Shang Ritual Bronzes: Casting Technique and Vessel Design
Author(s): Robert W. Bagley
Source: Archives of Asian Art, Vol. 43 (1990), pp. 6-20
Published by: University of Hawai'i Press for the Asia Society
Stable URL: http://www.jstor.org/stable/20111203
Accessed: 27/08/2010 12:59
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=uhp.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
University of Hawai'i Press and Asia Society are collaborating with JSTOR to digitize, preserve and extend
access to Archives of Asian Art.
http://www.jstor.org
Fig.
6
iA. Detail
of Fig.
ib, Fang
yi.
Shang Ritual
Casting
Robert
W.
Bronzes:
Technique
and Vessel
Design
Bagley
Princeton University
he past three decades have seen a considerable
in our understanding
of Shang
ritual
change
owes much
ar
to
recent
bronzes.
it
Although
was
this
discoveries,
chaeological
change
brought
about mainly
done outside
the field of
by work
in the study of fabrication methods.
archaeology,
a
The
starting point was
study of Shang mold
out
Orvar
carried
Karlbeck, who was
fragments
by
able to show conclusively
that Shang founders did
on the lost-wax
not
's paper
process.1 Karlbeck
rely
was
in
but
his
work
attracted
little
1935,
published
notice at the time; it was only in the 1960s that an
obvious correspondence
between
the appearance
of
to
bronzes
and
the
used
make
them
Shang
technique
was
This belated discovery
has
finally
recognized.
to
the
the
bronzes
rewritten
of
be
required
history
from a point of view which
takes casting technique
into account. The art historian can no longer ignore
or
it to an appendix,
for neither
technique
relegate
the character
of individual objects nor the history
as a whole
can be understood
of Shang design
to fabrication methods.
without
reference
no
But
if the art historian
the
longer enjoys
of
without
luxury
mentioning
discussing
design
neither can the historian of metal
tech
technology,
to ignore
so would
to
afford
do
be
nology
design:
to assume that
never made
casters
technical
Shang
on artistic
con
decisions
grounds.
Shang bronzes
front us inescapably with
the problem
of under
and design interacted,
and
standing how technique
a first step toward
is
this
solving
problem
perhaps
to recognize
that we are formulating
it in terms that
no
caster would
have understood.
The dis
Shang
tinction we make between
and
technique
design is
a construct
from our own
inherited
intellectual
not attempt
to do without
the words
"technique"
and "design,"
discussed
but the examples
should
to
it clear that if we continue
make
analyze Shang
bronzes in terms so artificial, we must proceed with
as
as
great care. No formulation
simple
"technique
influences design" will do justice to the experience
of casters who did not think in these terms.
the Shang caster learned the two things
art
The
firm line we draw between
together.
history
to do with
and the history of technology
has more
the structure
of our universities
than with
the
of
bronzes.
The
article
will
present
making
Shang
The object shown in Figure 1, a vessel of the type
can serve to introduce
the main features of
fang yi,
cast about
late Shang bronzes.
It was
1100 b.c.,
at
site
of
the
last
perhaps
Anyang,
Shang capital.
The principal motif
of its decoration
is an animal
1
tradition;
ib.
ca.
Fig.
Fang yi,
Harvard
Collection,
12th century
University
b.c. h. 29.8
Art Museums
cm.
Winthrop
(1943.52.109).
7
head (Sargon of Akkad?)
from Nineveh,
Fig. 3. Bronze
After Max Mallowan,
2370 b.c. Iraq Museum,
Baghdad.
and Iran (London,
Mesopotamia
1965), fig. 74.
ca.
Early
and all, the sunken lines should not be
to
were
resemble
lines cut in metal:
expected
they
a
not carved inmetal but in the
of
clay
preliminary
a
model.
The caster began with
of the
clay model
a mold
on the
to make,
vessel he wished
formed
to cast a bronze
and used the mold
model,
replica
of the model. The decoration
of Shang bronzes owes
to the fact that itwas executed
much of its character
in clay rather than in metal.
But this is only the most
influence of
superficial
a matter
the technique on the vessel's appearance,
decoration
b.c. Dimensions
of
of a fang yi, ca. 12th century
cm.
Museum
of Art, New
7
4.5
shown,
by
Metropolitan
Sackler
Gift of Arthur M.
York,
(1974.268.2).
Fig.
area
2. Detail
face with
occupies
staring eyes, the taotie, which
the main register on the vessel proper and reappears
on the lid. Narrower
registers contain
upside down
seen in
creatures
animal
motifs draw
These
profile.
Set
the viewer's
attention
off by a
irresistibly.
of
carved
dense,
pattern
ground
finely
spirals, they
do not interact with each other but are held fast in
a
array of clearly defined
compart
symmetrical
are bounded
ments.
The compartments
by heavy
vertical flanges and plain horizontal
strips, the plain
strips coinciding with gaps in the flanges.
An enlarged detail of the decoration
of another
one
connection
the
obvious
between
suggests
fang yi
of these objects and the technique used
appearance
to make
them (Fig. 2). The fine sunken lines with
their vertical walls
and sharp edges point imme
to
decoration
made by casting. The drafts
diately
of these energetically
drawn,
manship
precisely
is
lines
that
unlike
of a craftsman
angular
utterly
tools on cold metal,
and it immediately
using
bronzes
from
the products
of
distinguishes
Shang
other metalworking
traditions. The effect is indeed
so unfamiliar
a few observers
that it has persuaded
with
unacquainted
decoration
copies
another medium.
