonders whether the
placing
of
doors off axis were not
deliberate,
the
Saljuqs
were
unable to
distinguish
between
capitals
and
bases,
at least not at
first,
and used none at all if
a column fitted
without them. It was
not,
we
may
be
sure,
because of that
spurious hadith,
to
quote
Michael
Rogers,18
that
alleges
that
nothing
so much wastes the substance
of
a
believer
as
architecture.
The
saray
and citadel at
Konya
were built with care and so was the Ala'ettin
mosque,
which
required
a
great
number of columns of the
Byzantine
and Roman
periods.
At
Sivas,
when we
inspect
the Gok madrasa
or
the
Barugidi
madrasa,
we
also meet with the
rough
reuse of columns and marble
facets, carefully
recut
when
necessary.
It should be noted that the
Byzantines
and the
Saljuqs
did not
use true ashlar like the Greeks but economized
by insetting
a
rubble and mortar
core. Neither in
Saljuq
nor even in Ottoman
times,
when smooth columns
became
increasingly
hard to
find,
were
segmented
or
fluted columns reused.
Indeed the
borrowing
of an
antique
fluted column was so rare that their
appear
ance in the
mosque
of Ivaz Pasha at Manisa in the 15th
century
is almost the
only
occasion.
The worst
despoilers
of Anatolian
antiquities
were Western
museums such as
the Bode in East
Berlin,
which harbours a
great
amount of Miletus: so much that
Ilyas Bey,
who built
a
mosque
at Balat out of the ruined
city,
or Isa
Bey,
who
quarried
the ravished hulk of the church of St. John at
Ayaslug, by
contrast
behaved with restraint. At both
mosques
the marble is recut and remodelled in a
Syrian
manner
where
necessary.
Already
the
Beylik period
of these
mosques presaged
the exactitude and
engineering
skills of the
Ottomans, although
its
beginnings
were not
auspicious.
REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN 25
PLATE VII
Istanbul.
Suleymaniye Complex.
Tabhane
Courtyard
with reused Proconessian columns.
The
capital
in the
foreground
is in the
process
of
carving.
The
mosque
of the
Hudavendigar
was
built at Assos in the midst of unlimited
materials from moribund ruins and
yet,
because an
inscription
was
meaningless
to the
patron,
it was
used
clumsily
as a decorative lintel.
In such
circumstances,
it is
surprising
that,
although
old
buildings
from the
tomb of
Cyrus
at
Pasargadae
downwards were used
as
mosques,
and ruins used as
quarries,
the forms of
antique
monuments were
copied
so
rarely
that the
Aqsa
mosque
at Jerusalem
surprises just
because its
plan
relates to that of a
Constantin
ian basilica. But
spendthrifts
have
empty purses
and
by
the end of the 16th
cen
tury
materials were
increasingly
hard to
find,
which accounts for the
clumsy
reuse of
transoms,
akin to those in the church of Constantin
Lips,
and in the
mosque
of the Yeni
Valide,
both in Istanbul.
A
huge complex
like that of
Suleyman
I with ten columned
courtyards
required copious quantities
of
columns, including nearly
all
remaining
in the
Hippodrome,
while
Gyllius reported seeing
the
pillar, meaning
the column of the
Virgin
near the church of the
Holy Apostles,
on
top
of the Fifth Hill
being
broken
up
for the
mosque19
but elsewhere
says
of
Suleyman's
tomb,
"now build
ing",
that
elegant
marbles were sent from all over the
empire,
old
quarried
slabs
from
palaces
of
Byzantium,
Greece,
and all
Egypt.20
It was Dr.
Cyril Mango
who found the edict of Manuel which had
hung
in the
26 REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN
church of
Hagia Sophia by pursuing
an
observation of
Pigafetta's
in 1567.21 The
slab had been
placed
over the
porch
of
Suley
man's tomb.
Sinan,
who had
sur
veyed
the
cathedral,
had an
acquisitive eye.22
A less
conspicuous
but more
devouring
form of waste was the
employment
of
Byzantine capitals,
not
just
graciously
as coffin
rests,
but in foundations. Masses of miscellaneous
Byantine
objects pack
the
garden
of
Hagia Sophia
due to road
works,
where a
city
is built
over a
city.
