ACTA
LINGUISTICA
LITHUANICA
XLVII
(2002),
41-54
The
Indo-Eu opean
geni i e
o
agen
wi h
ini e
e bs
and
pa iciples
*
WILLIAM
R.
SCHMALSTIEG
The
Pennsyl ania
S a e
Uni e si y
Examples
o
he
use
o
he
ini e
e b
and
he
geni i e
wi h
agen
unc ion
a e
gi en
om
Old
Indic,
Old
Pe sian,
Tocha ian,
G eek
and
Sla ic.
I is
sugges ed
u he mo e
ha
he
geni i e
objec
o
ce ain
e bal
classes,
e.g.,
Indo-Eu opean
e bs
wi h
he
su ix
*-€-
may
o iginally
ha e
unc-
ioned
as
an
agen
and
ha
he
e ymological
subjec
was
he
pa ien .
In
his
pape
I
should
like
o
gi e
some
examples
o
he
geni i e
o
agen
(o
pe haps
be e ,
geni i e
o
sou ce)
wi h
ini e
e bs
in
some
Indo-Eu opean
languages
o he
han
Bal ic'.
The
e ymological
possessi e
in e p e a ion
is
always
possible
o
passi e
pa iciple
cons uc ions,
bu
i
would
seem
o
me
o
be
unnecessa y
o
ha e
one
ex-
plana ion
o
he
ini e
e b
cons uc ions
and
a
second
explana ion
o
he
passi e
pa iciple
cons uc ions,
al hough
I
suppose
ha
in
p inciple
he
cons uc ions
wi h
ini e
e bs
could
be
explained
as
being
analogical
o
hose
wi h
he
pa iciples.
Fi s
o
all
I
would
quo e
he
example
o
he
use
o
he
geni i e
o
agen
wi h
a
ini e
e b
om
Old
Pe sian
(Ca dona
1970:
2;
Ken
1953:
133):
(1)
a aiy
ii ji a
a ika
aha
NOM.
PL.
MASC.
NOM.
PL.
NOM.
PL.
MASC.
3
PL.
IMPERF.
he
Elami es
un ai h ul
we e
u- asam
Au amazda
naiy
_ayadiya
CONJ.-GEN.
PL.
MASC.
NOM.SG.MASC.
NEG.
3
SG.
PSV.
IMPERF.
and
by
hem
Ahu amazda
no
was
e e ed
‘The
Elami es
we e
un ai h ul
and
Ahu amazda
was
no
e e ed
by
hem.’
No e
he
Old
Indic
example
wi h
a
ini e
e b
(Jamison
1979:
134):
Ishould
like
o
hank
he ewi h
P o .
Vy au as
Amb azas
and
P o .
Algi das
Sabaliauskas
o
eading
and
commen ing
on
ea lie
e sions
o
his
pape .
I
should
also
like
o
hank
P o .
Gied ius
Subaius
o
ansc ibing
he
passage
om
Rhesa’s
Bible
in
he
Chicago
Newbe y
Lib a y.
Al hough
I
ha e
no
men-
ioned
him
di ec ly
in
his
pape ,
I
should
like
o
exp ess
my
hanks
o
P o .
Axel
Hol oe ,
wi h
whose
conclusions
I
equen ly
disag ee,
bu
whose
a icles
ha e
been
help ul
o
me
in
e ising
and
o mula ing
my
own
hough s.
His
2001
a icle
quo ed
in
he
e e ences
was
he
immedia e
s imulus
o
his
pape .
Haud y
(1977:
409)
w i es
ha
he
subjec i e
geni i e
and
he
geni i e
o
belonging
go
back
o
he
same
o igin
as
he
geni i e
o
agen ,
iz.
he
geni i e
o
sou ce,
which
when
used
wi h
he
passi e
pa iciples
deno es
he
au ho
o
he
ac ion
and
is
opposed
o
he
ins umen al
which
deno es
he
elemen
in
con ac ,
he
objec
possessed
o
he
ins umen
o
he
p ocess.
42
|
WILLIAM
R.
