Content uploaded by Terry Lustig
Author content
All content in this area was uploaded by Terry Lustig on Feb 28, 2021
Content may be subject to copyright.
New
Insights
into
les
Interminables
Listes
Nominatives
d'Esclaves'
from
Numerical
Analyses
of
the
Personnel
in
Angkorian Inscriptions
Eileen Lustig and Terry Lustig
great
number
of
people
were
employed
to
maintain
the Angkor kingdom's
much-studied
stone
temples-
many
living and working
in
their
shadow,
providing
ritual
food, clothing,
music
and dance
for
tens
of
thousands
of
stone
deities
housed
within.
For
example, inscription
K.
908 (CE 1191) records
that
people
in
8,176
villages provided
rice
for
20,400 gods
in
gold, silver,
bronze
and
stone
in
the
provinces.
Much
of
the
workforce
was
linked
to
or
centred
on
these temples,
and
it
is
conceiv-
able
that
the proportion
in
the temple-based
economy
increased
over
time
as
new
foundations
were
established.
In
the city of Angkor,
over
30,000
were
employed
in
A
just
two
ofJayavarman
VIl's
temples.
Our
knowledge
of
these people has rested largely
on
temple bas-reliefs
from
the 13th century,
on
an
account
written
by
a
Chinese
envoy
at
the end
of
that
century
(Zhou Daguan 2007), and
on
the
stone
inscriptions
of
temples,
which
provide
the
data
for
this
study. The records,
at
times
running
into
thousands
of
individuals,
often providing
name,
sex,
dependants and
even
occupation, proved
too
much
for
Aymonier, who
referred
to
the 'interminable
name
lists'
in
inscription
K.
183 (CE 928)
(Aymonier 1900, 408). This
was
also the
view
of
Coedès (Coedès 1937, 52), and
until
recently,
there
has been much less
attention
paid
to
non-elites than
to
elites.
The
images
of
ordinary people
in
bas-reliefs do
little
to
explain
their
status,
while
in
the
texts
they
are
rarely
active
agents. Probably
most
of the empire's population
was
not
exactly free,
as
we
understand the
term.
Many
were
bonded
to
others,
either
by inheritance, punishment,
as
a
tribal
or
war
captive, and possibly through debt
(Martin 1998; Mabbett 1983;
see
also
Khin
Sok 1998 and
Forest
1998
for
Cambodia's
Middle
and French
colonial
period). Many
lived
in
villages
that
were
traded
or
given
to
provide temple
labour.
Aséanie 31,juin 2013, p. 55-83
Eileen Lustig and Terry Lustig
Aséanie
31
|
2013
To
date,
much
of
the
discussion
about ordinary people has been based
on
Pre-Angkorian data (acob 1979; Vickery 1998).
The
recent
publication
of
the largely-
ignored
lists
of
personnel
at
Koh
Ker,
the
short-lived
Angkorian capital (Chhom
Kunthea
2011),
makes
it
timely
to
re-examine
Angkorian period non-elites.
In
this
study,
we
have applied
mathematical
techniques
to
give
some
shape
to
the often-
featureless
lists
of
people,
to
discern
something
of
their
circumstances.
This
paper
is
in
two
parts.
In
the
first,
we
address
questions
of
how the
demographic
characteristics
and
the
socio-economic
circumstances
of
these
personnel
varied
over
time.
In
the
second
part,
we
hypothesize
a
link
between
changes
in
status
of
two
categories
of
male workers,
si
and
gho,
the
rising
influence
of
some
officials,
and
Jayavarman
IV
remaining
at
Koh
Ker
after
his
enthronement
(circa CE 928).
The
Interminable
Lists
The
people
were
recorded
as
individuals
or,
more
commonly,
as
lists ofpersonnel
assigned
to
work
for
a
foundation
-
Aymonier's
'interminable
lists'-
rangingin
number
up
to
hundreds
of
people,
that
donors
had
given
as
acts
of
merit.
Figure 1
shows
the
data
used
-
over
17,000 people
identified
individually
in
over
100 temple
inscriptions,
all
assigned
somehow
to
work
for
religious
foundations.
The graph does
not
include
the
hundreds
of
thousands
of
people
claimed
by Jayavarman
VIl
in
1186,
because he
refers
to
the
rice
producers
in
his
villages
en
masse,
not
as
individuals
as
in
the
inscriptions analyzed
here.
Only
lists
or
totals
without
significant lacunae
in
the
texts
were
used
for
calcu-
lating proportions,
to
avoid
systematic
errors,
such
as
with
the
Koh
Ker
lists, where
70
60
50
9000
8000
7000
6000
S5000-
4000
3000
2000
Z
1000
0
30
20
10
1100-99
800-99 900-99 1000-99
Date
CE
Number
of
individuals
-Number
ofinscriptions
Figure
1.
Numbers oftexts and personnel for each century showing the steady decline
in
inscriptions
and
personnel recorded after the 10th century.
New Insights
into
les
Interminables
Listes
Nominatives
d'Esclaves 57
at
first
there
appeared
to
be
a
significant majority
of
adult
males
recorded.
However,
many
of
the
texts
were
on
standing pillars whose bases
were
buried
by
rubble
or
had eroded (e.g.
Chhom
Kunthea
2011,
75),
so
that
many
females, usuallylisted after
males,
were
not
being counted.
The
bulk
ofthepeople
in
the 9th and early 10th
centuries
are
from
inscriptions
in
royal temples, respectively
from
the Roluos
group
and
at
Koh
Ker.
From
the
mid
10th century
up
to
the 12th century,
texts
were
produced mainly by
officials
recording
how they
or
their
families
had established temple
foundations
on
lands, purchased
or
granted by kings (Figure 2). The
officials
frequently
offered
their
pious
work
(the
foundation)
to
the
monarch
as
a
'royal
foundation'
in
exchange
for
its
protection
(Coedès 1954, 242).
An
important
incentive
would
have been the
immunities
granted
to
the
foundations
and
their.
personnel, exempting
them
from
various
impositions
such
as
state
corvée
and
taxes,
which
the
population
was
generally subject
to
(Lustig
2009, 196-220).
The combined
taxation
relief
granted
to
all
the private
foundations
may
well
have posed
a
threat
to
the
income
collected by the
Khmer
rulers
(Sahai 1977b, 134).
This could have
contributed
to
central
weakness, making possible
the
coup
staged
by
the
founders
of
the
Mahīdharapura dynasty
circa
CE
1080.
The
new
dynasty
instituted
a
number
of
administrative
changes:
the
political
and
economic privi-
leges, including
immunities,
enjoyed by
some
officials
seem
to
have been
curtailed
from
this
period (Lustig 2009, 197).
Thereafter,
the
number
of
new
inscriptions
si
dominant gho
dominant
80
70
60
50
40
30
officials
Jayavarman
VII
Roluos
eriod
Koh
Ker
period
10
0
1050-1099
Date CE
800-849850-899900-945 950-999 1200-1249
1250-1299
1100-1149
1150-1199 1300-1349
1000-1049
Inscriptions non-royal
Inscriptions royal
Igure
2. Texts by kings and officials (adapted from Lustig 2009, Fig. 42).
Eileen Lustig and Terry Lustig
Aséanie 31 2013
and
of
personnel
declined
until
the
reign ofJayavarman
VII.
Either
new
privileges
granted
to
royal supporters
were
no
longer
written
in
temple inscriptions,
or,
with
the
loss
of
immunities,
there
was
less
incentive
to
establish
new
foundations.
Around
the
same
time,
some
of
the officials, perhaps
those
claiming descent from
Jayavarman I1, including those
titled
loñ, ten and vāp,
who
had
been
of
relatively high
status,
and
who
were
often
related
to
the
rulers,
declined
in
importance (Vickery
1998, 406;
and
2005, 4-5).
The
väp disappeared
from
the
texts
altogether
after
the
reign
of
Udayādtyavarman
II
(circa CE 1066). Following the
decline
in
influence
of
the
officials,
we
see
a
rise
in
the
number
of
royal
texts
at
the
end
of
the
12th century
particularly those by Jayavarman
VII.
Two
of
them, inscriptions
K.
908 (CE 1191) and
K.
273
(CE 1188), suggest his capacity
to
mobilize
substantial
resources
and
to
assert
authority
over
a
population
of
hundreds
of
thousands.
The Personnel: Angkorian Period
Commoners
Serving the Temples
The
individuals
in
the
inscriptions
were
generally grouped according
to
their
social
category
or
working
role.
Each had
an
appellation
that
described
their
gender
and
status:
with
juveniles,
the
appellation
often
indicated
their
age
as
well.
Some
descriptors changed
in
meaning
or
importance
over
time.
In
Angkorian period Khmer
texts,
commoners
were
most
commonly
called
khñum
or
sometimes
the
Sanskrit
dāsa
(male)
or
dāsī
(female). They
could
fill
a
variety
of
roles, each
with
its
own
social and
legal
status
(see
Annex
1).
titles
of
93%
of
the
individuals
recorded.
rable
1
shows appellatives used
int
Among
the
adults,
si
and gho (male) and
tai
(female)
are
the
most
profuse. Interpretations
of
si,
gho, gväl, and lap
are
discussed
below.
The relatively
few
individuals
with
other
titles
(e.g.
ye,
dvańand mu)
were
counted
in
the
analysis, according
to
their
gender
and
age
if
known,
otherwise
simply
as
'personnel'.
TABLE
1.
MOST
FREQUENT
APPELLATIVES
FOR
ANGKORIAN
PERIOD
COMMONERS
Juniors; Children (kvan)
gval
lap
Adults
junior male,
late 9th
c.
to
early 11th
c.
junior
female,
late
9th
c.
to
early 11th
c.
male
toddler
male,
cf.
