ACTA
LINGUISTICA
LITHUANICA
XLIV
(2001),
71-87
Li huanian
bi i
wi h
he
in ini i e
as
a
modal
exp ession
and
i s
La ian
coun e pa s
AXEL
HOLVOET
Lie u iy
kalbos
ins i u as,
Vilnius
In
Bal ic,
bi i
+
in ini i e
is
now
used
as
a
modal
cons uc ion
exp essing
necessi y.
I
de eloped
ou
o
an
exis en ial
cons uc ion
wi h
bi i,
expanded
wi h
an
in ini i e
o
pu pose
(namas
y a
s a y i
‘ he e
is
a
house
o
be
buil ’
—
‘a
house
can/should
be
buil ’).
This
cons uc ion,
o iginally
ague
be ween
possibili y
and
necessi y,
subsequen ly
became
specialised
in
he
meaning
o
necessi y.
The
o iginal
nomina i e
wi h
bi i
was
eanalysed
as
an
objec
and
eplaced
wi h
he
accusa i e.
Whe e-
as
he
ype
exp essing
necessi y
is
p oduc i e
in
bo h
Li huanian
and
La ian,
only
Li huanian
has
e ained
he
o iginal
cons uc ion
wi h
a
nomina i e
as
an
exp ession
o
possibili y
(man
namai
ma y i
‘I
can
see
he
house’).
La ian
coun e pa s
o
his
cons uc ion
a e
some imes
ci ed
in
he
li e a u e,
bu
hey
ep esen
a
di e en
syn ac ic
ype
which
is
an
inno a ion
o
La ian:
i
is
cha ac-
e ised
by
a
nomina i e
ins ead
o
he
o iginal
da i e
o
he
agen /modal
subjec
and
a
di e en
s a us
o
he
nega ion:
zi gam
ezumu
nepa ilk
>
zi gs
ezumu
ne
pa ilk .
1.
THE
CONSTRUCTION
ba i
+
INFINITIVE
IN
THE
LITERATURE
The
use
o
Bal ic
bi i
and
Sla onic
by i
wi h
he
in ini i e
as
an
exp ession
o
possibili y
o
necessi y
is
well
known.
In
he
g amma s
as
well
as in
monog aphs
on
his o ical
and
synch onic
syn ax
i
is
some imes
e e ed
o as
i
is
done
he e
(1.e., as
a
cons uc ion
in ol ing
he
e b
‘be’),
e.g.,
Miklosich
(1926:
859)
w i es
‘das
subjec lose
e bum
sub-
s an i um
bezeichne
in
e bindung
mi
einem
in ini i
die
no hwendigkei
ode
méglich-
kei
und
nega i
die
physische
ode
mo alische
unméglichkei
eine
handlung’.
A
simila
accoun
can
be
ound
in
B ugmann
(1916:
923-926).
In
o he
cases,
he
lack
o
an
o e ly
exp essed
e b
in
he
p esen
ense
(es e,
y a,
i
e c.)
leads
he
au ho s
dealing
wi h
he
ype
unde
discussion
o
e e
o
i
as
an
ins ance
o
he
p edica i e
in ini i e
( o
La ian,
c .
Endzelin
1951:
991;
o
Li huanian,
c .
Ul ydas,
ed.,
1976:
323)'.
I ,
alongside
a
cons uc-
In
some
(mainly
Russian)
wo ks,
he
basic
s uc u al
iden i y
o
cons uc ions
wi h
and
wi hou
o e
o ms
o
he
e b
‘be’
is
no
dis ega ded,
bu
bo h
a e
desc ibed
as
ins ances
o
an
‘indepen-
den
in ini i e’
(c .
Bo ko skij,
ed.,
1978:
278-283).
Such
a
ea men
is
appa en ly
based
on
he
assump ion
ha
he
pas
and
u u e
ense
o ms
o
he
e b
by i
used
wi h
he
in ini i e
a e
no
ins ances
o
an
au onomous
e b
bu
me e
ense
ma ke s.