SCHMALSTIEG
(2)
idam
ya
pa am
ind apanam,
—
ind asya
NOM.SG.NEUT.
NOM.
SG.
NOM.
SG.
GEN.
SG.
his
ha
cup
Ind a’s
d ink
by
Ind a
p iyam
am am
apayi
NOM.SG.NEUT. NOM.SG.NEUT.
3RDSG.
PSV.
AOR.
dea
immo al
was
d unk
‘This
e y
cup,
Ind a’s
d ink,
dea ,
immo al,
was
d unk
by
Ind a.’
A
Tocha ian
B
example
o
he
geni i e
o
agen
is
gi en
below
(K ause,
Thomas
1960:
83):
(3)
s uko
aisaumyepi
olypo
[ i] oy( a)
NOM.SG.
GEN.
SG.
ADV.
3RD
SG.
MIDDLE
OPTATIVE
dea h
by
wise
(man)
e y
is
o
be
sough
‘ehe
dii e
de
Tod on
einem
Weisen
gesuch
we den’
‘dea h
ough
a he
o
be
sough
by
a
wise
man.’
Now
i
is
well
known
ha
in
he
his o y
o
languages
in
he
cou se
o
ime
he
p epo-
si ion
comes
o
suppo
he
simple
case
ending.
Typically
language
eache s
ell
hei
s uden s
ha
such
and
such
a
p eposi ion
equi es
such
and
such
a
case
o
cases,
bu
in
ac
his o ically
i
is
usually
jus
he
e e se.
The
p eposi ion
(as
some
kind
o
e ymo-
logical
ad e b)
suppo s
an
o iginal
case.
The
eplacemen
o
La .
i e
Romam
by
i e
ad
Romam
‘ o
go
o
Rome’
is
well
known,
bu
he e
a e
many
o he
examples
also.
Fo
example,
K ys’ko
(1997:
121-122)
w i es
ha
o
Russian
du ing
he
p ocess
o
he
o ma ion
o
he
di ec
objec
ca ego y
he
con as
be ween
objec
and
ci cums an-
ial
o ms
is
s eng hened
as
a
esul
o
which
he
accusa i e
o
spa ial
objec
which
combines
he
lexically
loca i e
and
he
unc ional
objec
ca ego ies,
i.e.,
ma ked
bo h
by
ci cums an iali y
and
objec
quali y,
is
g adually
eplaced
by
he
p eposi ion
ce ez
‘ h ough’
plus
case
o m.
In e es ingly
enough
K ys’ko
p edic s
ha
he
simple
accu-
sa i e
o
objec
o
e bs
o
mo ion
will
g adually
be
eplaced
by
he
p eposi ion
plus
objec
cons uc ion
as
i
has
al eady
happened
wi h
ce ain
cons uc ions,
¢.g.,
pe elez ’
ée ez
zabo
‘ o
climb
o e
he
ence’,
pe ele e ’
Ge ez
mo e
‘ o
ly
ac oss
he
sea’.
In
G eek
he
p eposi ion
6x6
‘by’
has
come
o
suppo
he
geni i e
case
in
he
sense
o
agen .
Thus:
(4)
VAY
AAG
LAL
b o
TOY
yeye nyé oy...
ToUTo
—
xaTHYOpE y
1SG.PSV.AOR.
PREP.
GEN.
PL.
GEN.
PL.
GEN.
SG.
_
INF.
I
am
o ced
by he
happenings...
hisone
—_
o
accuse
‘I
am
o ced
by
wha
has
happened
o
accuse
his
man.’
(Goodwin,
Gulick
1958:
261.)
Paul
Ande sen
(1991:
123)
showing
how
in
G eek
he
in oduc ion
o
an
agen
can
change
he
in e p e a ion
o
an
ac i e
in ansi i e
cons uc ion
o
a
passi e
cons uc-
THE
INDO-EUROPEAN
GENITIVE
OF
AGENT
|
43
ion
gi es
he
ollowing
examples
(He odo us
3,
15,
4):
(5)
a a
a eou
TLOV
aneda e
TaeayoTjUA
ACC.