Pre-Angkorian
vā
Si
ghoda) (robust?)male
gho
female, cf. Pre-Angkorian ku
(me; 'ame) female,
mother
or
courtesy
title
male,
from
12th
c.,
previously
man
of
rank
female,
from
12th
c.,
previously
woman
of
rank
tai
Si
rat
female toddler
male
nursing baby
female
nursing baby
tai
rat
me
loñ
si
pau
ten
tai
pau
New
Insights
into
les
Interminables
Listes
Nominatives
d'Esclaves
The personnel could be given, bought and sold, and might be
inherited
with
their
children.
Yet
many
worked
within
temples
in
the
domain
of
the gods,
some
with high
status
functions,
so
a
moot
point is
whether
those
personnel
could
be
true
slaves.
Various
aspects
of
their
social and
economic
circumstances
have been debated,
including
how
those
of
the
Khmer
khnum
compared
with
those
ofthe
ir
South
Indian
counterparts (Jacques 1976; Jacob 1979;
Chakravarti
1980, 149-98; Mabbett 1983;
Vickery 1998,
225-74;
Sanderson 2003, 395-400). A
consensus
that
not
all
khñum
were
oflow social standing is
seen
in
Jenner's
broad
interpretation
of
the
term:
a
person
assigned
to
unpaid
labour:
slave, serf,
bondsman
or
a
person
assigned
or
offering
himself
to
the
service
of
divinities
or
sanctuaries
Jenner 2009a and 20095).
Sanctuary and Non-Sanctuary Workers
The
texts
of
many
inscriptions distinguish
between
sanctuary
and
non-sanctuary
personnel, the
latter
largely
ricefield
workers.
This
distinction
was
explicit
where
roles
were specified,
or
deducible
from
their
social category,
their
allocations
or
by
their
hierarchical positioning
in
the
text
(e.g.
Soutif
2009, 457
note
225).
In
order
to
examine
differences between the
two
groups,
all the personnel
in
the
texts
were
classified
as
sanctuary
or
non-sanctuary, generally according
to
the
criteria
set
out
in
Table
2.
TABLE
2.
CRITERIA
USED
FOR
DISTINGUISHING
BETWEEN
SANCTUARY
AND
NON-SANCTUARY
PERSONNEL
Sanctuary
indicators
nak
pamre
'servant
khnum vrah
"*slave'
of
god"
nak vrah
Sanctuary
roles
specified
Mode
of
acquisition
of
donated
worker
stated
Adults
nak
person'
nak
sre
'ricefield
person'
vnvak
"team
Village (sruk)
or
settlement
named
(usually)
Attached
to
donated
land
Listed
with
ritual
objects Duty
to
produce
a
quantity
of
a
resource
and
provisions
for
god
Assigned
to
upkeep of god
khlon sruk 'village chief', amrah
'toreman',
especially 9th
c.
especially 9th
c.
The clearest
distinction
was
in
the 9th century
Roluos
texts.
Thereafter, greater
ambiguity is
found.
For
example,
a
donated
sruk,
which
would
usually be designated
non-sanctuary
in
the 9th century,
may
have provided sanctuary
or
non-sanctuary
staff
in
later
centuries.
Again, whereas
a
foreman
khloñ sruk would normally
appear
in
early
sanctuary
texts
and
amrah usually
in
non-sanctuary
texts,
from
the
late
9th
century,
Aséanie
31
|
2013
Eileen Lustig and Terry Lustig
amrah
are
seen
with
sanctuary
lists
and
in
inscription
K.
832
(CE 889-915) preceded
by
a
khloñ
sruk.
The
individuals
named
in
sruk
lists
did
not
necessarily
constitute
the
entire
village,
and
were
often
in
teams.
Sometimes
the
same
sruk featured
in
more
than
one
donation.
The
lists
ranged
in
size
from
a
handful
to
hundreds.
The signifi-
cance
of
uncertain
classifications
to
our
results
is
discussed
below.
Not
only
were
there
declining
numbers
of
inscriptions
and
personnel recorded
after
the
10th century (Figure 1),
there
was
also
a
marked
reduction
in
the propor-
tion
ofthe
later
texts
that
listed
non-sanctuary
workers
(Figure 3). This raises the
question,
after
the
9th
century,
why
was
it
less important
to
inscribe
these names?
Fig. 1
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1000-99 1100-99
800-99 900-99
Date
CE
Sanctuary Non-sanctuary
Figure 3. Declining proportion of
texts
recording non-sanctuary personnel.
An
important
consideration
early
on
may
have been
to
ensure
legal ownership.
This
may
have
stemmed
from
the
shortage
of
labour
resulting
from
low population
densities, typical
of
Southeast
Asia
(Hall 1985, 4;
Reid
1988,
13-15
and 26).
It
makes
sense
to
record
the
names
of
people legally required
to
work.
Furthermore,
some
classes
could
be
inherited
and
there
are
cases
of
personnel
tied
to
inherited land (see
Annex
1). According
to
Mabbett, given
the
large
number
of
children
recorded with
their
mothers,
'slaves'
could
be
born
as
such (Mabbett 1983, 53),
which
may account
for
the
children
being
recorded
as
well.
New Insights
into
les
nterminables
Listes
Nominatives
d'Esclaves 61
During
the
Angkorian period,
as
the
number
of
people associated
with
craft,
trade
and bureaueracy
swelled
the
populations
around
temples, people
were
likely
more easily
controlled,
reducing
the
need
for
inscribing
the
names
of non-sanctuary,
low-status
workers
or
for
providing
an
inducement
tO
work
for
foundations.
In
addition,
whole
Communities
were
often
being
treated
as
assignable
resources
in
ways
not
apparent
earlier.
This
is
first
seen
in
the
late
9th
century royal inscriptions
of
Roluos,
where
among
the
large
number
of
temple
workers
are
people
in
territorial
groups,
at
times
from
relatively distant
localities
(Vickery 1999, 49).
We
also
begin
to
see
a
shift
from
the
donation
of
people
ith
land
to
the
provision
of
land
with
people.
In
later
texts,
there
are
instances
of
land
and
villages
with
attached
people, notably
K.
420
(CE 1078-1177)
or
whole
villages
allocated
by
rulers
to
officials,
as
in
K.
235 (CE 1052),
or
purchased
to
provide temple labour,
seen
in
K. 216 (CE 1006) and
K.
383 (CE 1121).
One
reason
for
Continuing
to
inscribe
the
names
of
sanctuary
workers
may
have
been
to
preserve
their
entitlements
or
to
detect
freeloaders.
In
many
cases,
sanctuary employees
may
have
worked
under
somewhat
better
conditions
than
other
indentured
people,
inasmuch
as
they
were
exempt
from
corvée
(e.g. Bongert
1959, 19)
and
various
official
taxes,
and
received
some
maintenance.
Aymonier
noted
this
for
the
Cambodian
colonial
period (Aymonier 1900, 102)
and
Turton
for
pre-modern
Thailand
(Turton
1980, 260).
Immunities
also
applied
to
temple
villages
and
their
lands.
Income
from
the
tracts
of
land
landj
from
the
servant
stafts
belonging
to
V.K.A.
Parameśvara
and
V.K.A.
Bhagavatī
are
not
to
come
under
the
authority
of
the
district
(vişaya)
for
purposes
of
exacting
taxes
lor]
employing[them]
on
cancüla [..]. (K.
1141B,
CE 972)
The
decline
in
the
proportion
of
texts
with
non-sanctuary
workers
might
be
viewed im light
of
the
fact
that
immunities
from
taxation
by
officials
applied
to
the
products ofthe
land
but
not
to
the
low
status
producers.
That
is,
these
workers
came
to
beregarded
more
as
commodities.
Recording of
Children
andJuniors
In
the Pre-Angkorian period (prior
to
CE 800), 87%
of
texts
listed
children,
mostly simply
as
'child'
with
its
mother
or
family
but
sometimes
also by
name,
age
range and gender.
This
is generally
a
higher proportion
than
in
the Angkorian period.
In
the 9th century,
about
80%
of
children
were
recorded
with
their
mothers
Figure 4a). Thereafter, the proportion oftexts
with
children
decreased gradually
with time,
while
the proportion oftexts
that
included
the
children
in
totals
reduced
Signiticantly
after
the
10th century.
Further,
while
many
children
were
documented
by
name,ageand gender
in
the
9th century,
this
was
less
common
in
subsequent
centu-
ies
((Figure46).
The
substantial
drop
in
the
naming
of
children
in
the
10th century
reilects
the
genre
ofalarge
number
of
early 10th century
texts
from
Koh
Ker,
where
no
children
were
named.
f
ig.
P.62
Eileen Lustig and Terry Lustig
Aséanie 31
|
2013
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
--------------
------.
1100-99
1000-99
900-99
800-99
Date
CE
Children counted
-
Inscriptions with children
Figure 4a. Decreasing recording and counting ofchildren with time.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
.
....
-
----
-----
-----
---
.
**
****.
***************************
1100-99
1000-99
900-99800-99
Date CE
Genderrecorded
Named with adults Age recorded
Figure 4b. Reducing
numbers
ofchildren
with
name,
age and gender noted.
New Insights
into
'les
Interminables
Listes
Nominatives
d'Esclaves 63
Figure 5a shows
a
marked
decline
in
the
ratio
of
children
per
female
for
rig. 5
P
.
64
non-sanctuary
children
after
the
9th
century,
but
apparently
none
for
sanctuary
children.