8
casting technique that the bronze
some lost art form executed
in
But
since the vessel was
cast,
of the caster's handwriting;
casting technique holds
the key to much more fundamental
features. Tech
a role in any art form, of course, but
nique plays
in the Shang design tradition
it seems for a time to
have played
the leading role. To understand
how
this came about we must consider for a moment
the
uses to which bronze was put in ancient China.
In China metal was used
the Bronze
throughout
to
make
ritual
and
vessels.
It was not
weapons
Age
to
for
other
which
the heirs
us,
purposes
employed
seem
of Egyptian
and Near Eastern
civilization,
more
familiar
and more
natural.
The Akkadian
bronze portrait head shown in Figure 3, dating from
the late third millennium
b.c., is a royal monument
of a kind abundantly
in the art of the
represented
are unknown
ancient Near East. Such objects
in
ancient China, where we find only a rather feeble
art and no interest at all
interest in representational
or the
in portraiture
of rulers. Shang
depiction
were not
metalworkers
required by their patrons to
nor were
the
human
depict
figure,
they expected
to describe
other features of the everyday world,
and as a result they enjoyed considerable
freedom
seem to
in certain directions.
In particular,
they
Fig.
4. Mold
diagram,
(London,
Artifacts
1976),
lost-wax
process.
p. 72, fig. 10.
After
Henry
Hodges,
free to experiment
with
their casting
con
and to develop
forms of decoration
to
are
it.
and
related
genial
Design
technique
closely
in Shang bronzes because
the designs emerged from
in casting technique. The early history
experiments
as an
of Shang bronzes might
almost be described
of
the
of
the
section-mold
exploration
possibilities
have
been
method
technique.
a formulation
But
is not very
of this kind
on
in
it
the
takes
substance
abstract;
meaningful
only
can be adduced. If the
if specific illustrations
designs
grew out of the casting technique, how exactly did
this happen? How
does a casting method
influence
a
Can
actions
direct
the
of the
design?
technique
craftsman who uses it?
To answer
these questions we must begin with
we believe
the technique
itself. Nowadays
that
were
cast
in
bronzes
section
but
this
molds,
Shang
is a fairly new conviction.
Before
i960 most West
ern students of the bronzes
took it for granted that
founders
used
the
lost-wax
process. Scholars
Shang
interest in technical matters
who had no particular
may
simply have thought
as synonymous with fine
of the lost-wax
process
casting.
in Figure 4 explains
The diagram
the lost-wax
process in its simplest form. A founder who wishes
a bronze cat
to make
a
begins by making
vaguely
core
core
is
This
then
covered
cat-shaped
clay
(1).
is given the exact
with a layer of wax,
and the wax
awax
shape desired for the finished cat: the result is
a
cat with
core
A
is
mold
constructed
clay
(2).
by
is
the mold
packing
clay all around the cat; when
is melted
baked the wax
out, but the core remains
in place, held skewered by pins called chaplets
(3).
is then turned upside down and bronze
The mold
is poured into the space previously
occupied by the
wax
Once
the
metal
is
has
solidified
the mold
(4).
cat
broken open to reveal the casting?a
bronze
with
a
clay
Before
assumed
starting
core.2
1935 it seems to have been universally
that Chinese bronzes were cast in this way,
from a model made of wax. As any foun
Fig.
5.Mold
diagram
Powell.
for
the fang
yi of Figure
1.
Drawing
by
Whitney
is
another
knows,
however,
procedure
to the bronze caster, one which
does not
wax models.
If the clay mold
in Figure 4
employ
so that it
in fitted sections,
had been constructed
in pieces and then
could be removed from the model
dryman
available
reassembled
around the core, there would have been
no need for amodel
that could be melted
away. The
caster would
have been at liberty to construct
his
model
from clay or any other convenient material.
in Figure 5 suggests how this
The mold diagram
alternative
could have been used to cast
procedure
the bronze vessel shown in Figure 1 (the body only:
the lid was cast separately). The vessel was probably
so that the
cast
core could be
upside down
larger
Since
from
below.
the
model
carried
supported
clay
all the decoration
that was to appear on the finished
the object at the center of the diagram can
bronze,
on which
either the model
is
the mold
represent
or the finished vessel
out
formed
of
being
coming
assume for the moment
the mold;
that it is the
was
model.
around
this
model
and
Clay
packed
removed in four flat sections; afterwards
the model
could be discarded.3 The mold
sections were
then
reassembled
around a suitable core,
the spacing
not by
between mold
and core being maintained
spacers of the thickness de
chaplets but by metal
sired for the casting (these would
be incorporated
9
for a fang yi, ca. nth
century
Fig. 6. Part of the mold
Yinxu Miaopu
Guo Baojun,
from Anyang
Beidi. After
Zhou
1981), pi. 31.
tongqiqun zonghe yanjiu (Beijing,
b.c.,
Shang
core was
in the finished
inserted
vessel). Another
to form the hollow foot of the vessel, and the bronze
was
poured.