SG.
GEN.
SG.
NOM.
SG.
MASC.
PAST
ACT.
PART.
3
SG.
AOR.
ADV.
blood
o
bull
ha ing
d unk
died
on
he
spo
“Ha ing
d unk
he
bull’s
blood
he
died
on
he spo .’
(He odo us
7,
213,
2)
(6)
anéGa e
o d
"Ady adew
—
d d
006
Tonye iou
3.SG.
AOR.
PREP.
GEN.
SG.
GEN.
SG.
GEN.
SG.
MASC.
died
by
A henades
man
T achinian
“He
was
killed
by
a
T achinian
(called)
A henades.’
Wi h
ega d
o
Old
Chu ch
Sla ic
Vaillan
w o e
(1964:
189)
ha
he
agen
o
a
passi e
e b
may
be
ende ed
by
he
ins umen al
o
o
plus
he
geni i e.
Vaillan
(1964:
195) gi es
he
example
om
Ma hew
3.6:
(7)
k es aaxo
se
o e
nego
3PL.IMPERF.
REFL.PTCL.
PREP.
—
GEN.
SG.
we e
bap ized
by
him
‘They
we e
bap ized
by
him.’
The
addi ion
o
om
me ely
s eng hens
he
o iginal
geni i e
o
agen
meaning.
Abo e
I
ha e
gi en
e idence
o
he
use
o
an
agen i e
geni i e
wi h
a
ini e
e b
in
Old
Pe sian,
Old
Indic,
Tocha ian,
G eek
and
Sla ic.
I
seems
o
me
hen
ha
an
Indo-
Eu opean
o igin
o
he
cons uc ion
is
qui e
likely.
Adhe ing
o
he
no ion
ha
unmo i a ed
syn ax
has
i s
o igin
in
mo i a ed
seman-
ics,
I
p opose
ha
he
unmo i a ed
geni i e
case
which
is
equi ed
by
ce ain
e bs
is
he
e lec ion
o
an
ea lie
agen i e
meaning.
F om
he
synch onic
poin
o
iew
one
would
say
ha
such
and
such
a
e b
equi es
a
ce ain
case,
bu
he e
is
no
mo i a ion
o
ha .
In
many
cons uc ions,
hen,
one
can
unde s and
he
subjec
as
o iginally
being
he
objec
(o
pa ien )
o
he
ac ion
o
s a e
and
he
geni i e
objec
as
he
agen
o
he
sou ce
o
he
ac ion
o
s a e.
When
he
Indo-Eu opean
language
shi ed
om
e ga i e
ypology
o
nomina i e-accusa i e
ypology,
he
o iginal
pa ien
came
o
be
el
as
a
subjec
and
he
analysis
o
he
old
e ga i e
ype
sen ences
was
changed.
Thus,
o
example,
ce ain
Old
English
cons uc ions
a e
adi ionally
called
‘im-
pe sonal’,
al hough
Visse
(1963:
20)
conside s
his
e m
a
misnome .
Acco ding
o
Visse
(1963:
23):
“This
cons uc ion
is
no
in equen ly
accompanied
by
a
comple-
men
in
he
o m
o
a
noun
o
p onoun
in
he
geni i e...
o
by
a
p eposi ion
( o
o ,
e ,
o
e c.)
+
noun...
Since
i
is
con enien
o
ha e
a
name
o
e e
o
he
complemen
in
he
o m
o
a
(p o)noun
in
he
geni i e
i
will
be
called
CAUSATIVE
OBJECT,
as
in
he
majo i y
o
cases
i
exp esses
he
cause
o
he
occasion
o
he
ac ion
o
s a e
deno ed
by
he
e b
i
quali ies.”
Visse ’s
e m
‘causa i e
objec ’
is
e y
app op ia e
since
i
i s
wi h
he
no ion
ha
he
geni i e
is
an
agen
o
a
sou ce
p oducing
he
ac ion
o
s a e.
Example:
44
|
WILLIAM
R.