This
again points
to
an
overall
decline
in
the
value
placed
on
the
resource
producers of
the
working population,
in
that
nom-sanctuary
children
became
less
important. The
alternative,
that
women
without
children
were
preferred
for
field
work,
seems
unlikely.
For
children
of
women
working
in
sanctuaries,
the
ratio
may
below because
skilled
people,
such
as
dancers
and
musicians,
had
few
accompa-
nying
children.
In
some
instances,
it
was
unclear
whether
kvan (child)
or
ph'van (younger brother,
sister, cousin)
indicated
juveniles
or
simply
a
sibling relationship, and
whether
the
personal
name
inscribed
directly
after
these descriptors belonged
to
the
same
or
a
different
person.
Generally,
it
is
clear
that
kvan
meant
a
dependent
child.
Where
a
total
of
personnel
was
said
to
include
children,
this
had
to
refer
to
the
kvan.
Other
unambiguous readings
were
where
the
child
was
explicitly
that
of
the
preceding
female eg.
'tai
Kandvat and
child
of
tai
Kandvat'
underlined
in
'sī
kanso
tai
kandvat
kvan
tai
kandvat
sī
kanjü
sī
kanthok' (K. 153,
CE
1001);
where
the
punctuation
made
the
relationshipexplicit,
e.g.
"..tai
Kandes and
child.
tai
Kanin...
underlined
in
'tai
kampitkvantaivrata
tai
kañjan
tai
kandes kvan
o
tai
kan'Tn
tai
phem' (K.
221N,
CE
1011);
oFwhere
a
male
kvan
appeared
among
a
group
of
adult
women,
e.g.
'the
child
gho
Paron'
in
'tai pandan kvan gho paron
tai
kanhyan
tai
th yak
tai
kan
in
kvan
tai
chke' (K. 221N).
Nevertheless,
there
remained
instances
of
ambiguity
in
6%
of
inscriptions.
For
example,
in
K.
222 (11th c.)
there
is
'tai
dharmma
kvan
gho gat
ti
vāpśikhāšiva (tai
Dharmaland)
child
gho 'Gat, given by
the
vāpSikhāśiva).
Should
we
read
this
as:
'tai
Dharma
with
child
followed
by gho
Gat'
or
as
'tai
Dharma
with
the
child
gho
Gat'?
Further,ifthesecond reading is
correct,
is gho'Gat
an
infant
or
an
adult?
Asensitivity
test,where the
uncertain
kvan
were
taken
to
be adults,
resulted
in
little
change
to
theratio of
children
per
female
(Figure 5b).
It
might be
noted
in
passing
that
each
of
theambiguous
texts
was
from
the
11th
or
12th
century,
a
further
indication
of
lessening
attention
paid
to
children.
There
were
very
few ph
van
(50
or
0.3%).
While
ph
van
following
adult
males
would not
be
young
children,
it
was
often
unclear
whether
they
were
children,
particularly
when
listed
with
adult
females.
Given
their
relatively
small
number,
it
wasdecided
not
to
count
them
in
the
analyses.
Dictionaries Jenner 2009a; 2009b;
Pou
2004)
define
gvāl
as
an
'animal
herder
based
on
the
modern
Khmer
ghvāl, possibly
related
to
vāl, 'field Jenner)
or
the
Prakrtagovala, linking
it
with
cattle
(Pou).
One
Pre-Angkorian
occurrence
where
herderwould
make
sense
has been
found
in
K.
155
(CE 578-777)
but
the
term
appears
as
a
qualifier
of
tnur
(cattle)
in
K.
689 (CE 578-777) (Vickery 1999, 73-74).
However,
fromtheAngkorian period, gval
appears
not
only
in
field
lists
but
also
within
the
Sanctuary
in
various
roles
(Vickery 1999, 73-4).
The
origin
oflap,
'an
unidentified
classofslave
children
of
both
sexes
Jenner
20095) is
unknown.
However,
the
term
Eileen Lustig and Terry Lustig
Aséanie 31
|
2013
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
----------------
-.
800-99 1000-99
900-99 1100-99
Date
CE
-
Sanctuary Non-sanctuary
Figure
Sa.
Children per female for sanctuary and non-sanctuary personnel. Dashes
on
the non-sanctuary
curve
indicate the relatively small number (one child and five adult females).
0.6
0.5
0.4
---*
2
0.3
0.2
.1
.0
1100-99
800-99 900-99
1000-99
Date
CE
Sanctuary Non-sanctuary
Figure
5.
Curves
where
uncertain
kvan
taken
as
adults.
New Insights
into
les
Interminables
Listes
Nominatives
d'Esclaves' 65
is
also
defined
in
Jenner's dictionary
as:
'to
repeat,
recur,
return',
while
tlāp/tlap
and
tanlap
are
thought
to
mean:
"to practice, exercise,
train,
discipline', suggesting
the
role
of
the
lap category
of
personnel
as
being
akin
to
trainees
or
apprentices.
Pou,
on
the
other
hand,
tentatively suggests
an
association
with
a
modern
meaning
of
lap.
'to
move
stealthily, secretly' (Pou 2004, 415).
In
the
lists
of
personnel,
the
positions
of
gvāl
and
lap
are
broadly
after
male
and
female
adults
respectively
and
before
children.
Vickery
argues
that
lap
could
be
a
category
of
young
adult
(Vickery 1999, 78)
and
gvālperhaps
a
category
of
males
and
females
not
defined
by
work
(Vickery 1999, 74),
while
Soutif
has suggested
that
gval and lap
are
respectively
male
and
female,
and
junior
in
some
way
to
adults
(Soutif 2009, 496).
We
concur
with
Soutif
on
this.
These
two
categories together
constitute
10-15%
of
all
individuals
in
texts
where
they
are
mentioned,
up
to
the
early
11th
century. They have
a
wide geographical
distribution,
and
are
found
as
far
as
the
provinces
of
Lopburi,
Sakhon
Nakhon, and
Prei
Veng. Figure 6
shows
a
declining
trend
in
gval and lap,
as
we
observed
for
children,
although
the
children
don't
disappear entirely,
as
these
do.
Gender
Ratios:
Adults
and
Juniors
Michael
Vickery has
observed
a
gender
imbalance
among
some
lists
of
Pre-Angkorian personnel (Vickery 1998, 259, 306, 3
10-312,
428),
and
Figure
7a
showS fia.7
that
there
are
generally
more
adult
females
than
males
in
the
Angkorian period
as
P.
00
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1100-99
1000-99
900-99
800-99
Date
CE
Figure 6. Declining percentages of inscriptions
with
gval
or
lap.
1.
One
of
the
reviewers
of
this
paper
disagrees
with
Pou's
interpretation,
on
the
grounds
that
'closed syllables
with
the
inherent
vowel,
i.e.
a,
in
ancient
Khmer
usually correspond
to
syllables
with
a
short
vowel
in
Modern
Khmer.
Eileen Lustig and Terry tig
Aséanie 31
|
2013
3000
2500
--.
2000 -**********
1500 ---------
1000 ----------.
500
0
1100-99
1000-99
900-99
800-99
Date
CE
Females
Males
Figure 7a. Numbersofadult males and females with
time.
400
300
200
O
---
100
---
1100-99
800-99 900-99 1000-99
Date CE
gval lap
Figure 7b. Numbers ofgväland lap with
time.
New Insights
into
les Interminables
Listes
Nominatives
d'Esclaves 61
well.
In
contrast
among
the
juniors, gvāl
are
significantly
more
numerous
than
lap
(Figure 7b). When gvāl
and
lap
are
combined
with
adult
males
and
females
(Figure 8a), Fig.s
the gender
ratios
in
the 9th, 10th and 11th
centuries
are
reasonably close
to
1
in
non-sanctuary
texts,
as
might be expected
in
a
natural
village population comprising
all ages.
In
sanctuaries
where
many
roles
are
allocated
to
women,
the
ratio
fluctu-
ates
around
0.6.
Again,
the
12th
century non-sanctuary plotting position may
not
be
representative,
as
few
of
the
texts
seem
to
have field personnel.
p.
68
In
many
cases
it
was
very
hard
to
distinguish
between
sanctuary
and
non-sanctuary
lists.
Altogether, 15
out
of
93,
or
18%
ofthe Angkorian
texts
could
not
be categorized
as
sanctuary
or
non-sanctuary
with
some
certainty. Sensitivity
tests
were
applied
in
order
to
examine
if
this
imprecision might affect the
results.
Doubtful
cases
were
all
re-categorized
first
to
sanctuary (Figure 8b) and
then
to
non-sanctuary
(Figure 8c). Where the
numbers
were
large enough
for
statistical
analysis, the
results
were
not
significantly
different.
Apreliminary
examination
of
the
non-sanctuary
male-female
ratios
in
the
Pre-Angkorian period
indicates
an
average
that is lower (-0.8)
than
for
the
corre-
sponding Angkorian
workforce
(~1). Vickery interprets
the
low
Pre-Angkorian
ratio
as
illustrating production
units
in
the
field
rather
than
communities,
as
is
indicated
above
for
the
corresponding Angkorian
workforce
(Vickery 1998, 310).
A
minor
problem
was
that
the
gender
of
inspectors
and
foremen
(e.g.
khlon
sruk, tamrvac, 'amrah)
was
usually
unstated.
Nevertheless, only
in
four
texts
ofCE
893
(K. 324,
K.
327,
K.
330,
K.
331)
were
inspectors
recorded
as
female,
tamrvac
tai,
and
heading
a
list
of
sanctuary
females.