Parts of the mold for a similar vessel have recently
at an Anyang
been unearthed
foundry site, and they
not quite
the
mold
that
suggest
diagram in Figure 5 is
a
correct
the
mold
for
fang yi
(Fig. 6). Apparently
been
would
have
with
decoration
high-relief
in eight sections
rather
removed
from the model
have carried
than four, and the sections would
mortises
and tenons to ensure that they could be
the
removal
from
reassembled
after
accurately
model. But in all essentials
the moldmaking
process
to that illus
casters corresponds
used by Anyang
trated in Figure 5, and it is a process which Karlbeck
was able to reconstruct
in his 1935 paper by studying
a collection
of mold
said to have been
fragments
Some
of his mold
found at Anyang
(see Fig. 11).
mortises
carried
and
tenons, many were
fragments
a
traces
and
of bronze
left in
few
had
scorched,
came
that the fragments
them. Karlbeck
concluded
in fitted sections and that
from molds
constructed
the molds had been used to cast bronze.
At least one reader saw immediately
that Karl
of
beck's paper had implications
for the history
an article
In
in
1937, Leroy
Shang casting.
published
Davidson
that vessels
like the tripod of
suggested
Figure 7 might be the earliest of decorated Chinese
bronzes.4 Davidson's
argument depends on the ob
servation that the section-mold
technique allows the
caster access to the interior of the mold: he can carve
decoration
10
directly
in the mold
surface,
and
lines
Fig. 7. He,
Collection,
b.c. h. 22.9 cm.
15th century
Brundage
Asian Art Museum
of San Francisco
(B60B53).
ca.
cut into the mold will
raised lines on the
produce
finished
bronze. The
lines and dots
thread-relief
seen in Figure 7 are therefore just what we
might
to find on the earliest decorated
if
bronzes
expect
at cast decoration
in their first attempts
Chinese
founders chose to carve on the mold rather than on
It should be added that thread relief is
the model.
not a form of decoration
likely to arise in lost-wax
casting.
The
lost-wax
caster
must
carve
on
the wax
model or the finished bronze; the mold is closed and
he does not have access to its inner surfaces.
In the light of subsequent
finds
archaeological
Davidson's
The
reasoning
seems
more
cogent
than
ever.
earliest decorated
bronze
vessel yet known
from China, a small pitcher of the shape called jue,
was discovered
in 1975 at Erlitou
in a level dating
b.c.
from about the middle of the second millennium
on
Its
which
the side
decoration,
appears
(Fig. 9).
of the vessel opposite
the handle, is shown in Figure
9 in a rubbing. This simple pattern of lines and dots,
which
reappears in Figure 7 as the border to amore
is more primitive
elaborate
than anything
design,
at the time Davidson
known
but it was
wrote,
as he foresaw,
direct
of the
produced,
by
working
Fig.
8. Undecorated
jue from Erlitou
(3rd stratum),
b.c. h. 12 cm. After Herum
chutu Shang Zhou
century
(Beijing,
1981),
no.
ca.
16th
qingtongqi
1.
3
b.c.
of the second millennium
that he could
had discovered
lines in mold
object by carving
conse
sections.
had far-reaching
The discovery
an
it
for
established
quences,
enduring preference
It
decoration.
than coldworked
for cast rather
ex
marks
the beginning
of the Shang caster's
of the section-mold
of the possibilities
ploration
mold. By the middle
caster
the Chinese
a bronze
decorate
technique.
At
the
no
wrote
there was
Davidson
to support his conjecture,
evidence
archaeological
and it does not seem to have attracted much notice.
Karlbeck's
for the use of section molds,
arguments
not fare much better.
decisive
though they were, did
in the West
into the 1960s most writers
Well
to speak of Chinese
bronzes as lost-wax
continued
or else
castings
ignored technique altogether. When
turn their attention
Western
scholars did finally
were
to section-mold
casting,
they
apparently
not by Karlbeck's
work
but by the work
prompted
of Chinese
scholars who
had long since taken
as their point of
Karlbeck
departure.5 Karlbeck's
that
had
used
1935 study
proved
Shang founders
to make
section molds,
the
but it had not managed
issue seem important.
In the course of the 1960s, however,
the section
status
of
achieved
the
mold
finally
theory
Karlbeck
after
suddenly
thirty years
orthodoxy:
and
lost-wax
seemed
seemed
obviously
right
it
is
quite impossible
obviously wrong.
Nowadays
to look at
and think of lost-wax
Shang bronzes
time
B
ca. 16th century
b.c.;
(3rd stratum),
Fig. 9. a, Jue from Erlitou
h. 22.5 cm. After
the handle,
of the side opposite
b, Rubbing
Kanan-sh? Hakubutsukan
Fong, ed.,
1983), pi. 1;Wen
(Tokyo,
The Great Bronze Age of China
17.
1980), fig.
(New York,
a
of students brought up
casting, and to generation
on the section-mold
it is puzzling
that
theory
scholars could ever have done so. The complete
success of the new
to forget
theory has made it easy
once stood in its way and the
the obstacles which
overcame
them.
insights which
The delayed but complete victory of the section
seems to have been brought
about
mold
theory
was
two
One
these
of
contributions.
John
chiefly by
used
careful study of the joining methods
Gettens's
a
like
vessel
casters.6
Chinese
Thirty years ago
by
the four-ram zun of Figure 10would have been cited
as
casters used the
proof that Shang
unhesitatingly
in sections
to remove a mold
lost-wax
technique:
from amodel with four sets of spiralling rams' horns
But the conclusion
that the zun is
is unthinkable.
on the assumption
a lost-wax
rests
that it
casting
was cast in one piece, and Gettens
showed that such
In Figure 10 the
bronzes are not one-piece
castings.
horns and ears of each ram and the four small dragon
heads on the vessel
shoulder,
twenty
altogether
cast individually
and then
separate pieces, were
in the mold
of the
for the remainder
embedded
of the zun thus involved
vessel. The fabrication
11
v ,.,
Ahvuil,;MouId
KARI.IIKCK:
Fig.
io. Four-ram
century
(Beijing,
zun
from Hunan
b.c. h. 58.3 cm. After
no. 17.