SCHMALSTIEG
(8)
him
sceamode
Owes
mannes
DAT.
SG.
3
SG.
PRET.
GEN.
SG.MASC.
GEN.
SG.
o
him
i
shamed
o
ha
man
‘he
was
ashamed
o
ha
man.’
This
‘impe sonal
cons uc ion’
wi h
a
‘causa i e
objec ’
was
p obably
p eceded
by
a
pe sonal
cons uc ion
also
wi h
a
‘causa i e
objec ’.
In
Old
English
a
da i e
objec
eplaced
an
o iginal
nomina i e
subjec .
Compa e
he
Go hic
example
wi h
a
cogna e
e b
(Luke
9.26):
(9)
sae...
skamaip
sik
meina
NOM.
SG.
3
PRES.
REFL.
GEN.
SG.
whoe e
shames
himsel
o
me
‘whoe e
is
ashamed
o
me’
(S ei be g
1919:
131)
Acco ding
o
E ha
(1980:
182)
Indo-I anian
ou h
class
e bs
( hose
in
-ya)
we e
o iginally
s a i e,
bu
some
had
de eloped
in o
passi es
al eady
in
he
Indo-I anian
pe iod.
Le
us
compa e
he
ollowing
sen ence
om
Old
Indic
(Macdonell
1916:
319):
(10)
dnnasya
pya i
GEN.
SG.
3RD
SG.
PRES.
wi h
ood
he
e eshes
himsel
‘He
e eshes
himsel
wi h
some
ood.’
Acco ding
o
he
ex -book
in e p e a ion
he
e b
7p-
go e ns
he
geni i e
case,
bu
no e
ha
his
e b
belongs
o
he
Indic
ou h
class
and
ha
one
could
e e se
he
syn ac ic
in e p e a ion
assuming
a
geni i e
agen
(o
in
Visse ’s
e ms,
a
causa i e
objec )
and
a
nomina i e
pa ien .
One
could
hen
unde s and
he
sen ence
as:
‘ he
ood
e eshes
him’.
Conside
now
he
ollowing
La in
example:
(11)
pude
mé
ui
3
SG.
PRES.
ACC.
SG.
GEN.
SG.
shames
me
you
‘I
am
ashamed
o
you.’
(Benne
1945:
140)
This
also
could
be
unde s ood
as:
‘you
shame
me’.
Compa e
he
Old
English
ex-
ample
(8)
and
Go hic
example
(9)
abo e.
He mann
(1926:
286-287)
connec s
he
La in
e bs
in
*-é-
wi h
he
Li huanian
s a i e
e bs
in
-é-
which
equen ly
equi e
a
geni i e
objec ,
c .
he
Li huanian
sen ence
om
Basana icius:
(12)
0
jéi
manés
aip
bjau iesi
CONJ.
CONJ.
NOM.
SG.
GEN.
SG.
ADV.
2
SG.
PRES.
bu
i
you
by
me
so
a e
epelled
‘bu
i
you
a e
so
epelled
by
me.’
THE
INDO-EUROPEAN
GENITIVE
OF
AGENT
|
45
The
e b
bjau é is
‘ o
be
epelled’,
is
an
-é-
s em
e b
(<*-é-).
This
could
be
unde -
s ood
as:
‘bu
i
I
epel
you
so’.
O he
Li huanian
examples
include:
(13)
Vaikas
s eciq
d o isi
(in .
d o é is)
NOM.
SG.
GEN.
PL.
3
PRES.
child
by
gues s
is
made
shy
(DZ
1993:
13)
‘The
child
is
made
shy
by
he
gues s,
i.e.,
he
child
is
shy
in
he
p esence
o
he
gues s.”
This
could
also
be
e e sed
and
unde s ood
as
‘The
gues s
make
he
child
shy’.
(14)
Manés
isi
bédis
(in .
bodé is)
GEN.
SG.
NOM.
PL.
3
PRES.
by
me
e e yone
is
annoyed
(LKZ,
966)
‘E e yone
is
annoyed
by
me’
which
could
be
unde s ood
as
‘I
annoy
e e yone’.