The
others
were
likely
to
be male, because
of
their
position
in
the
list
ahead
of
males
or
males
and females, and also
as
others
in
the
same
roles generally
were
stated
to
be
male.
As
it
is,
these
supervisory personnel
constituted
only
1.5%
of
the
workforce,
so
a
misallocation
of
their
gender
would
not
affect
the
results
appreciably.
Determination of Ages
The
age
ranges
for
the
different
categories
of
inscription personnel
were
estimated from the proportions
of
non-sanctuary
individuals
of
the 9th century
who
were
adults,juniors and
children
(see
Table
1). The
calculations
used
the
9th century
data
with
the highest proportion ofchildren
together
with
the
age and
sex
distribution of the 1950 population pyramid
for
Cambodia (Figure 9a), the oldest rig, 9
Such
statistics
available
(DESA nd, Cambodia).
In
recognition ofthe possibility
that
the
intant
mortality
rate
in
medieval
times
might have been higher,
we
repeated
the
analysis
with
asensitivity
test
using
a
modified
pyramid,
with
the percentage
of
intants
aged 0-4
years
-
an
indicator
ofinfant mortality- higher
than
the
16.7%
p. 69
taken from the 1950 data.
Eileen Lustig and Terry Lustig
Aséanie31|
2013
1.2
1.0
0.8
--*------
0.6
0.4
0.2
0.0
1100-9
800-99 900-99 1000-9
Date CE
-
Sanctuary Non-sanctuary
1.2-
1.0
0.8
0.6
0.4
0.2
0.0
1000-9
9
-99
900-99
1100-9
Date
CE
-
Sanctuary
Non-sanctuary
1.2
1.0
0.8
0.6
-**
0.4
0.2
0.0
800-9 900-
-99 1100-9
1000-*99
Date
CE
***Sanctuary
Non-sanctuary
Figure 8. Male
to
female
ratios
with
sensitivity
tests:
a)
initial
results; b)
all
doubtful
cases
taken
as
sanctuary:
)
all
doubtfiul
cases
taken
as
non-sanctuary.
69
New Insights into
les
Interminables Listes Nominatives d'Esclaves
80-84
75-/9
70-7
4
65-69
60-64
55-59
50-54
45-49
40-44
35-39
30-34
25-2
9
20-2
15-1
I9
10-1
5-9
0-4
8
12
10
46
2
10 42
12 6
8
Males
Females
Percentage
of
population
80-8
75-7
70-74
65-69
60-64
55-59
50-5
45-49
40-44
35-39
30-3
25-2
9
20-24
15-19
10-14
5-9
0-4 12
10
8
6
2
12 10
4
8
Males
Females
Percentage
of
population
reyPopulation pyramids: a) Cambodia 1950; b) assuming higher infant
mortany.
Eileen Lustig and Terry Lustig
Aséanie
31
|
2013
The
shape
of
a
population pyramid depends
on
wealth,
nutrition,
medical
knowledge,
urban
or
rural
residence,
education
and
social
customs.
Ideally, societies
which
are
similar
to
Angkor's
could
provide
data
for
comparison (see
for
example
Vickery 1998, 257-258).
Ancient
and
medieval
age-sex
distributions
have been
inferred
from
skeletal
remains
worldwide,
but
relying
on
such
data
is
considered
hazardous,
since
the
results
obtained
by palaeodemographic techniques have been
found
to
differ
appreciably
from
distributions
obtained
from
public records
at
the
same
sites
(Bocquet-Appel
and
Masset
1982). Although Angkorian population life
expectancies
cannot
be
known
accurately,
it
is
reasonable
to
expect
that
they
were
no
less
than
what
has
been
found
in
countries
with
the
world's
poorest
condi-
tions
of
health
and
nutrition.
Accordingly,
the
percentage
of
the
population
in
the
age
range
0-4
in
1950
was
examined
for
each
country
listed
on
the
DESA
website,
excluding
those
at
war
at
the
time.
The
largest proportion
was
for
Djibouti, being
just
under
20%
of
the
population (DESA n.d., Djibuti).
From
this,
we
judged that
recalculating
the
age
ranges
using
20%
should
provide
a
reasonable
upper
bound
Fig.9
for
the
sensitivity analysis (Figure 9b).
p.
69
The
results
are
set
out
in
Table
3.
For
both
population pyramids,
the
ages of
the
dependent
children,
pau
and
rat
corresponded
with
what
would
be expected for
nursing
infants
and
toddlers,
while
the
transition
from
gvāl
and
lap
to
adulthood
accorded reasonably
well
with
the
ages
of
puberty.
When
the
proportion
of
inscrip-
rig. 6
tions
that
recorded
gvāl
and
lap decreased
after
the
9th
century
(Figure 6), either
*
0"these
individuals
were
no
longer partofthe workforce,
which
seems
unlikely,
or
the
age
of
transition
from
gvāl
and
lap
into
the
adult
workforce
decreased,
until
these
categories
were
ultimately
subsumed.
This
might be
associated
with
a
reduction
in
the
age
of
transition
between
the
status
of
'child'
and
that
of
full
time
worker,
which
in
turn
might be
linked
to
a
more
general
trend
whereby
elites
were
regarding
the
workforce
increasingly
as
units
of
labor,
which
in
turn
implies
a
tightening of
controls
over
individuals.
TABLE
3.
AGE
RANGES
FOR
CATEGORIES
OF
PERSONNEL
1950 Cambodia
0-2
3-5
High infant mortality
0-2
pau
3-5
rat
gval
lap
si
and
gho
6-14
6-12
6-10 6-9
15+ 13+
tai
11+
10+
Changing Status of Workers
Whereas
personnel
in
sanctuaries
were
organized according
to
their
occupa-
tions, non-sanctuary
workers
were
probably employed along family
and
communal
lines. This might be
inferred
from
the facts
that
the male-female
ratio
was
near
to
1
at
least
for
the period
from
the
9th
to
the 11th centuries, and that the estimated
age ranges of nursing infants, toddlers and (probable) adolescents, albeit for the
9th century,
conformed
to
what
might be expected in
a
natural
population. This
contrasts
with
the
Pre-Angkorian period,
where
lower
male-to-female
ratios
suggest
what may often have been production
units.
The
status
of
the
Angkorian non-sanctuary
communities
themselves
seems
to
have changed, however. The declining numbers, after the 9th century,
of
children,
gval and lap, people's
names
and
even
the adult personnel
-
relative
to
the number
ofinscriptions
point
to
adecrease in
status
of non-sanctuary
workers
from that
of
individual families attached
to
land,
to
components of production for each parcel of land.
Changes
in
Status
of
Adult
Males
In
order
to
learn
something
of
the
differences
in
status
and
work
assignment
between
the
male
categories si/sīand gho,
we
analyzed the
contexts
in
which
these
categories
were
recorded.
In
dictionaries, the
term
ghoda/gho is
'robust
male slaves
(Pou 2004), 'one of
a
category of
male
slaves and
'to
be
fit
for
manual
labour (Jenner
2009a and 20095)
for
both the Pre-Angkorian and Angkorian periods. These
inter-
pretations
are
possible
from
many
contexts.
On the
other
hand,
K.
41 (CE 877) lists
a
tmon ghoda
'male
percussionist and K. 134 (CE 781)
records
gho
who
are
nakk
pamre
kamlun vrah 'persons serving
within
the sanctuary' Jenner, n.d.). The
term
si/siy/sī
is
a
male person, especially
a
'commoner (Pou;Jenner)
or
'slave Jenner).
It
too
is
attested
from
the Pre-Angkorian period,
where
it
is used
for
male
children
and rarely
to
refer
to
adult
males,
as
in
K., 451 (CE 680).
In
the
Pre-Angkorian period,
while
both
terms could signify males
in
general,
individual
males
were
given the appellative
vã: thus
for
example, ghoda: vāKantān; vāSagom; vāSampuh; vāChep (K. 375, 7th c.).
Antelme
notes
that
sīis
still
used
to
strengthen
the
word
ptī'husband'
in
Modern
Khmer, and
that
there
are
cognates of si
in
various
Austronesian
languages, including
those of Bali, the Philippines and indigenous Taiwan, where the appellative
can
be
used for both males and females (Antelme 2001, 253).
From
the Angkorian period, the
terms
are
more
clearly
differentiated.
In
the
Roluos texts, si
tends
to
be in sanctuary
roles
and gho mostly
in
village
and
field
teams, exemplitied
in
K.
809N (CE 878-87),
where
male
inspectors, cooks,
musicians
and dancers
are
sī,whereas
rice
field
hands
living
in
specific villages
are
gho. Vickery
noted
an
exception
to
this 'rule'
in
a
Roluos
text,
K. 328 (CE 893), and thought
that
the
gho
warranted
further
investigation (Vickery 1999, 73).
A
ditference
between the categories
had
already
been
noted
by Aymonier
who, perhaps
not
taking
into
account
many
Roluos
and
Koh
Ker
texts,
thought
that
Eileen Lustig and Terry Lustig
Aséanie31|2013
the
condition
of
the
gho
was
less 'abject'
than
that
of
the
si
and
the
tai
(Aymonier
1904, 547).
He
observed
that
from
the
12th
century,
the
previously high
status
loñ
and
ten
appeared
in
texts
as
'serfs'
and
that
they
too
were
socially superior
to
the si
and
tai.
Bongert
later
questioned Aymonier's
observation
on
the
grounds that there
were
cases
where
si
and
gho
held
positions
of
inspector
or
supervisor
(Bongert 1959,
17).