The
Emperor
Theodosius'
triumphal
arch was
dug
out of the widened
road at
Beyazit
and related reliefs can be
seen in the
exposed
foundations of the
hammdm across the road. The
porphyry sarcophagi
of the first Eastern
emperors
have been found in the courts of the
saray
and
on the
railway
embankment. The
foundations of the baths of Sinan Pasha at
?ehzadeba?i proved
to be
a storehouse
of
Byzantine
cut stone.
The same
wastage
was true of
Egypt,
where
no
marble was
quarried
after the
Islamic
conquest.
This resulted in the curious Mamluk
technique
of
cutting
thin
strips
of marble
veneer sometimes from columns. This method
simplified
the
intricacies of
arching
with
joggled
voussoirs,23
which otherwise needed the
gift
of
magic
to
erect,
although
Roland Mainstone has
now
offered a solution to the
problem.
I would here add that
Egypt
was so
short of wood that it came to be
prized
as much as
marble and bronze and to be
purchased
or
stolen when needed
in new
buildings.24
The Ottoman sultans
were
thrifty
too. When an
old kiosk was
demolished,
the marble and tiles
were
stored for
reuse
elsewhere,
the latter in beds of fern
and moss. When Ahmet III built his
library
in the Third Court of
Topkapisaray,
the columns of the Kiosk of the Pool which
preceded
it were re-erected
as the
colonnade before the
present Treasury.25
Elsewhere the dearth of columns had a
more serious
effect,
like that on the otherwise
important mosque
of Semiz Ali
Pasha at Babaeski built
by
Sinan where the
portico
is
sadly ignoble
for a work
acknowledged by
the
master,
although
he
certainly delegated
the actual
erec
tion to a
subordinate. Even at the
saray,
where the redoubtable Murat IV was
the
patron,
the famous Revan and
Baghdad
Kiosks were faced with
a
hotchpotch
of marbles and awkward
pieces
of
poryphyry
that were too intractable
to be
recut.
Villagers
have borrowed stones from monuments and still
do,
and this
accounts for those
kervansaray
and other walls where the
footings
and several
courses above them have been nibbled
away
all over Anatolia to
expose
the
rubble core of the
masonry
for the instruction of architectural historians. Just
so,
but with
concepts
of
grandeur,
did Selim the Grim
strip
the chamber where
the
Caliph kept
the momentoes of the
Prophet
in Cairo and
brought
them back
along
with the loot to be reset in the outside wall of the Pavilion of the
Prophet's
Mantle at
Topkapisaray.
There
they
are less
unexpected
than are similar revet
ments at the
mosque
of the
Beylerbey
of
Egypt, forban
Mustafa
Pasha,
at
Gebze.
REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN 27
PLATE VIII
Istanbul.
Edirnekapi.
The
Mosque
of Mihrimah. Reused
Byzantine
Proconessian
columns,
one
clearly showing
the holes to which crosses and icons were
fixed.
IV
Just as the
Byzantines began
a
building by buying
columns for
it,
so
did the
Ottomans collect columns before
starting
to erect their
buildings
-
the same
28 REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN
PLATE IX
Istanbul. The
Mosque
of
Hekimoglu
AH
Pasha, showing
a reused
Byzantine granite
column cut
roughly
at the neck to fit the
capital.
columns. With the
imperial mosques
of
Beyazit
II or
Suleyman
I
-
and of the
erstwhile
mosque
of Fatih
-
the
preparation
of the
pair
of
huge
columns between
the central
piers
had to be related to the
height
of all other
columns, including
REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN 29
those of the
courtyard.
This also affected the
span
of the arches and therefore
the area of both
mosque
and court.
With viziers'
mosques
there
might
be no
important
internal
columns,
with the
result that those of the
portico,
when it was
ideally
related to the
proportions
of
the
mosque itself,
were modules of the whole.26 Where the
relationship
is success
fully
achieved at the
mosque
of Selim II at
Konya
or
Karapinar,
or of Rustem
Pasha at
Tekirdag,
the result is a concordance of
parts:
if the
relationship
is not
achieved,
the
result,
as at the
Kur?unlu
Cami at
Kayseri,
is a
discordance between
the bulk of the
prayer
hall and the
elegance
of the double
portico.
The same is
true to some extent of the
mosque
of Pertev Pasha at Izmit where the inner series
of columns is
insufficiently grand
to sustain
aesthetically
an
outer,
subsidiary
portico.