The
e bs
in
examples
(9)
h ough
(14)
abo e
all
belong
o
he
same
Indo-Eu o-
pean
class
o
s a i e
e bs
and
all
he e
ake
geni i e
objec s,
o
in
Visse ’s
e m
‘caus-
a i e
objec s’.
See
also
Amb azas
(2001a:
29-34)
o
many
o he
examples
and
some-
wha
simila
conclusions.
In
my
iew
an
analysis
o
he
geni i e
o
agen
canno
be
sepa a ed
om
an
analysis
o
he
ins umen al
wi h
which
i
compe es
in
ce ain
syn ac ic
colloca ions2.
No e
he
ollowing
Li huanian
sen ence
(Paulauskiené
1979:
99):
(15)
Zemé
es i
sniegi
(sniégo)
nuklo a
NOM.SG.
3
PRES.
INSTR.
SG.
GEN.
SG.
NOM.
SG.
FEM.-T-PART.
ea h
is
wi hsnow
(bysnow)
co e ed
‘The
ea h
is
co e ed
wi h
(by)
snow.’
In
his
case
he
use
o
he
ins umen al
snieg
would
deno e
ha
snow
is
conside ed
he
ins umen ,
whe eas
use
o
he
geni i e
sniégo
would
deno e
ha
snow
is
he
agen
2
Now
one
may
apply
a
simila
dicho omy
o
he
*-o
s em
nouns.
Then
he
nomina i e
singula
in
*-5
(p obably
an old
e ga i e)
may
be
conside ed
he
sou ce
o
he
ac ion
as
opposed
o
he
nomina i e
singu-
la
in
*N
(=
*-m
o
*-n,
p obably
an
old
ins umen al)
which
may
be
conside ed
he
ins umen .
The
eason
o
he
assignmen
o
anima e
nouns
o
he
*-s
mo phological
ca ego y
is
ha
hey
a e
mo e
likely
o
unc ion
as
he
sou ce
o
he
ac ion.
This
may
explain
hen
why
such
La in
nouns
as
gladium
‘swo d’,
scu um
‘shield’,
caelum
‘sky’
ha e
olde
al e na e
nomina i es
singula
gladius,
scu us,
caelus.
C .
also
Gk.
Guy6
‘plow’
beside
Cuyés.
O iginally
he
choice
o
he
nomina i e
singula
depended
on
he
speake ’s
pe cep ion
o
he
unc ion
o
he
subjec ,
iz.,
ei he
as
sou ce
(wi h
he
ending
*-os)
o
ins umen
(wi h
he
ending
*-oN).
One
is,
o
cou se,
amilia
wi h
he
phenomenon
o
gende
change
in
he
adjec i e,
iz.
La .
bon-us,
bon-um,
bon-a,
e c.,
bu
o
he
noun
I
p esuppose
di e en
o iginal
unc ions
o
he
nomina i e
singula
o ms
*-os,
*-oN,
*-d,
al hough
a
he
momen
I
would
ha e
no
explana ion
o
he
di e ence
be ween
*-oN
and
*-4,
o
he
la e
o
which
Haud y
(1977:
449)
sugges s
an
o iginal
ins umen al
unc ion.
He
compa es
he
o m
o
he
*-4
s em
ins umen al
unc ion
and
nomina i e
singula
in
he
La in
sen ences
(1)
miles
sagi a
hos em
ulne a
‘ he
soldie
wounds
he
enemy
wi h
an
a ow’
and
(2)
mili is
sagi a
hos em
ulne a
‘ he
soldie ’s
a ow
wounds
he
enemy’.
This
would
explain
why
he
same
oo
may
appea
in
ela ed
languages
as
masculine
o
neu e
*-o
s ems
o
(mos ly
eminine)
*-a
s ems.

46
|
WILLIAM
R.
SCHMALSTIEG
o
he
sou ce
o
he
ac ion.
I
should
be
no ed
ha
he
agen i e
use
o
he
geni i e
is
no
es ic ed
o
anima e
beings.