Si
and
tai
appear
as
having
lower
status
than
loñ
and
ten
in
the
two
texts
where
they
both
appear
(K. 383, CE
1119;
K.
366,
CE
1139),
but
loñ
and
gho
are
not
seen
in
the
one
text.
In
addressing
these
matters,
neither
researcher
took
account
of
the
appellatives
of
individuals
in
relation
to
their
working
functions
or
to
the changes
that
took
place
over
time.
As
elsewhere
in
Southeast
Asia
(Errington 1983, 199;
Ebihara
1984, 284),
status
and
hierarchy
were
important
in
Khmer
society, and emphasized
in
the
Angkorian
period inscriptions (Sahai 1978, 18;
Wolters
1982, 19). Especially
evident
are
the many
levels
and
classes
of
positions,
real
or
symbolic,
the
regalia, and
the
titles
given
to
officials.
Thus
the
status
of
the
recipient might
determine
how
much
merit
is
acquired
(K. 842,
CE
968),
the
amount
paid
for
services
(K. 206, CE 1042)
or
for
the
purchase of
land
(K. 374,
CE
1042) and,
within
a
monastery,
the
maintenance
for
resident
clerics
and
students
(K. 701,
10th
c.),
or
the
level
of
punishments
for
misdemeanours
(K. 323,
CE
889). Hierarchy
is
evident
in
the
order
of
listing people (high
to
low
status
titles;
for
commoners,
supervisor
-
males
-
females
-
children). Vickery has
noted
a
single
hierarchy
of
both
gods
and
human
elites
as
well
as
one
from
elites
down
to
commoners
in
the
Pre-Angkorian period (Vickery 1998,
207
and
271).
For
the
period
and
region
dominated
by Angkor,
we
consider
that
the
place
ment
of
si
ahead
of
gho,
or
gho
ahead
of
si
in
an
ordered
list
is
significant.
We
also
interpret
sanctuary
personnel
as
having higher standing
than
the
non-sanctuary,
since
of
those
texts
recording
workers,
the
proportion
with
sanctuary
staff
increased
with
time
(Figure 3),
while
those
listing non-sanctuary
workers
decreased, implying
Fig. 3
that
sanctuary
staffwere
more
important
to
the
authors
of
texts.
From
the
9th
century,
both
si
and
gho
are
seen
in
sanctuary and non-sanctuary
roles.
Up
to
the
end
ofJayavarman
IV's
reign,
the
majority
of
sanctuary
lists
contain
only si,
while
gho
continue
to
constitute
the
males
in
most
non-sanctuary lists. Several
sruk
lists, apparently
of
field
workers,
from
one
Roluos
text
and
from
Koh
Ker
under
Jayavarman
IV,
are
headed
by
one
or
two
si,
followed
by
many
subordinate
gho. We
argue
that
up
to
the
end
of
Jayavarman
IV's
reign (circa
CE
941),
si
were
of
higher
status
than
gho,
but
that
afterwards
gho
had
greater standing
than
si
(Table 4). Up
to
941,
where
si
and
gho
are
in
the
same
text,
there
are
five
inscriptions
where
only
si
are
found
in
sanctuaries
and
only gho
are
non-sanctuary
workers.2
There
are
no
inscriptions
in
this
period
where
gho
are
in
sanctuaries
and
si
are
not.
Furthermore,
K.
320 (CE 879),
K.
314
(CE 891),
K.
328 (CE 893),
K.
318
(CE 897),
K.
270
(CE 921).
Seven
inscriptions
have
si
and
gho
unsorted
and
have
not
been
counted.
2.
73
NewInsights
into
'les Interminables
Listes
Nominatives
d'Esclaves
fiveinscriptions
have
si
listed
ahead
ofgho',
but
there
are
none
with
gho
listed
before
Si.After941,
there
are
nine
sanctuary
and
three
non-sanctuary inscriptions
with
gho
before si
but
none
with
si
recorded
ahead
of
gho.
Where
si and gho
are
listed
in
the
sameinscription,si
seem
to
have
higher
status
than gho up
to
Jayavarman
IV,
while
after his reign. gho
seem
to
be always
above
si.
TABLE4.
NUMBER
OF
INSCRIPTIONS
CONTAINING
BOTH
SI
AND
GHO,
UP
TO
AND
AFTER
THE
REIGN
OF JAYAVARMAN
IV
Tot a l
with
gho
higher
status
than
si
gho
ahead
si
aheadTotal
with gho
Period
Si
m
in
si higher
status
than
gho
sanctuary of gho sanctuary
and
si
not
in
sanctuary
of
si
and
gho
not
in
Sanctuary
Up
to
end
Jayavarman
IV
5
5
After
12
12
0
0
Jayavarman
IV
Might
the
prevalence
of
si
or
of
gho
be
a
regional phenomenon
rather
than
a
general
trend?
While
no
regional
concentrations
are
evident,
this
could
be
because
there
are
relatively
few
instances.
Dominant
si
and
gho
are
generally
found
in
areas
ofthat developed during
the
Angkorian period,
north
of
the
Tonle
Sap
in
present
day Cambodia.
The
si
are
at
Angkor
and
Koh
Ker,
while
the
gho
are
distributed
more
widely.
Another
possibility is
that
the
perceived
shift
from
dominant
si
to
dominant
gho
is
associated
with
the
change
from
inscriptions being mainly royal
to
being
mainly non-royal
after
the
reign ofJayavarman
IV,
when
Rājendravarman
moved
the
capitalback
from
Koh
Ker
to
Angkor.
We
find
this
unlikely,
however.
While
there
are
hardly
any
royal
texts
with
lists
of
personnel afterJayavarman
IV,
the
male
personnel
given by sūryavarman I
to
an
ashram
are
gho
in
K.
290 (CE 1015). Likewise,
six
gho
arelisted ahead
of
four
si
in
a
royal
donation
to
a
goddess
in
K.
353S (10th c.)
under
Jayavarman V
or
Rajendravarman.
3. K.313 (CE 879), K.
677
(10th c.),
K.
681
(10th c.),
K.
182
(10oth c.),
K.
183 (10th c.).
K.
52 (CE 918)
Was
not
counted, being
a
largely illegible fragment,
with
gho
and
tai
precedingasi, suggesting
a
child.
4.
K.1151 (CE 942),
K.
350
(CE 944-68),
K.
239 (CE 966),
K.
240
(CE 968-77), K. 343 (CE 974),
K.
214
(CE 982),
K.
263 (CE 984),
K.
352
(10th c.),
K.
353 (10th c.),
K.
1198
(CE 1015),
K.
258 (CE 1113),
K.
241 (11th c.).
Aséanie 31
|
2013 Eileen Lustig and Terry Lustig
A
Hypothesis:
Factions
and Shift
of
the
Centre
to
Koh
Ker
What is
of
some
note,
however, is
that
after
Jayavarman
IV,
the influence
of
Fig,2
certain
officials appears
to
have
grown,
as
suggested by Figure 2,
as
these officials
*produced
most
ofthe inscriptions
in
the
next
century and
a
half.
Was
it
more
than
coincidental that
this
rise
and that
of
the gho
took
place
over
much the
same
period?
We
hypothesize that during the 9th century, possibly resulting from
territorial
amalgamation and the need
to
organize
labour
in
the early Angkorian period, si
were
accorded higher
status
than
gho.
In
the early 10th century, opposition by
some
elites,
perhaps
to
the royal
succession
by Yaśovarman's
sons,
was
mounting
at
Angkor and
was
supported by si. Indeed,
we
might expect
some
reaction, as kingship normally
passed among members of
one
generation before moving
to
the
next
(Vickery 1986,
108).
Between
921
and 928, Jayavarman IV,
a
contender
for
the
throne
and ruling
at
Koh Ker, expanded his power base by allying
with
these
two
opposition
groups.
By
928, he had become strong enough
to
become 'king
of
Khmer kings' (Jacques 1985,
278). Tra nsfe rri ng the capital
to
his home base served
to
reduce the influence
of
those formerly supporting
Yaśovarman's
sons.
In
944, Rājendravarman
took
power
and
returned
the capital
to
Angkor, supported by
some
officials who had been
gaining
in
power
and wealth, and perhaps by gho, who
now
had
a
higher
status.
The
gho
now
assumed roles previously
held
mostly by si and
were
henceforth
not
seen
listed
behind them
The political geography leading
to
Koh
Ker
becoming the
centre
has long
puzzled Angkorian
researchers.
Agricultural and
mineral
resources
may have been
relevant
to
its
location
(Coe 2004, 108; Jacques 2004, 156). Hendrickson
notes
that
Koh
Ker
is close
to
a
major road
to
the
copper
and gold
resources
near
the well-
travelled
pilgrimage
site
of
Wat
Phu (Hendrickson 2007, 245).
He
suggests
that
Koh
Ker
may
have been placed
to
cut
Wat
Phu
off
from
Angkor.
Evans
has demonstrated
that
Koh
Ker
had adequate
water
resources
and productive
rice
fields sufficient
for
a
large population, and has noted
that
local quarries could produce sandstone
of
high
quality
for
monumental
construction
(Evans 2010-2011, 101-115).
Evans
argues
that
the Koh
Ker
region
was
settled well before the 10th century,
and
that
rulers
based
at
Angkor
took
an
interest
in
the
site
over
a
period both before
and after its
time
as
the
centre.
The
ceramics
found
at
Koh
Ker
indicate
intensive
occupation
from
the 9th
to
the 13th century. Included
are
Chinese
ceramics
which
date from the 11th
to
the 19th century (Evans 2010-2011, 116).
An
inscription,
K.