TZ
|
AXEL
HOLVOET
ion
wi hou
an
o e ly
exp essed
p esen
ense
o
‘be’,
he e
is
a
compa able
cons uc ion
wi h
an
o e
ense
o m
(say,
Li huanian
Man
bu o
ei i
‘I
had
o
go/was
o
ha e
gone’
alongside
Man
aziuo i
‘I
mus /should
go’),
hen
i
is
clea
ha
he
o me
mus
con ain
a
ze o
ealisa ion
o
he
3 d
pe son
o m
o
he
p esen
ense
o
‘be’
( his
ze o
ealisa ion
is
a
ea u e
well
known
om
bo h
li ing
Bal ic
languages
as
well
as
om
se e al
Sla onic
languages),
and
ha
we
a e
dealing
wi h
a ie ies
o
one
single
cons uc ion
a he
han
wi h
wo
di e en
cons uc ions.
Howe e ,
no
all
ense
o ms
a e
used
wi h
he
same
egula i y
( o
example,
he
cons uc ion
may
be
i ually
es ic ed
o
he
p esen
ense,
i.e.,
he
a ie y
wi hou
o e
o m
o
‘be’);
u he mo e,
addi ional
unc ional
di e ences
may
appea
ha
may
obscu e
he
undamen al
iden i y
be ween
hose
cons uc ions
ha
con ain
some
o m
o
he
e b
‘be’
and
hose
ha
do
no *.
Fo
Li huanian,
bo h
Palionis
(1972)
and
Amb azas
(1995)
conclude
ha
he e
is
no
eason
o
se
he
wo
ypes
apa .
In
o he
wo ds,
man
aziuo i
should
be
in e p e ed
s uc u ally
as
man
(y a)
aziuo i
( o
Old
Russian
his
was
emphasised
al eady
by
Po ebnja
1958:
394).
Ano he
eason
why
pa icula
a ie ies
o
he
ype
discussed
he e
a e
some imes
se
apa
as
independen
cons uc ions
is
ha
he e
occu
objec s
in
di e en
case
o ms.
In
Li huanian
( es igially
in
La ian)
and
in
some
Sla onic
dialec s
we
ind
a
nomina i e
objec
alongside
an
accusa i e
objec
wi h
he
in ini i e:
man
namai
s a y i
alongside
man
namus
s a y i
‘I
ha e
o
build
a
house/should
build
a
house’
( he
examples
used
o
ep e-
sen
hese
models
a e
ci ed
a e
Amb azas
1995).
I
he
nomina i e
is
in e p e ed
as
a
subjec
and
he
accusa i e
as
an
objec
(which
is,
in
a
way,
accu a e
bu
means
pu ing
phenomena
om
di e en
s ages
o
language
de elopmen
on
one
le el,
as
he
eanalysis
o
an
o iginal
subjec
as
objec
is
in ol ed
he e,
.
in a),
hen
he e
is
no
way
o
accoun ing
o
bo h
ypes
in
a
simila
way. This
led
F aenkel
(1928:
118-119)
o
say ha he
in ini i e
a ai i
in
kaip
bua iam
a ai i
is
he
subjec
o
bi i
whe eas
in
iumus
bus
us us
sudas
pas
Diewa
kel i
he
in ini i e
exp esses
a
p edica e,
he
subjec
o
which
is
sudas
(F aenkel
1928:
14;
bo h
examples
om
he
Wol enbii el
Pos il,
wi hou
p ecise
loca ion).
The
in e p e a ion
o
he
in ini i e
as
he
subjec
o
bi i
(sugges ed
by
F aenkel
o
kaip
bua
iam
a ai i)
canno
be
an
accu a e
desc ip ion
o
he
oldes
s age
o
he
cons uc ion
deal
wi h he e
because
he
Bal ic
and
Sla onic
in ini i e
was
o iginally
he
da i e
o
a
e bal
noun
in
- -
and
i s
o iginal
scope
was
he e o e
es ic ed
o
wha
is
now
desc ibed
as
he
in ini i e
o
pu pose.
In
econs uc ing
he
oldes s ages
o
he
Bal ic
in ini i al
cons uc ions
one
should
always
be
awa e
o
he
ela i ely
ecen
o igin
o
he
in ini i e
as
a
e bal
o m
and
o
he
ange
o
possible
syn ac ic
unc ions
he
in ini i e
could
ini ially
pe o m
by
i ue
o
i s
mo phological
na u e,
iz.
ha
o
he
da i e
o
a
e bal
noun.