1976),
Ningxiang,
Zhongguo
ca.
,,
12th
gu qingtongqi xuan
Fig.
ii. Plate
2 from Orvar
Karlbeck,
Anyang
Moulds,
Bulletin
of theMuseum ofFar EasternAntiquities, Stockholm7 (1935).
the twenty
casting operations,
in
the
locked
precast
pieces being
place during
of
the
casting
twenty-first.
which was
Gettens's
paper on joining methods,
to the
one obstacle
in
removed
1967,
published
the
section-mold
away
theory
by
explaining
features
that had seemed to require the lost-wax
twenty-one
separate
But the observation
turned the
which
technique.
been
had
made
tide against lost-wax
casting
already
in 1962 in a short paper of fundamental
importance
Fairbank.7 Mrs. Fairbank pointed out that
byWilma
an
is openly
the Shang moldmaker's
technique
nounced by his designs: the fang yi of Figure 1almost
shouts that it was cast in a mold divided vertically
on the axes of its
heavy flanges.
The simplest ideas can sometimes be very elusive.
In Karlbeck's
between
1935 paper the relationship
not leap to the eye
and
decoration
does
technique
in
because
the mold fragments
he illustrated were
11 it is not at all
in Figure
very poor condition:
of the decoration
obvious
that the compartments
to
But
scholars who
mold
sections.
correspond
also overlooked
studied bronzes
rather than molds
12
ca. 14th century
12.
detail of leg. h. 20.8
b.c., with
Ding,
Arthur M. Sackler Gallery,
Smithsonian
Courtesy
D.C.
The
rounded
Institution, Washington,
tips of the legs
are modern
restorations.
Fig.
cm.
a
seems obvious to us in Figure 1;
relationship which
must
the explanation
simply be that before Mrs.
com
Fairbank no one who
looked at the bronzes
a
bined an interest in their design with
knowledge
of how they were made. Nowadays
the connection
seems self-evi
between
decoration
and technique
do not remember
Karlbeck
's
dent; scholars who
Fig.
14. Mold
Artifacts,
Fig.
12.
for the ding of Figure
13.Mold
diagram
Drawing
by
in the
After
Powell.
Bagley,
Shang Ritual Bronzes
Sackler Collections
Mass.,
1987),
(Cambridge,
100.
Whitney
Arthur M.
fig.
or who
arguments,
have
never
even
read
his
paper,
can see the
in section molds
because
they
in
But we
the
bronzes.
of
the
method
consequences
as
see the connection
writers
such
because
only
our
drew
Fairbank
and John Gettens
Wilma
believe
attention
to
it.
bronzes were
the 1960s few pre-Anyang
known to scholars outside China, and the researches
centered on the
and Fairbank necessarily
of Gettens
of
the
bronzes
Anyang
period.
highly sophisticated
to a
in 1962 could draw attention
Fairbank
While
In
between vessel design and casting tech
relationship
was
in Anyang
obvious
which
instantly
nique
at
that rela
how
could
she
bronzes,
only guess
came about. But the excavations
of
the last
tionship
full
twenty years,
by increasingly
supplemented
to
have
found
of
material
earlier,
begun
publication
of the Shang bronze
throw light on the beginnings
can now refine our understanding
of
industry. We
and
between
the relationship
technique by
design
a historical
dimension: we can reconstruct
adding
it into being.
the sequence of events which brought
at primitive
To do so we must
look more
closely
castings.
The mold
for a simple
round vessel,
such as the
diagram,
p. 73, fig. 11.
section-mold
process.
After
Hodges,
12, was normally
ding of Figure
Erligang-phase
into three identical sections, each of which
divided
carried the same decorative
pattern
(Fig. 13). The
sections were
the
divisions
between
aligned with
so that the sections could be
of
the
vessel
easily
legs
The decoration
of the
from the model.
removed
three times in the cir
finished
bronze
repeats
cumference
of the vessel, and since the units of the
tomold sections, the legs fall
decoration
correspond
at the same points as the vertical divisions between
units (Fig. 12, detail).
the sec
The relationship just described,
involving
of the
mold
the
of
the
tioning
assembly,
placement
of
the
calls
for
the
subdivision
and
decoration,
legs,
to
two comments.
First, it is a relationship
specific
re
the ding shape; the same features are differently
lated in other vessel types. Second, the relationship
is
by
no means
an
automatic
consequence
of
the
use
Consider
the diagram used in a
of section molds.
to explain
the
handbook
standard archaeological
The
section-mold
technique
(Fig. 14).
object being
cast is a portrait head like the one shown in Fig
ure 3. The head is
is
shaped in clay, then the mold
in sections. A consid
formed on it and removed
to free
erable number of sections will be required
but the sectioning
of the
the mold from the model,
in the appearance
of
mold will not be expressed
the finished head (the head will not have as many
If the caster's as
faces as the mold has sections!).
a
to
is
the com
make
signment
king's portrait,
nature
will
be auto
of
the
mold
assembly
posite
matically
suppressed.
In other words,
the section-mold
technique does
on
not inevitably
it
its
the objects
character
stamp
case of a portrait head we may
In
the
produces.
or
to discover
indeed find it difficult
impossible
was
what
from
the finished
procedure
object
in constructing
If the section
the mold.
followed
mold
technique did express itself in the decoration
13
Fig.
ca. 15th
from Hubei
15. Ding
Huangpi
Panlongcheng,
b.c. h. 54 cm. After
no.