72
from
Prei
Veng, dated
to
Harśavarman I(CE 912-23), is what appears
to
be
a
royal
directive
to
grant land
at
Chok Gagyar (Koh Ker), while
K.
674 (CE 966) is
an
edict
from Rājendravarman concerning sanctuary lands
at
Prasat
Dan, Koh
Ker.
Later
texts
K.
2575 (CE 979),
K.
814 (CE 979) and
K.
31 (CE 1019)-
mention
officials atChok
Gargyar, demonstrating continuing royal
interest
in
the
area.
15
New
Insights
into
"les Interminables
Listes
Nominatives d'Esclaves'
Yet
it
seems
unlikely
that
anticipated
economic
benefits
were
a
strong
motiva-
tion
for
the
shift
to
Koh
Ker.
First, capital
cities
are
rarely moved mainly
on
economic
grounds,
because
of
the
large
costs
entailed
(Schatz 2004, 116). Second,
in
several
respects, Angkor
was
better
placed strategically.
Its
greatest
asset
was
its
location
on
the
Tonle
Sap
which
floods annually, increasing
from
about 2,700 ha
to
more
than
10,000 ha between May and
November
(Delvert 1961, 80), the flooded forests providing
a
habitat
for
the
spawning
of
many
fish.
The
area
inundated
receives
rich
deposits
of
silt, and
the
receding
waters
can
be
retained
in
artificial
ponds
for
planting subsequent
rice
crops.
The
area
around
the
lake
is relatively flat,
with
hills
projecting
from
the
plain providing
excellent
lookouts.
The
location
is close
to
the
sandstone quarries
of
the
Kulen
Hills
and good
sources
of
laterite.
Access
to
the north, and
routes
to
the
east
and
west,
may
also have
influenced
the choice
of
site
(Mabbett and
Chandler
1996, 87; Hendrickson 2007, 91-92). The fact
that
the
centre
of
power
moved
away
from
Koh
Ker
after
a
couple
of
decades, whereas Angkor
was
the
centre
for
some
500
years,
suggests
that
the
latter
was
more
sustainable
as
a
capital.
Capitals
are
normally relocated
to
achieve several objectives (Joffe 1998, 552;
Wolfel
2002; Schatz 2004, 117).
Two
common
motives
are
to
undermine
or
control
rival
power
bases
while
strengthening personal authority, and
to
consolidate
the loyalty
of
the
population (Joffe 1998, 549; Schatz 2004, 119). Concerning the
first
objective,
we
argue
that
the
move
to
Koh
Ker
was
meant
to
undermine
the
influence
of
a
faction,
perhaps associated
with
supporters
of
Harśavarman
I
and Iśānavarman I1, the
sons
of
Yaśovarman,
who succeeded
him
ahead ofJayavarman
IV.
Indeed,
one
might
infer
from
the
inscriptions
that
there
was
some
dissent
during the period between
the
reigns
of
Yaśovarman
and Rājendravarman.
Initially, Coedès tentatively proposed
that
Jayavarman
IV
could
have already
been reigning
at
Koh
Ker
as
a
usurper
while
Harśavarman
Iand Iśānavarman
II
were
still
alive, assuming
that
succession
was
through primogeniture. Coedès based
this
assumption
on
Sdok Kak
Thom
inscription
K.
235 (CE 1052),
which
refers
to
the enigmatic kamraten jagat
ta
raja 'High
Lord
of
the World, protector
of
the king'
Jacques 1985, 278),
who
was
said
to
have
remained
with
all
the
various
rulers
from
Jayavarman
II,
wherever
their
seat
of
power
was
located. Coedès had initially
inter-
preted
the
similar
term
kamratenjagat
ta
räjya High Lord
of
the
World, protector
of
the kingdom' (Jacques 1985, 278),
a
deity appearing
in
Jayavarman's inscriptions
at
Prasat
Thom,
as
being
the
same
as
the kamrateńjagat
ta
rāja (Coedès 1931, 16),
leading
him
to
this
provisional suggestion.
However
if,
as
Jacques
argues,
Caedès
had
conflated
two
different
gods (Jacques
1971,
169; and 1985, 277-8), Jayavarman
was
not
a
usurper
in
921.
As
it
is,
the
veracity
of
much of the Sdok Kak
Thom
text
is
considered
questionable (e.g. Ccedès 1970; Vickery 1985, 232-233), casting
serious
doubt
on
the
historical
role
of
the kamratehjagat
ta
rāja.
5.
K.
184 (CE 921),
K.
682 (CE 921),
K.
189 (CE 928),
K.
188
(CE 92),
K.
186 (CE 930).
Aséanie
31
|2013 Eileen Lustig and Terry Lustig
Some
events
may
well
have been
omitted
from
the
historical
record
or
invented
in
order
to
legitimize the
status
and
actions
of
the
four
kings between the reigns of
Yaśovarman and Rājendravarman.
None
of
these kings have yet been found
to
produce
much
in
the
way
ofgenealogies,
as
was
the
norm
previously and subsequently. Apart
from
one
eulogy
from
a
non-royal inscription
at
Koh
Ker
(K. 675, CE 928-41), there
are
no
known
mentions
of
the immediate predecessors ofJayavarman
IV
during his reign,
particularly by Jayavarman himself,
even
though the reigns
of
Harśavarman Iand
Tsānavarman
II
lasted
from
around CE 910-2
to
the period CE 925-8, between 13 and 18
years.
This might be understandable
if
Jayavarman had
a
higher
claim
to
the throne
than
Yaśovarman's
sons,
as
they would then have been
usurpers
(Vickery 1986, 108).
Likewise, Rājendravarman
may
have had
a
stronger claim
to
legitimacy than
Jayavarman's
son
Harśavarman
II
(Vickery 1986, 108). His stronger claim could be the
reason
why Jayavarman
IV
and his
son
were
keen
to
demonstrate
a
close relationship
with
their
successor.
Rājendravarman is honoured
with
the
very
high
status
title
of
dhülijen kamsten 'añ
in
inscriptions
K.
957 (CE 941),
K.
999 (10th c.) and
K.
1151
(CE 942).
Jayavarman
erects
a
linga
in
honour of 'his
brother'
Rājendravarman (K. 677S,
10th
c.). Rājendravarman describes
himself
as
the
older
brother
of
Harśavarman
II
in
K.
286 (CE 947),
K.
806 (CE 961),
K.
522 (CE 944-66),
K.
686 (CE 944-66) and
reinstates
the
tradition
of
referring
to
all
his
predecessors beginning
with
Jayavarman
II,
suggesting
a
desire
to
distance
himself
from
any
instability by depicting
a
long
unbroken
pedigree, and leaving
out
anomalous
events
such
as
might explain the
shift
of
the
capital
to
and from Koh
Ker.
The
unease
hinted
at
by the apparent
omissions
and
inventions
in
the
texts
is also perceived by Thomas Maxwell
in
what he describes
as
'rivalrous
dialogue,
conducted
in
terms
of
religious symbolism' between Angkor and Koh
Ker
to
gain
legitimacy and political
power
(Maxwell 2010, 2). Principal shrines
at
the non-royal
Prasat
Kravan
at
Angkor
were
dedicated
to
the deities
Tribhuvanasvāmi
(K. 269, CE 921)
and Trailokyanātha (K. 270, CE 921)- both formsofVişnu, both meaning 'Lordofthe
Three
Worlds'.
Maxwell
considers
it
is unlikely
to
be
coincidental
that
at
almost the
same
time,
Jayavarman consecrated the
main
god
in
Koh Ker's
Prasat
Thom
as
Siva
under
the
name
Tribhuvaneśvara
(also meaning Lord
of
the Three Worlds)
-
and
perhaps
more
pointedly
soon
after, erected
a
Vişnu Tribhuvanaikanātha, meaning
Only Lord
of
the Three Worlds',
in
Kompong Cham (K. 99, CE 922).
If
the political
tensions
intimated
here
were
more
than supposition, moving the
centre
of
power
might have appeared
a
viable strategy
for
undercutting the opposi-
tion.
Jayavarman,
who
needed support
from
powerful
interest
groups,
particularly
given
the
great financial, logistical and political
costs
required,
may
have found
common
cause
with
elites unhappy
with
the
status
quo
and
with
some
si.
The
Koh
Ker
texts,
mainly village lists,
maintain
the superior
status
of
si.
The
subsequent
reversed
positions
of
the
si
and gho
at
the end
of
the Koh
Ker
interlude
might reflect
a
change
in
the
social order.
New
Insights
into
les Interminables Listes
Nominatives
d'Esclaves'
Another
impetus
for
the
relocation
to
Koh
Ker
could bethatJayavarman, having
been established
there
as
local
ruler
fromat
least CE 921,
was
seeking
to
empower
his
followers.
We
do
not
know
the
extent
of
his
control
in
921, because
the
locations
of
many
of
the
sruk
and pramāntoponyms,
referred
to
in
his
texts
of
this
date (K. 184,
K.
682)
cannot
be
ascertained.
By CE 928 however,
the
texts
(K. 183,
K.
186,
K.
187)
refer
to
the
distant
provinces
of
Bhimapura
and
Amoghapura
in
northwest
Cambodia
(e.g.
Wolters
1974). A
text
from
Take o
(K. 35) acknowledges Jayavarman
as
king
of
the
Kambujas
in
CE 928 and
another
in
932
from
Kompong Cham (K. 99)
records
an
order
he
issued.