Whe e e
an
in ini i e
unc ions
as
a
subjec
in
he
sense
o
occupying
he
place
o
a
nomina i e
subjec
(as
in
Tu e i
daug
dedziy
[...]
nemaza
ga bés
i
malonumo
‘ o
ha e
many
uncles
...
is
no
small
honou
and
pleasu e’
—
Vaizgan as,
ci ed
in
Ul ydas,
ed.,
1976:
292),
his
s a e
o
a ai s
is
a
ela i ely
ecen
de elopmen .
Amb azas
(1995)
dis inguishes,
wi hin
he
syn ac ic
ype
unde
conside a ion
he e,
wo
sub ypes:
(man)
y a
/
@
/
bu o
/
bus
namai
ma y i
and
(man)
y a
/
@/
bu o
/
bus
namai
-
As
an
ins ance,
one
could
ci e
he
unc ional
di e en ia ion
be ween
cons uc ions
wi h
a
ze o
o m
o
he
e b
‘be’
(p esen
ense
o ms)
and
hose
wi h
he
u u e
bis
in
La ian
( .
in a).
LITHUANIAN
ba& i
WITH
THE
INFINITIVE
|
73
s a y i.
Bea ing
in
mind
he
o igin
o
he
Bal ic
in ini i e,
he
analyses
he
cons uc ion
(man)
y a
|
@
|
bu o
/
bus
namai
ma y i
as
an
expanded
a ie y
o
a
cons uc ion
wi h
a
subs an i e
e b
and
a
nomina i e
subjec .
The
nomina i e
subjec
subsequen ly
unde -
wen
a
eanalysis
and
began
o
unc ion
as
he
objec
o
he
in ini i e.
This
is
one
o
he
sou ces
o
he
nomina i e
objec
cha ac e is ic
o
in ini i al
cons uc ions
in
Bal ic.
The
eplacemen
o
he
nomina i e
objec
wi h
an
accusa i e
objec
ga e
ise
o
he
newe
a ie y
(man)
y a
/
@
/
bu o
/
bus
namus
s a y i.
Alongside
he
abo e-men ioned
models
wi h
ansi i e
e b,
Amb azas
discusses
a
model
wi h
in ansi i e
e bs:
(man)
y a
/@/
bu o
/
bus
ei i.
This
model
is
undoub edly
o
mo e
ecen
o igin,
and
i
mus
ha e
de elo-
ped
on
he
basis
o
he
ansi i e
model
a e he
eanalysis
had
aken
place.
Amb azas’
accoun
o
his
p ocess
as
an
ins ance
o
eanalysis
(basically
e lec ing
ha
o
B ugmann
1916:
925)
is
qui e
con incing,
hough
al e na i e
explana ions
ha e
been
o e ed
as
well:
Timbe lake
(1974)
explains
he
nomina i e
objec
as
an
a eal
con e gen-
ce
linking Finnic,
Bal ic
and
No h
Russian.
As
Amb azas
has
poin ed
ou ,
hese
wo
explana ions
a e
no
necessa ily
con adic o y.
Amb azas’
econs uc ion
is
he
mos
plau-
sible
explana ion
o
his
exp ession
o
necessi y
anyway,
as
a
cons uc ion
con aining
he
e b
‘be’
combined
wi h
an
in ini i e
o
pu pose
o
some
equi alen
cons uc ion
exp essing
pu pose
is
a
common
means
o
exp essing
possibili y
and/o
necessi y,
c .
English
wha
is
o
be
done,
Ge man
die
F age
is
nich
zu
ldsen
e c.).
We
may
assume
ha
such
cons uc ions
o igina e
as
exis en ial
cons uc ions
and
ha
hey
a e
subsequen ly
eanalysed
as
modal,
1.e.,
he
o iginal
model
is
y a
namai
s a y i
/
ma y i,
which
would
mean
some hing
like
‘ he e
is
a
house
o
see/build’.
In
o he
wo ds,
namai
was
o iginally
he
subjec
o
an
exis en ial
cons uc ion.
Timbe lake’s
hypo hesis
is
he e o e
equi ed
no
so
much
o
explaining
he
o igin
o
he
use
o
he
nomina i e
as
o
explaining
he
e en ion
o
he
nomina i e
a e
he
cons uc ion
had
been
eanalysed
as
a
modal
exp es-
sion,
i.e.,
i s
e en ion
a
a
s age
when
he
nomina i e
subjec
had
become
a
nomina i e
objec .