And indeed one of the
many
touches of
genius
about the
organization
of the
Selimiye mosque
at Edirne is the manner in which Sinan ordered a
disparate
col
lection of columns in the
courtyard
with
deceptive
ease
by resorting
to corner
piers
of unusual
design.
We can then
enjoy
the
paradox arising
from the
power
of the
past
to control
the future
through
the
raw
materials of architecture matched
by
an architect of
a new era who was
capable
of
solving
the
problems
created
by
the
scarcity
of
precisely
these materials.
NOTES
1
N.
Davey,
A
history of building materials, London, 1961,
5.
2
The column is also
important
because its use
by Byzantines
and then Ottomans was a
revolutionary
advance in terms of structure. The inertia of Roman
building
was
replaced by
a
quasi-ribbed
construction which was to be the
greatest
achievement of Gothic architects.
3
The
depot
at
Ephesus may lighten
the darkness of the balderdash about the
proven
ance of the columns in
Hagia Sophia
since some
may
have been
purchased
there. The author
of
myths
of
Hagia Sophia
was the
Anonymous
of
Banduri,
11th
century,
who invented what
he did not know.
4
I have been unable to locate
large
circular
slabs,
such as those before the
great
door of
Suleyman's mosque
or in the
Pantheon,
which are neither cracked nor
patched.
s
At least
nowadays,
it is
impossible
to
predict
before
cutting
whether a block will
open
in this manner or not.
6
M. Harrison and N.
Firath,
1964-1965 Sarachane
Arastirmalan, Istanbul, 1966,
133
n.8,
and 134. The debris
layers
fit a
13th-century
date for the
transportation
of the Pilastri
Acritani to Venice.
7
Davey, op. cit.,
16.
8
See the unfinished columns of the
f
ifte Minare
Medrese, Erzerum,
for
example.
In
Cairo there is the
Mosque
of Sultan
Hasan;
see M.
Rogers,
The
spread of Islam, London,
1976,103.
9
Ahmet
Refik,
Istanbul
hayati,
on
altinci asirda
(1553-1591), Istanbul, 1935,
lists a
number of orders in
respect
of this
mosque.
,0
ibid., 21,
Sect. 15.
11
Ahmet
Refik,
Istanbul
hayati,
onbirinci asirda
(1,000-1,100), Istanbul, 1930-1, 26,
Sect. 29.
12
ibid., 33,
Sect.
64-5; 34,
Sect.
66;
and
36-7,
Sect. 70.
30 REUSE OF MARBLE IN THE EASTERN MEDITERRANEAN
13
G.
Sandys,
A relation
of
a
voyage begun
Anno Domini
1610, London, 1627,
22.
14
A. de la
Mottraye,
Travels
through Europe, Asia,
and into
parts of Africa, trans.,
3
vols.,
London, 1723-32,
344.
,s
J. B.
Lechavalier, Voyage
de la
Propontide, Paris, 1800,1,
24.
16
C.
Texier, AsieMineure, Paris, 1882,
162.
17
Information
supplied by
the Muhtar of
Palatya, quarry
owner.
18
Rogers, op. cit.,
12.
19
P. Gilles
(Gyllius),
The
antiquities of Constantinople, London, 1729,
219.
20
ibid.,
51.
21
C.
Mango,
"The conciliar edict of
1166",
Dumbarton Oaks
Papers, XVII, 1963,
315-30.
22
So had the architect of Ibrahim Pasha's
mosque
at
Nev?ehir,
who for this
18th-century
monument
purloined
the columns of the
nearby Sungur Bey
foundation of the 14th
cen
tury.
See A.
Gabriel,
Les monuments turcs
d'Anatolie, Paris, 1931-4,
156.
23
Joggled
voussoirs occur as
early
as A.D. 526 in the mausoleum of Theodoric at
Ravenna.
24
Rogers, op. cit.,
104.
25
S. H.
Eldem,
Koskler ve
kasirlar, I, Istanbul, n.d.,
104.
26
It is worth
considering
whether the abandonment in
mosques
such as that of Ahmet I
of the lateral columns of a model such as the
Suleymaniye,
at the cost of a contraction of
the diameter of the
dome, may
not
simply
be due to the exhaustion of columns after the
pair
in the
Piyale
Pasha
mosque
of 1573.