Zhongguo gu qingtongqi xuan,
century
Fig. 16. Ding,
Freer Gallery
D.C.
(60.18).
b.c. h. 24.8 cm. Courtesy
12th century
Smithsonian
Institution,
Washington,
2.
can
of Shang bronzes,
the explanation
only be that
was invented
casters
the decoration
whose
pur
by
not
to
did
them
conceal
the
poses
require
technique.
On this point the evidence
of the most primitive
seems
clear.
bronzes
The
earliest substantial
fairly
are ten small jue
in
unearthed
China
castings yet
from Erlitou,
objects easily cast in section molds.
casters
The Erlitou
evidently
began by making
undecorated
vessels (Fig. 8), for nine of the jue have
no decoration,
but as soon as some careless mold
maker
scratched
the interior of a mold
section, he
a
section makes
found that a scratch in the mold
raised line on the finished bronze. The tenth Erlitou
bears the simple decoration
of dots and
jue, which
lines already discussed, must bring us very close to
the time of this discovery
(Fig. 9).
was first carved
The fact that cast decoration
on the model
in
the
mold
than
rather
directly
might
seem
but it is to this accident
that
inconsequential,
the Shang artistic
its unique char
tradition owes
acter. Notice
that in Figure 9 only one side of the
came
vessel carries decoration. When
the moment
to execute
the decoration,
the craftsman
had the
sections in front of him, but the bronze vessel
mold
did not yet exist: at that moment
he must have been
less
about
the
vessel
than about the
finished
thinking
was
to
content
mold
and
he
decorate only
sections,
one of them, the
one.
largest
H
ca.
of Art,
In the case of a later and more
sophisticated
an
a
site, the en
casting,
ding from
Erligang-phase
tire circumference
but
of the vessel was decorated,
to think in terms of self
continued
the decorator
sections
lines are
contained
mold
(Fig. 15). The
were
carved
that the patterns
raised,
showing
same
in
the mold
sections; the
pattern was
directly
on
so
carved
each section,
the decoration
repeats;
in the
divisions
12, the vertical
and, as in Figure
a
are
decoration
the legs, establishing
aligned with
between
the
important
relationship
simple but
decoration
and the shape of the vessel.
is
This alliance between
shape and decoration
it
is
characteristic
of Shang bronze
and
design,
almost
the Shang moldmaker's
by
guaranteed
to
invited Shang founders
technique. The technique
The
repetitive,
produce
designs.
compartmented
resulted were
related to
intimately
designs which
their
the shapes on which
because
they appeared
and the
the sectioning of the mold,
layout reflected
sectioning of the mold had been decided already by
the shape of the vessel which was to be cast. Later
the bronze dec
of casters elaborated
generations
oration far beyond anything
imagined by Erligang
but the alliance with vessel shape
phase craftsmen,
remained
the ding
Anyang
intact
the Shang period. On
throughout
a
from the
of Figure
vessel dating
16,
the
coincide
with
the
period,
legs again
units of decoration,
and in this
are marked
with
emphatically
same
Additional
the
flanges.
give
flanges
heavy
note of
to the central axis of each unit.
emphasis
Introduced
the Anyang
just before
(see
period
caster a dramatic way
offered
the
Fig. 21), flanges
to announce
the organization
of his designs
and
to stress the
between
deco
thereby
relationship
ration and shape. The prevailing
of
interpretation
boundaries
between
case the boundaries
the flange does not, however,
concede
that flanges
were
It holds
for
the
sake
of
this
adopted
emphasis.
is the Shang caster's way of
instead that the flange
the scars which
appear
dealing with mold marks,
on a
if
into the space between
bronze
leaks
casting
In Figure
sections.
fitted mold
15 the
imperfectly
on the
band of decoration
has three discontinuities
axes where
the mold was divided,
and it is com
seen in Figure 16were
that
the
monly argued
flanges
to hide such discontinuities.
In other
introduced
the Shang caster is supposed to have made
words,
a virtue out of a defect,
the mold
exaggerating
accents.8
marks and converting
them into vertical
a
This theory, which
proposes
specific relation
is open to a
between
and
ship
technique
design,
number of objections,
both general and particular.
ca. 12th century
Fig. 17. Guang,
Freer Gallery
of Art,
Smithsonian
D.C.
(38.5).
b.c. h. 23.5 cm. Courtesy
Institution,
Washington,
It overlooks
the actual source of the
is to be found in Neolithic
pottery;
equation of mold marks and flanges
the two do not
many vessels on which
most
seriously,
it supposes
that
flange, which
its mechanical
the
disregards
coincide; and,
the evolution
of
15
Fig.
18. Elephant-shaped
Freer Gallery
Courtesy
Washington,
D.C.
zun,
of Art,
b.c. h. 17.5 cm.
12th century
Smithsonian
Institution,
ca.
(36.6).
influenced
the drawbacks
invented.9
If it could be shown that a Shang founder ever
to save a casting from disfigurement
used flanges
to
indeed be forced
mold
marks, we would
by
influenced by short
conclude
that his designs were
in his technique.
But the testi
inherent
comings
and unflanged
of
mony
surviving bronzes, flanged
no
saw
casters
is
connection
that
be
alike,
Shang
tween mold marks
first a
and flanges.