And
his
is
p ecisely
he
ole
asc ibed
o
he
Fennic
subs a um/ads a um
by
Amb azas
(1997:
99).
2.
THE
PROBLEM
OF
ba i
AS
A
MODAL
VERB
I
he
change
in
he
s a us
o
he
nomina i e
in
namaiy a
s a y i
(i.e.
i s
eanalysis
as
a
nomina i e
objec
and
i s
e en ual
eplacemen
wi h
an
accusa i e
objec )
is
clea ,
no
a en ion
has
been
paid
un il
now
in
he
li e a u e
o
he
changes
in
he
s a us
o
he
e b
‘be’.
O iginally,
i
was
a
subs an i e
e b
exp essing
an
exis en ial
p edica e.
Be o e
he
eanalysis
o
he
nomina i e
subjec
as
an
objec
he e
mus
ha e
been
a
ce ain
pe iod
when
he
o iginal
syn ac ic
s uc u e
o
he
clause
was
e ained
bu
a
shi
had al eady
occu ed
in
seman ic
s uc u e,
he
whole
cons uc ion
ha ing
acqui ed
he
meaning
o
an
exp ession
o
possibili y
o
necessi y.
A
simila
si ua ion
can
be
obse ed
in
he
case
o
o he
cons uc ions
ha
ha e
de eloped
o
a e
de eloping
a
modal
meaning.
So,
o
ins-
ance,
a
modal
meaning
( ha
o
lack
o
necessi y)
has
de eloped
in
Bal ic
and
Sla ic
cons uc ions
whe e
an
in ini i al
ela i e
clause
is
adjoined
o
he
implici
subjec
o
an
exis en ial
cons uc ion.
Li huanian
né a
ko
skai y i,
La ian
na
ko
lasi
means
‘ he e
is
no hing
o
ead’,
bu
also
‘ he e
is
no
poin
in
eading
(any hing,
o
some
pa icula
ex )’.
74
|
AXEL
HOLVOET
The
la e
eading
leads
o
he
use
o
his
cons uc ion
in
modal
unc ion,
in
which
case
he
ela i e
p onoun
loses
i s
g amma ical
ela ion
as
objec
o
he
in ini i e
skai y i,
lasi
and
he
in ini i e
may
ake
an
objec
o
i s
own: Né a
ko
skai y i
okiy
knygy,
Na
ko
lasi
adas
g ama as*.
The
in oduc ion
o
an
objec
(dis inc
om
he
ela i e
p onoun)
wi h
he
in ini i e
is
an
ins ance
o
he
ac ualisa ion
ollowing
upon
he
modal
eanalysis
o
he
cons uc ion.
Those
ins ances
whe e
no
new
objec
is
in oduced
hough
he
in ini i e
is
ansi i e
(pa ly
as
a
esul
o
ellipsis,
e.g.,
né a
ko
skai y i
‘ he e
is
no hing
o
ead’
o
‘ he e
is
no
poin
in
eading
[i ]’,
iz.
he
book
you
jus
men ioned)
a e
now
syn ac ically
ambiguous,
bu
he e
mus
e iden ly
ha e
been
an
in e media e
s age,
p io
o
syn ac ic
changes,
in
which
only
a
shi
in
meaning
had
occu ed. A
simila
si ua ion
may
be
e-
cons uc ed
o
namai
s a y i
‘ he e
is
a
house
o
build’,
‘a
house
mus
be
buil ’.
This
new
modal
meaning
led,
in
i s
u n,
o
a
syn ac ic
eanalysis
in
such
a
way
ha
y a
s a ed
unc ioning
as
an
exp ession
o
modali y,
i.e.,
as
ep esen ing
a
modal
p edica e,
he
a gumen
o
which
was
ep esen ed
by
he
embedded
s uc u e
[namai
s a y i};
his,
in
u n,
led
o
he
eanalysis
o
namai
as
a
nomina i e
objec
and
o
i s
e en ual
eplacemen
wi h
he
accusa i e.
The
p ocess
is
exac ly
pa allel
o
ha
in ol ing
he
e b
u é i
‘ha e’.