Nevertheless,
regardless
of
when
Jayavarman
became
king,
it
is
safe
to
assume
that
making
Koh
Ker
the
first
city
of
his
domain
would
have
been
a
move
to
shore
up
his
support,
The
other
important
motive
for
the
move
to
Koh
Ker,
to
establish
a
new
allegiance
in
the
population,
can
be
discerned
readily. The
monuments
of
Koh
Ker,
inevitably
symbols of kingship,
are
well
known.
Yet
Koh
Ker
as
capital
appears
to
have
been
efficacious only
in
the
short
term
in
undermining dissenting
elites
and
forestalling
changes,
which
were
later
manifested
in
alterations
in
status
of
the
si
and
the
gho.
The
names
and
titles
ofthousands
of
commoners
recorded
on
stone
in
temple
precincts have been regarded
until
recently
as
uninteresting and incapable
of
providing
insights
into
the
lives
of
these
commoners.
This
study
indicates
that
numerical
analyses
of
data
on
personnel
can
enhance
understanding
of
their
circumstances.
The continuing
near
equal proportions
of
males and
females
amongst
the
non-sanctuary personnel,
and
the
fact
that
the
9th century age
ranges
of
minors
conform
to
what
might be expected
for
natural
communities,
indicate
that
these
people
were
whole
families
or
communities.
By
contrast,
there
are
more
females
than
males
in
sanctuary
lists,
conceivably
because
some
roles
are
gender specitic.
The
reduced
recording
of
non-sanctuary
workers,
non-sanctuary
children
and
of
juveniles
after
the
Roluos
period suggests
a
greater
focus
on
sanctuary operations,
with
non-sanctuary operations being
taken
more
for
granted, indicating
that
labour
was
easier
to
procure,
and
workers
regarded increasingly
as
factors
of agricultural
output. This is
comsistent
with
the increasing
number
of
temple complexes
estab-
lished from the 10th century,
as
well
as
the
supporting
infrastructure
and industry,
all suggesting growing local populations.
Even
though the proportions
of
gvāl
and
lap declined, they
were
likely counted
as
part
of
the
adult
workforce.
The
shift
to
Koh
Ker
cannot
be readily explained
as
driven
by anticipated
economic
benefits.
The
omissions
and
inventions
in
the
royal eulogies
between
the
Eileen Lustig and Terry Lustig
Aséanie
31 |2013
reigns ofYaśovarman and Rajendravarman indicate disquiet,
if
not
discord, between
opposing factions. The
move
of the
centre
from Angkor
to
Koh
Ker
might be best
understood
as
a
strategy by Jayavarman
IV
to
weaken
an
opposing
group
and
bolster
his
power
base, allied with si. Following the
return
from Koh
Ker
to
Angkor,
a
change
in
the socio-political
power
structure, perhaps
seen
in
the rising influence of
many
officials and gho
commoners,
becomes apparent. Available inscriptions from the
first
half
of the 10th century make
no
explicit reference
to
unstable political conditions
during that period, but anyone
familiar
with the Khmer
texts
could
not
expect
to
find them.
Acknowledgments
We
first
wish
to
acknowledge the generous assistance of the late Phillip Jenner.
Our
thanks also go
to
Doug Cooper and the SEALang Old Khmer Project,
Kate
Domett,
Roland Fletcher, Thomas Maxwell, Christophe Pottier, Dominique Soutif, and
two
reviewers
who provided insightful
comments-
one
ofthem may
wish
to
lay claim
to
a
statement
summarising
one
of
our
points.
New Insights
into
les
Interminables
Listes
Nominatives d'Esclaves'
Bibliography
ANTELME, Michel
2001-
AStudy ofNaming Systems from
Ancient
to
Modern Cambodia,
Thesis
(PhD), SOAS,
London.
AYMONIER,
Etienne
1900 Le Cambodge. 1.
Le
royaume
actuel,
Paris,
Ernest
Leroux,
478
p.
1904
Le
Cambodge. 3. Le
groupe
d'Angkor
et
l'histoire,
Paris,
Ernest
Leroux,
818
p.
BocQUET-APPEL, Jean-Pierre and MASSET, Cclaude
1982-
Farewell
to
Paleodemography', Journal of Human Evolution,
11,
pp.
321-333.
BONGERT,
Yvonne
Note
sur
l'esclavage
en
droit
khmer', Etudes d'histoire du droit privé offertes à
Pierre
Petot,
Paris,
Montchrestien,
pp.
7-26.
1959
CHAKRAVARTI,
Adhir
1980-The
Sdok Kak Thom Inscription.
I.
Astudy in Indo-Khmèr Civilization,
Calcutta,
Sankrit
College, 357 p.
CHHOM
Kunthea
2011-InscriptionsofKoh
Ker
1, Budapest,
The
Hungarian
Southeast
Asian
Institute,
335
p.
2004 Angkor and the
Khmer
Civilization, London, Thames &Hudson, 240 p.
1931-
'Etudes cambodgiennes.
XXV,
Deux
inscriptions
sanskrites
du
Fou-nan.
XXVI,
COE,
Michael
D.
CEDES, Georges
La date de Kôh
Ker.
XXVII,
La
date du Bàphûon', Bulletin de l'Ecole française
dExtrême-Orient,
31,
pp.
1-23.
Paris,
Ecole
française
d'Extrême-Orient,
323
p.
Paris,
Ecole
française
d'Extrême-Orient,
337
p.
H.B.
Sarkar
(ed.),
R.C.
Majumdar Felicitation Volume,
Calcutta,
Firma
K.L.
1937-
Inscriptions du Cambodge. Collection de
textes
et
documents
sur
l'Indochine,
vol.
I,
1954- Inscriptions du Cambodge.
Collection
de
textes
et
documents
sur
l'Indochine,
vol.
VI,
1970-Le
véritable
fondateur
du
culte
de la royauté
divine
au
Cambodge',
Mukhopadhyay,
282
p.
DELVERT,Jean
1g6lLe
Paysan cambodgien,
Paris,
Mouton,
740
p.
DESA
ha,World
Population Ageing 1950-2050: Cambodia,
New
York ,
United
Nations
Department
of
Economic
and
Social
Aftairs,
https://www.un.org/esa/
population/publications/worldageing 19502050/countriesorareas. htm
accessed 10.5.12].
EBIHARA, May
04
Societal
Organisation
in
Sixteenth
and
Seventeenth
Century
Cambodia,
Journal
of Southeast
Asian
Studies 15, May 1884, pp.
280-295.
Eileen Lustig and Terry Lusti
Aséanie 31
|
2013
ERRINGTON, S. erarchies:
1983
The
Place of Regalia
in
Luwu',
L.
Gesick (ed.), Centers, Symbols,
and
t:.
niversity
Essays
on
the Classical States ofSoutheast Asia, New Haven,
Yale
Southeast Asia Studies, pp. 194-241.
EVANS, Damian
2010-2011-
The
Archaeological Landscape ofr
Koh
Ker,
Northwest
Cambodia', Bsllas
lletin
de l'Ecole française d'Extrême-Orient, 97-98,
pp.
91-150.
FOREST,
Alain
1998 Esclavage
et
société dans le Cambodge duxıx*siècle', Georges Condomina.
(dir.), Formes extrêmes de dépendance. Contributions àl'étude de l'esclavage
er
du Sud-Est,
Paris,
Éditions de l'Ecole des
hautes
études
en
sciencessociales,
Pp. 343-356.
Asie
HALL,
Kenneth
R.
1985
Maritime
Trade and State Development in EarlySoutheast Asia, Honolulu,
University
of
Hawaii' Press, 384
p.
HENDRICKSON,
Mitch
2007
Arteries
ofEmpire.
An
Operational Study of Transpo rt and Communication in
Angkorian Southeast
Asia
(9th to 15th centuries CE), PhD Thesis, Sydney,
University
of
Sydney,
312
p.
(text)
and
96
p.
(figures).
JACOB, Judith
M.
1979
Pre-Angkor
Cambodia:
Evidence
from
the
Inscriptionsin Khmer concerning
the
Common
People and
their
Environment',
R.B.
Smith
and
W.
Watson
(eds),
Early South East
Asia:
essays
in
archaeology, history
and
historical
geography,
New
York,
Oxford
University
Press,
Pp.
406-426.
JACQUES, Claude
1971
'Sur
les
données
chronologiques de la
stèle
de
Tûol
Ta
Pec
(K. 834)', Bulletinde
1976-Ā
propos
de l'esclavage
dans
l'ancien
Cambodge',
P.-B.
Lafont
(ed.),Actes
1985The
KamratehJagat
in
Ancient
Cambodia',
N.
Karashima
(ed.), Indus Valley
I'Ecole française d'Extrême-Orient, 58,
pp.
163-176.
du
29
Congrès
international
des Orientalistes.
Asie
du Sud-Est continentale,
Paris,
L'Asiathèque,
1,
pp.
71-76.
to
Mekong Delta Explorations
in
Epigraphy, Madras,
New
Era
Publications,
Pp. 269-286.
P.
Lafond
(eds), L'Empire khmer: cités
et
sanctuaires,
v'-xiur
siècle,
Paris,
Fayard,
2004-"TheMasters of
Iron
and
Fire:
Preah
Khan
of
Kompong Svay', C.Jacques and
Pp.
177-198.
JENNER, Philip
N.
2009a-A DictionaryofPre-Angkorian Khmer, Canberra,
Pacific
Linguistics, Research
School
of
Pacific
and
Asian
Studies,
Australian
National University,XXVI
+620 p.
2009b-
ADictionaryofAngkorian Khmer, Canberra,
Pacific
Linguistics, Research scro
hool
of Pacitic and Asian Studies,
Australian
National
University,
xxii
+
778 P:
n.d.
-Unpublished
translations
ofOld
Khmer
language inscriptions.