Consider
the guang of Figure 17. It is
vessel without
flanges,
that the maker of this vessel was concerned
obvious
to avoid or eliminate mold marks;
it is also obvious
that his way of dealing with mold marks did not
of technically
fine
involve flanges. The existence
16
Arthur
was
by the caster's desire
of a flawed
casting
assessment of the
The
distinctly negative
technique.
on which
this last point rests
section-mold
technique
can
if
shared
have
been
by Shang founders,
hardly
on a comparison
it
because
implicitly
only
depends
Yet
with
the lost-wax
method.
despite
general
intimate
connection
be
of
the
acknowledgement
tween the
and
his
caster's
designs,
technique
Shang
writers
seduced by the theoretical
simplicity of lost
wax casting and unaware of its
practical difficulties
to assume
continue
that the lost-wax
technique
than
offers a better way of casting Shang bronzes
were
in
the
those
bronzes
which
technique
Shang design
to overcome
ca. 13th century
b.c. h. 30.1 cm.
Courtesy
Fig. 19. You,
M. Sackler Gallery,
Smithsonian
Institution,
Washington,
D.C.
to reconcile with
castings is difficult
unflanged
as the solution
which
describes
flanges
theory
technical
problem.
pressing
any
to a
do carry flanges
of vessels which
The testimony
zun shown in
is no different.
The elephant-shaped
is no reason
18
there
but
has
several
Figure
flanges,
to do with hiding
to believe that they have anything
mold marks. They were
added, in locations where
could easily have been ground away,
mold marks
to the curve of the elephant's
to draw attention
trunk and tail. Other mold divisions fell at locations
to the finisher,
less accessible
yet those locations
were not
if the Shang caster
with
flanges:
supplied
to save his
had been in the habit of using flanges
it is on the ele
from disfigurement,
decoration
to find
that we should expect
phant's hindquarters
the
them. As in the case of the guang, however,
omission of flanges could hardly be said to have left
to find where
It is not difficult
this vessel disfigured.
the mold joins fell, but only a twentieth-century
look for them.
observer would
The you of Figure 19 ismore regular in shape than
the guang and the elephant
zun, and thus more
it
shows no more
but
of
sign
Shang bronzes,
typical
and flanges.
between mold marks
of a connection
it
is on the swing handle, where
The only flange
zun from
20.
M3,
Unflanged
Zhengzhou
Baijiazhuang
b.c. h. 27.7 cm. After Herum
chutu Shang Zhou
14th century
qingtongqi, no. 33.
Fig.
ca.
a mold mark;
cannot have been meant
to conceal
to the curve of
it serves instead to draw attention
the handle, which
repeats the shape of the vessel
is intricately
The
vessel
and
decorated,
proper.
at least four vertical mold divisions must
though
it
have been required to release mold from model,
is unflanged. Here as on the guang and zun, flanges
and mold marks
lead independent
lives. Flanges
were
added wherever
the caster felt the need of a
a silhouette.
to dramatize
vertical accent or wanted
The objects shown in Figures 17-19 are among the
most
of Shang bronzes,
and the
lavishly decorated
zun
are
moreover
and
guang
unusually
complicated
in shape. Few vessels can have posed greater
tech
at the
nical difficulties
but
the
stage,
moldmaking
the use of flanges.
difficulties were resolved without
it clear that Shang founders
Such examples make
a
did not automatically
add a flange
everywhere
mold division fell, and the examples
could be mul
not
Yet numbers
alone will
tiplied
indefinitely.
are
the
that
and
mold
marks
disprove
theory
flanges
it
in
because
still
be
that
connected,
might
argued
the beginning the flange was a device for hiding mold
In other words,
marks.
the fact that the flange was
at some stage used as a
not rule
design element does
out the
that
the
element
possibility
design
originated
as a device for
hiding mold marks.
To deal with
this possibility we must turn to the
zun from
21.
Renmin
Gongyuan,
Flanged
Zhengzhou
b.c. h. 24.9 cm. After Herum
chutu Shang Zhou
14th century
no. 76.
qingtongqi,
Fig.
ca.
earliest
its first
flanged bronzes. The flange made
before
the
appearance
Anyang
shortly
period,
around the end of the Erligang phase. The unflanged
vessel shown in Figure 20 dates from that time. To
con
it is essentially
judge from its decoration,
and the
the first flanged bronzes,
temporary with
to introduce
must
decision
therefore
have
flanges
in producing
been taken by casters engaged
vessels
in Figure 21 shows
like this one. The zun illustrated
the result of their decision:
large curvilinear
flanges
project from the decorated
registers on the axes of
the vertical
mold
were
divisions.
as
mold marks,
we
must
believe
the prevailing
theory maintains,
that a caster found the mold marks on the vessel of
that he added the flamboyant
Figure 20 so disturbing
seen in
21 to hide them.
devices
Figure
Surely just
is true. The caster who made the vessel
the opposite
of Figure 20 had no need to hide mold marks; it takes
an expert
to find any trace of mold marks. What
to
to clarify
mattered
the caster was
the organi
a
at
zation of his design
time when
the growing
to make
of the patterns had begun
the
intricacy
to
lines between
difficult
compartments
dividing
was
to
not
His
find.
hide flaws but to
problem
motive
which prompted
emphasize boundaries. The
the introduction
of flanges was not concealment
but
If flanges
added
to conceal
advertisement.
17
22. Gui from Hubei
ca. 14th century
Fig.
Huangpi
Panlongcheng,
in the Arthur M. Sackler Collections,
Shang Ritual Bronzes
fig. 214.