We
may assume
he
cons uc ion
u iu
duonos
algy i
o
ha e
o iginally
con ained
a
da i e
o
pu pose
as
well:
‘I
ha e
some
b ead
o
ea ’
—
‘I
ha e
o
ea
b ead’.
In
ac ,
he
o iginal
meaning
(p esupposing
he
segmen a ion
[[é iu
duonos]
algy i])
s ill
exis s
alongside
he
new
one
([ u i
[duonos
algy i]]).
Nowadays
he
meaning
o
a
modal
e b
(‘ha e
o,
mus ,
should’)
is
asc ibed
o
u é i
as
a
sepa a e
meaning
alongside
he
basic
possessi e
one,
and
in
iew
o
he
basic
seman ic
simila i y
o
‘habeo’
and
‘mihi
es ’
(which should
be
desc i-
bed
as
syn ac ic
con e ses,
as
emphasised
by
Ben enis e
1966)
i
should
be
possible
o
asc ibe
a
simila
modal
meaning
o
‘be’
as
well.
The
in e p e a ion
o
bi i
as
a
modal
e b
calls,
o
cou se,
o
a
u he
explana ion.
The
e b
u é i
co esponds
o
ou
no ions
abou
modal
e bs
because
i
displays
some
ea u es
cha ac e is ic
o
wha
is
desc ibed
as
he
Aux
componen
o
gene a i e
g amma ,
iz.
i
s ands,
as
i
we e,
apa
om
he
syn ac ic
s uc u e
o
he
sen ence
and
does
no
in luence
i .
In
his
sense,
i
is
simila
o
a
ense
and
mood
ma ke ,
and
indeed
ense
and
mood
a e
o en
exp essed
by
auxilia ies
as
well.
Thus,
we
may
speak
o
a
modal
auxilia y
he e.
Ob iously
he
case
o
bi i
is
di e en
because
i
seems
o
in luence
syn ac ic
s uc u e.
Howe e ,
he
equi emen s
and
con en ions
connec ed
wi h
he
gene a i e
mode
o
desc-
ip ion
a e
no
ele an
he e.
A
modal
e b
ypically
ep esen s
a
modal
p edica e
wi h
one
p oposi ional
a gumen ,
he
syn ac ic
ep esen a ion
o
which
is
an
embedded
clause.
I
will
assume
he e
ha
such
a
p edica e
can
in luence
he
assignmen
o
case
o ms.
I
should
be
no ed
ha
such
an
assump ion
is
necessa y
i
we
a e
o
accoun
o
he
syn ac ic
p ope ies
o
he
La ian
debi i e:
his
is
a
o m
he
unc ion
o
which
is
pu ely
modal,
ie.,
i
exp esses
a
modal
p edica e.
I
changes,
howe e ,
he
assignmen
o
case
o ms.
The
sys em
o
modal
p edica es
(as
dis inc
om
sen ence
modali y)
is
based
on
he
cen al
meanings
o
necessi y
and
possibili y.
Modal
e bs
p ope
end
o
be
specialised
in
C .
also
Russian
ne ezo
yumamb
maxue
xuueu,
Whe e
he
nega ed
subs an i e
e b
nem
has
used
wi h
he
ela i e
p onoun
in o
wha
is
now
conside ed,
in
his
pa icula
use,
o
be
a
p edica i e
exp ession
wi h
he
meaning
‘ he e
is
no
poin
(in
doing
some hing)’.
LITHUANIAN
ba i
WITH
THE
INFINITIVE
|
75
e ms
o
ei he
necessi y
o
possibili y.
These
wo
modal
meanings
a e,
o
cou se,
in e e-
la ed
in
se e al
ways
h ough
he
ac ion
o
he
nega ion.