O he
examples
quo ed
by
Paulauskiené
(1979:
99)
include:
(16)
Daik inis
a
eiksminis
simbolis
bu o
uzs elb as
NOM.
SG.
CONJ.
NOM.
SG.
NOM.
SG.
3PRET.
NOM.
SG.
-T-
PART.
ma e ial
o
ac i e
symbol
was
choked
o
iF
uz émdy as
Z6diniu
ap alkalu
(z6dinio
dp alkalo)
CONJ.
NOM.
SG.
-T-
PART.
INST.
SG.
INST.
SG.
GEN.
SG.
GEN.
SG.
and
eclipsed
e bal
co e ing
( e bal
co e ing)
‘The
ma e ial
o
ac i e
symbol
was
choked
o
and
eclipsed
by
he
e bal
co e ing.’
(17)
Bend uomenés
_is o ijos
ki imas
y a
GEN.
SG.
GEN.
SG.
NOM.
SG.
3
PRES.
o
he
socie y
o
he
his o y
es ablishmen
is
nejz elgiama__paslap im
_(nejz elgiamos
paslap ies)
_
p ideng as...
INST.
SG.
INST.
SG.
GEN.
SG.
GEN.
SG.
NOM.
SG.
MASC.
-T-
PART.
wi h
an
impene able
sec e
(by
an
impene able
sec e )
hidden
‘The
es ablishmen
o
he
his o y
o
he
communi y
is
hidden
wi h
(by)
an
impene able
sec e .’
One
mus
say
ha
he e
he
au ho
and
he
ins umen
(o
pe haps
sou ce
and
pa h)
a e
no
clea ly
di e en ia ed,
al hough
in
mos
cases
he
Li huanian
language
would
di e en ia e
qui e
clea ly
(Paulauskiené
1979:
99).
S ill
e en
wi h
e bs
in
-é i
one
no es
he
compe i ion
be ween
he
geni i e
and
he
ins umen al,
c .
he
ollowing
example
wi h
(14)
abo e:
(18)
Kai
jisai
si go,
isi
jud
bodéjosi
CONJ.
NOM.SG.
3
PRET.
NOM.
PL.
INSTR.SG.
3
PRET.
When
he
was
sick
e e ybody
wi hhim
was
annoyed
“When
he
was
sick,
e e yone
was
annoyed
wi h
him.’
(LKZ,
967)
In
example
(3)
abo e
I
ha e
gi en
a
Tocha ian
example
wi h
a
geni i e
o
agen .
Bu
in
addi ion
o
he
geni i e
in
Tocha ian
he
ins umen al
( o
inanima e)
and
he
pe la i e
( o
anima e)
cases
may
unc ion
as
ins umen
and
agen
cases
also.
No e
he
ollowing
example
om
Tocha ian
A
(K ause,
Thomas
1960:
84;
Lane
1947:
49):
(19)
ma
po yo
skdmsam a
ma
wad yo
ADV.
INSTR.SG.
3
PL.
MIDDLE
PRES.
ADV.
INSTR.
no
by
i e
a e
bu ned
no
by
wa e
THE
INDO-EUROPEAN
GENITIVE
OF
AGENT
|
47
sikam é
ma
lancsa
pa si
ya enc
3
PL.
MIDDLE
PRES.
ADV.
PERLATIVE
PL.
INF.
3
PL.
PRES.
washed
away
no
by
kings
o
ake
can
be
‘They
a e
no
bu ned
by
i e,
no
washed
away
by
wa e ,
canno
be
ca ied
(away)
by
kings...’
I
ha e
no ed
abo e
in
G eek
examples
(4)
and
(6)
he
geni i e
o
agen
go e ned
by
he
p eposi ion
x6.
The
G eek
da i e
unc ions
as
ins umen al,
since
G eek
has
no
inhe i ed
he
ins umen al
om
P o o-Indo-Eu opean.
Thus
simila
o
he
Li huanian
compe i ion
o
geni i e
and
ins umen al
is
he
G eek
compe i ion
o
he
geni i e
wi h
he
da i e.