Another
vessel of about the same time
unflanged
appears in Figure 22. The rubbing of the decoration
shows immediately
where
the mold was divided.
to the
The vertical
line at that point was important
two units of deco
caster
it
because
Shang
separated
on Chinese bronzes have fallen
ration. Most writers
into the habit of calling this line a mold mark, but
is that really correct?
If we use the term "mold
to mean a line that the caster would
mark"
like to
are
then this is not amold mark. Neither
eliminate,
the horizontal
the band of
lines above and below
None of these lines is accidental;
all of
decoration.
them could have been removed. The vertical
line
is not an unfortunate
of
the
pro
casting
by-product
cess, it is an essential part of the decoration. When
to overwhelm
such
intricate
threatened
patterns
caster
them
the
the
marked
with
lines,
boundary
accent he could find?as
in Figure 21.
heaviest
18
b.c. h.
17.4 cm. After
Bagley,
But where
did the Shang caster find this par
ticular accent? The prevailing
theory holds that the
an
is
flange
simply
extravagantly
enlarged mold
the enlarged mold
mark, but it cannot explain why
seen in
mark
takes the distinctive
hooked
form
and
the
hooked
of
the
earliest
2i,
Figure
profile
flanges is in fact the key to their ancestry. The same
on the
a common
profile appears regularly
legs of
bronze
the
vessel
type,
early
flat-legged
ding tripod
a
(Fig. 12), and such tripods copy in metal
shape
at least two thousand years earlier
which originated
in Neolithic
of the most
archaic
pottery. Typical
versions
Neolithic
of the shape is a pottery
ding
b.c.
site of the fourth millennium
from an east-coast
vessel are
(Fig. 23). The flat legs of the Neolithic
and
of
embellishment
their
way
radially placed,
by
are
or
outer
serrated.
edges
pinched
Legs
a feature of the
in this way
embellished
remained
its history.
throughout
flat-legged
dir?g shape
flat
with
serrated
Pottery
legs have been
examples
found at Erlitou
and other early Bronze Age sites
are
ding in bronze
(Fig. 24), and the first flat-legged
The
faithful copies of the pottery vessels
(Fig. 12).
pottery
tripod shown in Figure 24 is contemporary
with
bronze jue vessels from Erlitou
the primitive
than the Erligang-phase
and only a little earlier
12. The bronze
re
of
bronze
Figure
ding
tripod
in
it
and
the
metal,
preserves
pottery
shape
produces
the jagged outline of the leg.
and exaggerates
Recall how the mold for such a flat-legged
tripod
were
was divided
aligned
(Fig. 13). The divisions
so that the legs would
with
the legs of the model
not hinder the removal of mold
each
from model;
two
in
of
faces
the
lateral
its
left
imprint
leg
in the band
the
breaks
sections.
Since
mold
adjacent
the legs and the
of decoration
also coincided with
a
be
could
be
introduced
mold
divisions,
flange
tween two adjacent units of decoration
merely
by
the curly part of the corresponding
leg
extending
upward.
seen in Figure 21 have a curly
the flanges
because
they copy the serrated legs of flat
which
could be
An
embellishment
legged tripods.
added to the leg of a tripod could be added at the
mold joins of any vessel type. From a technical point
the leg on the ding and the flange on the
of view,
a
zun are
is merely
the flange
features;
equivalent
not reach to the bottom of the vessel.
that
does
leg
From an artistic point of view,
the flange
supplies
it
is
where
needed
most, at the
heavy emphasis just
vertical break between
pattern units.
Ifwe consider for amoment
only the early history
of the flange, how should we go about describing
this design element and the
the relationship between
are
two
The
caster's
certainly
technique?
things
at
not
all simple.
is
but
the
related,
relationship
to do with mold marks;
they
Flanges have nothing
were
in the deco
added to stress vertical divisions
at
ration, and when
they lie mold join lines it is only
be
because
the mold join lines are the boundaries
Thus
outline
tween units of decoration.
In other words,
flanges
were
were
not added for technical
reasons,
they
their
added for reasons of design. But what made
char
the compartmented
addition
desirable was
acter of Shang decoration,
and that character owed
to a series of experiments
its existence
with a par
ticular casting technique.
The history of the flange is evidence
enough that
the formulation
influenced
design"
"technique
of the events
would be a very misleading
description
which make up the history of Shang bronze casting.
if Shang casters had for some reason
Certainly
Fig.
23. Pottery
ding
from
Fig. 24. Pottery
from Luoyang
Shanghai Qingpu Songze
(middle level), 4th
millennium
ding
first half
Donggan'gou,
of the 2nd millennium
b.c. h. 31.7
b.c. After
cm. After
Feng Xianming
taoci shi
et al., Zhongguo
1982), pi. 7:3
(Beijing,
Kaogu
1959.10,
pi. 7:1.
to use the lost-wax method,
Shang bronzes
Yet if we ask how the
look very different.
we
section-mold
influenced vessel design,
technique
the tech
too easily fall into the habit of regarding
as a known quantity,
fixed
and
unalterable,
nique
the caster's intentions; we
which
actively modified
was forced by the limitations
imagine that the caster
to some sort of compromise
with
of his technique
ideas
of artistic
his original
idea. But the notion
in the abstract
and then imperfectly
conceived
from
is an irrelevant
realized inmatter
importation
art theory. A Shang caster would
Renaissance
to hear that his technique had
probably be puzzled
did not
The
limitations.
section-mold
technique
force him to produce
decoration,
compartmented
nor did it force him to use flanges or to carve sharp
the bronzes bet
edged lines. We will understand
ter if we
rather than negative
think in positive
not limitations,
terms. The caster saw possibilities,
is the history of
and the history of Shang bronzes
of the section
of the possibilities
his exploration
chosen
would
mold
technique.