In
pa icula ,
he
ollowing
equi-
alences
a e
o
be
obse ed:
‘Op’
=
‘0p’
(‘i
is
necessa y
ha
no -p’
=
‘i
is
impossible
ha
p’)
“O-p’
=
‘0p’
(‘i
is
no
necessa y
ha
no -p’
=
‘i
is
possible
ha
p’)
Though
languages
end
o
show
a ious
es ic ions
o
a
o mal
na u e
as
o
he
in oduc-
ion
o
a
nega ion
in
cons uc ions
wi h
modal
e bs,
he e
is,
in
p inciple,
a
possibili y
o
adding
a
nega ion
ei he
o
he
modal
e b
(EXTERNAL
NEGATION)
o
o
he
in ini i e
(INTERNAL
NEGATION),
as
illus a ed
by
he
ollowing
Li huanian
examples:
jis
negalia ei i
(‘=p’)
‘i
is
impossible
o
him
o
come
/i
is
impossible
ha
he
will
come’
Jisgalinea ei i
(‘0-p’)
‘i
is
possible
o
him
no
o
come
/i
is
possible
ha
he
will
no
come’
jis
negalinea ei i
(‘=0—p’)
‘i
is
impossible
o
him
no
o
come
/i
is
impossible
ha
he
will
no
come’
Wi h
exp essions
o
necessi y,
howe e ,
he
nega ion
may
o en
be
in e nalised,
i.e.,
i
may
o mally
be
a ached
o
he
modal
e b
while
being
seman ically
in e p e ed
as
belon-
ging
o
he
in ini i e,
e.g.
he
s uc u al
meaning
o
jis
ne u i
a ei i
is‘Op’,i.e.,
‘i
is
no
necessa y
o
him
o
come‘
=
‘he
need
no
come’,
bu
i
can
as
well
mean
‘O-p’,
i-e.,
‘i
is
necessa y
o
him
no
o
come’,
‘he
should
no
come’.
In
addi ion
o
he
modal
e bs
b ie ly
cha ac e ised
he e
we
may
no e
a
se ies
o
al e na-
i e
means
o
exp essing
necessi y
and
possibili y.
In
La in,
o
ins ance,
he
meaning
o
necessi y
is
o en
exp essed
by
means
o
a
cons uc ion
wi h
he
p edica i e
ge undi e:
Ka hago
delenda
es
“Ca hage
should
be
des oyed’.
A
s uc u al
and
unc ional
coun e -
pa
o
he
La in
cons uc ion
wi h
he
p edica i e
ge undi e
is
he
La ian
cons uc ion
wi h
he
p esen
passi e
pa iciple:
Siens
i
plaujams
‘ he
hay
mus
be
mown’.
The
La ian
cons uc ion
is
in e es ing
in
ha
i
may
se e
as
an
example
o
an
exp ession
ha
is
ague
be ween
necessi y
and
possibili y.
A
pa iciple
o ,
o
be
mo e
p ecise
(as
he
pu ely
pa i-
cipial
use
o
his
o m
is
bu
a ely
obse ed)
e bal
adjec i e
in
-ms
ma ks
a
gene al
p edisposi ion
o
an
objec
o
unde go
a
ce ain
ype
o
ac ion,
bu
he
p ecise
in e p e a-
ion
o
his
p edisposi ion
in
e ms
o
possibili y
(‘ ha
can
be
mown’)
o
necessi y
(‘ ha
mus
be
mown’)
is
con ex -dependen .
This
inde e minacy
subsis s
in
hose
cases
whe e
he
e bal
adjec i e
is
used
in
a
copula
cons uc ion,
and
Siens
i
plaujams
may
also
ha e
bo h
meanings,
ha
o
possibili y
and
ha
o
necessi y.
In
he
case
o
cons uc ions
wi h
he
e b
‘be’
and
an
in ini i e
o
pu pose,
a
simila
modal
agueness
is
a es ed
om
se e al
languages,
c .
Ge man
Die
F age
is
zu
lésen
‘The
p oblem
can
be
sol ed’
o
‘The
p oblem
mus
be
sol ed’.
The
specialisa ion
o
such
cons uc ions
as
exp essions
o
neces-
si y
is
p obably
associa ed
wi h
hei
inc easing
g amma icalisa ion.
I
is
p obably
in
his
way
ha
one
should
explain
he
specialisa ion
o
he
La in
cons uc ion
wi h
he
p edica-
i e
ge undi e
as
a
means
o
exp essing
necessi y.
The
La ian
cons uc ion
wi h
p esen

76
|
AXEL
HOLVOET
passi e
pa iciples
seem
o
be
unde going
a
simila
p ocess.