JOFFE, Alexander
H.
1998-
Disembedded Capitals
in
Western
Asian
Perspective"', Society for Comparat
ative
Studies
in
Society and History, 40 (3),
pp.
549-580.
New Insights
into
Tes
Iiterminables
Listes Nominatives d'Esclaves' 81
KHIN
SOk
1998 Lesclavage
au
Cambodge
a
l
epoque moyenne à
travers
les
codes,
les
inscriptions
modernes
d'Angkor
et
les
chroniques royales', Georges
Condominas
(di.),
Formes
extrémes
de dépendance. Contributions àl'étude de
l'esclavage
en
Asie
du
Sud-Est,
Paris,
Editions
de
l'Ecole
des
hautes
études
en
sciences
sOCiales,
pp.
315-341.
LUSTIG,
Eileen
2009 Power
and
Pragmatism
in
the
Political
Economy of Angkor, Ph.D Thesis, Sydney,
University of Sydney, 281
p.
(Text)
and
63
p.
(Appendices).
MABBETT,
lan
W.
1983-SomeRemarks
on
the
Present State ofKnowledge about Slavery
in
Angkor'
A.
Reid
(ed.). Slavery, Bondage and Dependency
in
Southeast Asia,
St
Lucia,
Queensland
/
London/
New
York , University ofQueensland
Press,
pp. 44-63.
MABBETT,
lan
W.
and
CHANDLER,
David
P.
1996- The Khmers, Cambridge, Mass., Blackwell, 289 p.
MARTIN,
Marie
A.
1998Lesclavage
au
Cambodge
aux
temps
anciens
d'aprèslessources épigraphiques
et
bibliographiques', Georges
Condominas
(dir.),
Formes
extrêmes
de
dependance. Contributions à
l'étude
de l'esclavage
en
Asie du Sud-Est,
Paris,
Editions
de
l'Ecole
des
hautes
études
en
sciences
sociales,
Pp.
285-3
14.
MAXWELL,
Thomas
S.
2010
Angkor
Inscriptions Survey and Documentation,
The
A.I.S.
Annual
Report
for
2010-2011:
1,
Prasat
Kravan
Central
Temple, K.270|1|
and
Iconography,
Research
Report
to
the
19th
Session
of
the
UNESCO
International
Coordinating
Committee
for
the Safeguarding and Development
of
the
Historic
Site
of
Angkor,
Siem
Reap. Working draft, unpublished.
POU,
Saveros
2004-
Dictionnaire
vieux
khmer-français-anglais.
An
Old Khmer-French-English Dictionary.
Vacanānukram
khmaer cās-pārāmn anles,
L'Harmattan,
Paris,
iv
+
XXV
+
732
p.
1st
ed.
Paris,
Cedoreck, 1992].
REID, Anthony
1988 Southeast
Asia
in
the Age ofCommerce, 1450-1680, vol. 1, The lands below the
winds,
New
Haven,
Yale
University Press, 272 p.
SAHAI, S.
1977-Piscal
Administration
in
Ancient
Cambodia, The South
East
Asian Review, 1(2),
Pp.
123-137.
1978'Central
Administration
in
Ancient
Cambodia', The South East
Asian
Review
3(1),
Pp. 15-40.
SANDERSON,
Alexis
2003 TheSaiva Religion
among
the
Khmers', Bulletin de l'Ecole firançaise d'Extrêëme-
Orient,
90-91,
pp.
349-462.
SCHATZ,
Edward
2004
What
Capital
Cities
Say about
State
and
Nation
Building, Nationalism and
Ethnic
Politics, 9:4,
pp.
111-140.
Eileen Lustig
and
Terry Lustig
Aséanie 31|
2013
SoUTIF, Dominique siècle, PhD Thesis
2009 Organisation religieuse
et
profane du temple khmer du
vir
au
xi 51S,
Université Sorbonne Nouvelle, Paris, 3
vol.,
390
p.,
634
p.
and
cLxVI
p.
TURTON, Andrew
1980
Thai
Institutions
of
Slavery', Asian
and
African Systems ofSlavery,J.L. Watson
(ed.), London, University
of
London, School
of
Oriental
and
African
Studies,
PP. 251-292.
VICKERY, Michael
1985-The
ReignofSüryavarman
I
and Royal Factionalism
at
Angkor,Journal of
1986 "Some
Remarks
on
Early
State
Formation in Cambodia',
D.G.
Halland
A.C.
1998
Society,
Economics, and Politics
in
Pre-Angkor Cambodia: the 7th-8th Centuries,
1999
"The
Khmer
Inscriptions ofRoluos (Preah
Ko
and
Lolei):
Documents
from
2005
Cambodia
and
its
Neighbours
in
the 15th Century, Working Paper
Series
No.
27,
Southeast Asian Studies, 16 (2), Pp. 226-244.
Milner
(eds), Southeast
Asia
in the 9th
to
14th Centuries, Singapore,
Institute
of
Southeast
Asian
Studies,
pp.
95-115.
Toky o,
Centre
for
East
Asian
Cultural
Studies
for
UNESCO,
The
Tokyo Bunko,
486
p.
a
Tra nsi tio n
Period
in
Cambodian History'., Seksa Khmer, Nouvelle Série, 1,
Pp. 48-87.
Singapore,
Asia
Research
Institute,
National
University
of
Singapore,
71
p.
WOLFEL,
Richard
L.
2010
North
to
Astana:
Nationalistic
Motives
for
the
Movement
ofthe Kazakh(stani)
capital, Nationalities Papers: The Journal ofNationalism
and
Ethnicity,
3o
(3),
Pp.
485-506.
WOLTERS,
Oliver
W.
1974-
"North-Western Cambodia
in
the Seventh Century',
Bulletin
ofthe School of
Oriental and African Studies, University of London, 37,
pp.
355-384.
1982
History,
Culture, and Region
in
Southeast Asian Perspectives, Singapore,
Institute
ot
Southeast
Asian
Studies, 275
p.
ZHOU Daguan
2007
A
RecordofCambodia, the Land and its People,
Peter
Harris
(tr.), Chiang Mai,
Silkworm Books, 184
p.
Authors
Eileen Lustig: Affiliate
at
the University ofSydney
Terry Lustig: Affiliate
at
the University ofSydney
New
Insights
into
les Interminables
Listes
Nominatives
d'Esclaves" 83
Annex
I.
Commoners
and
their
roles
in
the
Angkorian
texts
The following cxamples have been chosen
to
illust
rate
the ranges ofoccupation and
status
ofcommoners in Angkorian society. Asquare symbol (n) marksa lacuna.
A
tract
of land
in
the settlement
of
Jeń
was
given [him] by the kamsten
of
Lamvān
for
vaudi [and] cuspidor(s) together [weighing)
seven
jyani,
borrowed
from VK.A.
at
Vrac
when [he] inaugurated
a
holy dam. Khnum
on
it:
si
Kan'ā,warder, [with]
seven
males, eight females, [and] 1gval. (K. 420 [12th c.J)
Gho
Krau,
hereditary hand ('nak mrtakadhana) belonging
to
the lord SrīGunavijaya
of the sruk of
anoo.
(K. 291N [910])
Field hands (nak sre)
for
the village (sruk) of Pānī:foreman Nagadhara: gho 'Nit; gho
Smau: gho Kansāt: ghoSrīgho 'Amdah: gho Kamvik; gvālcke; gvālKampas;
tai
Kamvrau
and) male
rat,
female
rat,
male pau,
female
pau;
tai
Kañjā
|.].
(K. 809N [878-87])
Servants
(nak pamre)
for
the fortnight of the waning
moon:
the headman
of
the sruk
ofKanjā;storekeepers,
sī
Tankū,
sī
Damluh; inspector,
sī
Kanlon; cooks,
sī
Kañjai,
tai
Kandai,
tai
'Anrok; grinders [of aromatics],
tai
Kamprvat,
tai
Ras; gleaner,
sī
Kanlap.
Percussionist,
sī
Ghoda; dancers,
tai
Kapkep,
tai
Kansū,
tai
Kanak; singers,
tai
Pandan,
tai
'Gat [and) pau
[.]
(K. 809N [878-87])
Two
vaudi; 1cuspidor; 4
footed
trays;
[..]
Three
water
buffaloes. Slaves ofthe divinity
(khnum vrah):
si
Tanker;
si
Slācra;
si
Prakat;
si
Samap;si Th'yak;
si
Kamvrau;
tai'Gat;
tai
Kampit;
tai
Kana,
tai
Kampin.
All
of
these
are
the
property and khñum
of
the
divinity. (K.
742
[994])
The price
of
tai
Kantem
was
1
yau
ofjñasira, 2
yau
of
amval, 1thlvan
of
milled
rice,
which
gave [her|
title
to
tai
Kantem
until
she died, and
to
her
children
and grand-
children.
(K. 158 [1003])
HisMajesty vrah pada kamraten
añ
šrī
Udayādityavarmadeva
gave
[him]
the
village
(sruk)named
Gnan
Cranān
Vo
with
its residents and
151
transportees and
its
land
revenues, land) gave [them]
as
a
royal offering
to
the said holy linga
in
Bhadraniketana.
(K. 235 1052|)
Servants bound by law ?(nak
pamre
āgama):
Dancers:
tai
Ranga;
tai
Bhā[and] female
suckling;taiKanhey;
tai
Ajya;
tai
Kanuin;
tai
Sresthā;
tai
Kanhyan;
tai
Thkep [and]
tai
van
Kantim
[and]
tai
lvanJen
Kres
o
(K.
270s
[921])