No e
he
examples
om
Schwyze
(1966:
526):
(20)
=
édauy,
oxo
yEeo
TOdaxE06
Aiaxidao
3.SG.
PSV.
AOR.
PREP.
DAT.
PL.
GEN.
SG.
GEN.SG.
was
slain
by
hands
o
swi - oo ed
Aeacus
“He
was
slain by
he
hands
o
swi - oo ed
Aeacus.’
(Iliad
II,
860)
(21)
o
ono
do gi
TOALY
éoSa
DAT.
SG.
PREP.
DAT.
SG.
ACC.
SG.
MIDDLE
INFINITIVE
hy
by
spea
ci y
o
be
laid
was e
“ci y
o
be
laid
was e
by
hy
spea ’
(Iliad
XVI,
708)
As
wi h
he
geni i e
case
he
p eposi ion
4x6
me ely
comes
o
s eng hen
he
o igi-
nal
ins umen al.
The
ac
ha
6x6
is
used
wi h
bo h
he
geni i e
(wi h
agen i e
mean-
ing)
and
he
da i e
(wi h
ins umen al
meaning)
shows
ha
he
o iginal
di e ence
in
meaning
was
de e mined
by
he
case
and
no
he
p eposi ion.
The
use
o
he
ins umen al
wi h
he
passi e
(as
con as ed
wi h
he
geni i e
as
in
sen ence
[2]
abo e)
is
a es ed
in
Old
Indic
(Jamison
1979:
131):
(22)
adbhi
gobhi
mujya e
INSTR.
PL.
INSTR.PL.
3
PRES.
wi hwa e
wi h
cows
__is
wiped
“He
is
wiped
wi h
wa e
(and)
cows
(=
milk).’
(23)
sd
mujyémanah
ka ibhih...
NOM.
SG.
MASC.
NOM.
SG.
MASC.
PSV.
PART.
INSTR.
PL.
he
(=
Soma),
being
wiped
by
he
ka is...
Acco ding
o
Haud y
(1977:
414)
he
use
o
he
ins umen al
as
an
agen
in
Old
Indic
is
a
esul
o
he
ex ension
o
he
subjec
o
o igin
o
he
subjec
o
ins umen .
Haud y
w i es
ha
his
esul s
om
he
ac
ha
sayo
hin an i
‘ he
see s
inci e’
and
he
inno a ion
ma ayo
hin an i
‘ he
hough s
inci e’
a e
pa allel,
so
ha
he
pa allel
o
ma ibhi
hi d
‘inci ed
by
he
hough s’
would
be
sibhi
hi d
‘inci ed
by
he
see s’.
48
|
WILLIAM
R.
SCHMALSTIEG
In
Russian
he
compe i ion
be ween
he
geni i e
o
agen
and
he
ins umen al
is
e iden
in
his
example
om
he
Igo
Tale,
c .
he
Jakobson
econs uc ion
(1966:
170):
(24)
Posképani
sabljami
—_kalenymi
Selomi
NOM.
PL.
MASC.
PAST
PSV.
PART.
INSTR.
PL.
INSTR.
PL.
NOM.
PL.
cle
wi h
sabe s
empe ed
helme s
O a eskii
o e
ebe,
Ja o
u e
Vose olode!
NOM.
PL.
MASC.
PREP.
GEN.
SG.
VOC.
SG.
VOC.
SG.
VOC.
SG.
A a
by
you
Fie ce
Bull
Vse olod!
‘Cle
wi h
empe ed
sabe s
a e
[ hei ]
A a
helme s
—
by
you,
Fie ce
Bull
Vse olod!’
In
mode n
Russian,
o
cou se,
his
dis inc ion
has
been
los
and
co espondingly
such
mode n
Russian
ph ases
as
(25)
and
(26):
(25)
napisano
mnoju
NOM.
SG.
NEUT.
PAST
PSV.
PART.
INSTR.
SG.
‘w i en
by
me’
(26)
napisano
ka andasom
NOM.
SG.
NEUT.
PAST
PSV.
PART.
INSTR.