Author's
Note:
An
earlier
version
of
this
article
was
to the conference
La Civilt? Ci?ese Antica,
Venice,
presented
am
to the
of that con
I
1-5,
1985.
organizers
April
grateful
to participate,
and to the Publications
for inviting me
ference
Princeton
of Art
and Archaeology,
Fund of the Department
University,
possible
for a grant toward publication
illustrations.
the use of color
costs which
has made
19
Notes
Bronze
i. Orvar
Bulletin
Karlbeck,
Moulds,
Anyang
Far Eastern Antiquities,
Stockholm 7 (1935): 39-60.
2. The
lost-wax
method
direct
described
of theMuseum
here
of
is
today
and
duplicate
castings
in destroying
because
of the risk involved
the original
model.
in Greek
The
lost-wax
indirect
foundries
method,
employed
as
as the seventh century
the original model
b.c., preserves
early
a section mold
on it and then
intact by forming
using the section
a wax model
or models.
to
See Ruth Whitehouse,
mold
produce
The Macmillan
Dictionary
of Archaeology
1983), p. 112;
(London,
Art Foundry
C.
Christian
H?user,
(New York,
1974); Carol
seldom
used
because
it does
not
allow
Greek Bronze Statuary
Mattusch,
(Ithaca, N.Y.
1988).
seem never
to have
reason, models
3. For no obvious
been
are unknown
founders;
castings
Shang
duplicate
in the
among
(see R. Bagley,
Shang bronzes
Shang Ritual Bronzes
in
Arthur M.
Sackler Collections,
Mass.,
1987,
Cambridge,
are too much
section
dam
troduction,
2.2). Molds
ordinarily
reused
by
to be used
from a casting
again.
aged in removal
a
Toward
of Early Chinese
4. J. Leroy Davidson,
Grouping
Parnassus
cited
Bronzes,
9.4 (April
51- Davidson
1937): 29-34,
a similar vessel
not the
in Berlin.
tripod of Figure 7 but
formerly
was
the most
notable
Shi Zhangru,
whose
these
5. Of
dates from
industry
comprehensive
study of the Shang bronze
1955: Yin
dai
Fabrication
de
zhu
Lishi
tong gongyi,
Yanjiuyuan
Zhongyang
26 (1955): 95-129.
in
Methods
the
Gettens,
John
Joining
Bronze
of Ancient
Chinese
Ceremonial
Vessels,
Yuyan Yanjiusuo
6. Rutherford
jikan
Application of Science inExamination ofWorks ofArt (Museum of
Fine Arts,
The
is treated
Boston,
1967), pp. 205-217.
subject
more
in the same author's
The Freer Chinese Bronzes,
extensively
Volume II, Technical Studies (Washington,
D.C,
1969).
Piece-Mold
and Shang
7. Wilma
Fairbank,
Craftsmanship
20
Archives
16
Design,
of the Chinese Art Society of America
Bronze
Noel
Barnard's
Bronze
in
and
8-15.
(1962):
Casting
Alloys
a more
Ancient
account
China
of
(Tokyo,
1961),
speculative
no
section-mold
makes
mention
of the con
explicit
casting,
nection
between
and design.
later writings
technique
Among
on the
the papers
in Cyril
Smith's A Search
subject,
reprinted
Mass.,
for Structure (Cambridge,
1981) should be singled out for
on
comments
their illuminating
and artistic
invention.
nology
8. This
which
interpretation,
Noel
and Wilma
Barnard
Barnard
accepted.
a list of
expedients
p.
(Bronze Casting,
for dealing
with
pp.
(Piece-Mold
Craftsmanship,
element
"evolved
from
craftsmanship."
9. Fairbank's
1962 paper
to the lost-wax
in William
implicit
seems
Fairbank,
design
superiority
the relationship
between
to have
with
originated
been
very widely
on
117) included
flanges
mold
Fairbank
marks;
has
them
12-13) described
the practical
requirements
(pp. 9-10)
method.
Watson's
tech
as a
of
an
attributes
The
same
discussion
of
unqualified
assessment
Shang
is
metal
in Ancient
East Asia,
(Cultural Frontiers
technology
Edinburgh,
to cite a more
recent
and it forms
1971, pp. 72-79),
example,
of Noel
Barnard's
the cornerstone
for
argument
oft-repeated
the independent
of Chinese
in Barnard's
origin
metallurgy:
casters
if Chinese
view,
never
they would
S?rica 22,
have
had
used
known
of
section
molds
the
lost-wax
process,
(see e.g. Monumenta
the confusion
arises
1963, pp. 225-227).
Perhaps
textbook
from comparing
abstractions
ultimately
mold
the lost-wax
rather
than
process,
process)
Most
techniques.
process"
the one
seem
as "the
writers
used
to be
aware
lost-wax
2 and
Bagley,
who
term
by that
to cast the four-ram
understand
discuss
a
"the
far
procedure
zun of
Figure
(the section
actual
casting
section-mold
than
simpler
10, and few
to
the procedures
referred
commonly
note
section
often
involve
molds
process"
(see
section
introduction,
2.6).
Shang Ritual Bronzes,
that