SG.
‘w i en
wi h
a
pencil’
would
be
ende ed
in
Li huanian
by
(27)
and
(28)
espec i ely,
see
Paulauskiené
(1979:
99).
(27)
pa asy a
mano
NOM.
SG.
NEUT.-T-PART.
GEN.
SG.
‘w i en
by
me’
(28)
pa asy a
pies uki
NOM.
SG.
NEUT.
-T-
PART.
INSTR.
SG.
‘w i en
wi h
a
pencil’
In
mode n
Russian
we
see
a
de elopmen
pa allel
o
ha
o
Old
Indic,
iz.
he
e-
placemen
o
(o @
plus)
he
geni i e
by
he
ins umen al.
The
compe i ion
be ween
he
geni i e
and
ins umen al
is
also
encoun e ed
in
he
Old
Indic
sen ences
quo ed
om
Pa a ijali
by
Wacke nagel
and
Deb unne
(1954:
582):
(29)
idam
aheh
s p am
ADV.
GEN.
SG.
|
NOM.-ACC.
SG.
NEUT.
-T-
PART.
he e
se pen
c awled
THE
INDO-EUROPEAN
GENITIVE
OF
AGENT
|
49
(30)
ihahina
(=
ha
+
ahina)
s p am
ADV.
INSTR.SG.
|
NOM.-ACC.SG.NEUT.
-T-PART.
he e
se pen
c awled
‘The
se pen
c awled
he e.”
In
mode n
A menian
he
pa iciple
in
-c
is
used
wi h
a
geni i e
o
agen
which
ap-
pa en ly
does
no
go
back
o
Indo-Eu opean
imes
and
mus
be
conside ed
an
inno a-
ion,
di e en ly
om
he
use
o
he
geni i e
wi h
he
pa iciple
in
-/
which
is
mos
likely
ancien .
No e
he
mode n
A menian
example
(Fai banks
and
S e ick
1975:
227):
(31)
im
ba ekami
—
g ac
gi k‘a
he ak‘aki
é
POSS.
PRON.
|
GEN.SG.
—PASTPART.
~—
NOM.
SG.
NOM.SG.
3
PRES.
my
by
iend
w i en
book
in e es ing
is
‘The
book
w i en
by
my
iend
is
in e es ing.’
Such
a
sen ence
could
be
ansla ed
in o
Li huanian
wi h
almos
exac ly
he
same
syn ac ic
s uc u e:
(32)
mano
bici lio
—_—ppa asy a
knyga
y a
idomi
POSS.
PRON.
|
_GEN.SG.
_
PAST
PART.
NOM.SG.
3
PRES.
NOM.
SG.
my
by
iend
w i en
book
is
in e es ing
Wi h
he
pa iciple
in
-c
we
also
encoun e
he
ins umen al
usage
in
A menian
(Tumanjan
1962:
58):
(33)
angin
K‘a o
Sa ac
ADJ.
INSTR.SG.
PAST
PART.
aluable
=
wi hs one
s udded
‘s udded
wi h
aluable
jewels’
C .
Li h.
(34)
papuoS as
b angakmeniais
NOM.
SG.
INSTR.
PL.
‘ado ned
wi h
jewels’
One
a gumen
o
he
possessi e
o igin
o
he
geni i e
o
agen
in
Bal ic
is
he
accep ed
use
o
he
possessi e
p onouns
Li h.
mano,
La .
mana
‘my,
by
me’
and
Li h.
a o,
La .
a a
‘you
(sg.),
by
you
(sg.)’
as
agen s,
a he
han
he
s anda d
geni i es
Li h.
manés,
La .
manis
‘my,
by
me’ and
Li h.
a és,
La .
e is.
The e
a e,
o
cou se,
excep ions.
Endzelins
(1923:
384)
no es
he
acilla ion
be ween
La .
dziésma
mana
padziédd a
‘ he
song sung
by
me’
and
la e
in
he
same
place
manis
ja ina
padziédd a
‘a
me
ju enco
can a a,
sung
by
me,
a
young
( ellow)’.