In
hose
cases
whe e
he
cons uc ion
wi h
a
passi e
pa iciple
is
copula ,
i
is
also
ague
be ween
possibili y
and
necessi y,
bu
when
i
ceases
o
be
copula
and
begins
o
unc ion
as
a
MODAL
CONSTRUC-
TION
p ope ,
only
he
necessi y
eading
is
e ained.
This
is
he
case
when
a
cons uc ion
o
his
ype
is
de i ed
om
an
in ansi i e
e b,
so
ha
he e
is
no
nomina i e
subjec
and
he
cons uc ion
canno
he e o e
be
in e p e ed
as
copula
any
mo e:
[k iens
jau
zindja,
ku a
baznica
inam
ejams
(P.
Rozi is)
‘Bu
e e yone
o
hem
know
o
which
chu ch
he
was
o
go’.
I is
possible
o
ind
mo e
examples
o
modal
cons uc ions
ha
a e
o iginally
ague
be ween
necessi y
and
possibili y
and
subsequen ly
unde go
specialisa ion*.
The
cons uc-
ions
unde
discussion
in
his
a icle
a e
one
o
hem.
3.
TWO
STAGES
IN
THE
GRAMMATICALISATION
OF
ba i
+
INFINITIVE
AS
A
MODAL
CONSTRUCTION
The
Bal ic
cons uc ions
wi h
ba i
and
he
in ini i e
seem
o
con i m
he
no ion
o
an
o iginally
undi e en ia ed
cons uc ion,
suscep ible
o
a
wo old
modal
in e p e a ion
(possibili y
o
necessi y)
de eloping
owa ds
a
modally
unambiguous
cons uc ion
ex-
p essing
only
necessi y.
Ins ances
o
he
meaning
o
possibili y
should
be
in e p e ed
as
esidual
and
imp oduc i e.
I
will
s a
my
exposi ion
wi h
La ian,
whe e
he
cons uc ions
discussed
he e
a e
mo e
equen
han
in
Li huanian
( hough
hey
also
show
aces
o
some
new,
speci ically
La -
ian
de elopmen s,
o
be
discussed
below).
In
Li huanian,
he
cons uc ions
wi h
o e
o ms
o
bii i
a e
now
excep ional;
hey
a e
well
a es ed
in
B e kiinas’
w i ings
(c .
Palio-
nis
1972:
125-126).
In
he
La ian
Academy
G amma
(1959:
622),
he
‘ ela ed
mea-
nings’
o
ine i abili y
and
impossibili y
a e
said
o
be
exp essed
by
he
p edica i e
in ini-
i e.
The
ollowing
examples
a e
ci ed:
(1)
A ak
pa
nakamo
pa asa i
...
naudu
ne edzé .
‘Be o e
nex
sp ing
he e
is
no
hope
o
seeing
any
money’.
(2)
Jums
manu
ilgu
mizam
nesap as .
(Rainis)
“You
will
ne e
unde s and
my
longing’.
I
is
clea
ha
we
a e
no
dealing
wi h
wo
ela ed
bu
dis inc
cons uc ions
he e,
as
he
o mula ion
o
he
Academy
G amma
seems
o
sugges .
Two
di e en
o mula ions
a e,
o
cou se,
possible,
as
he
impossibili y
o
p
may
also
be
o mula ed
as
he
ine i abili y
o
—p.
Now
i
we
assume
ha
he
cons uc ions
ci ed
he e
a e
ins ances
o
bi
wi h
he
in ini i e
a he
han
o
independen
p edica i e
in ini i es,
hen
he
occu ence
o
he
nega ion
wi h
he
in ini i e
sugges s
ha
we
a e
dealing
wi h
an
in e nal
nega ion,
and
ha
‘0p’
(‘i
is
impossible
ha
p’)
is
exp essed
as
‘O—p’
(‘i
is
necessa y
ha
no -p’),
and
he
O
cou se,
his
does
no
mean
ha
such
modally
ague
cons uc ions
mus
necessa ily
unde go
such
a
specialisa ion,
o
ha
hey
mus
necessa ily
de elop
in o
exp essions
o
necessi y
a he
han
possibili y.
The
Ge man
‘modal
in ini i e’
(is
zu
/ésen),
o
ins ance,
shows
no
signs
o
unde going
an
e olu ion
in
ei he
di ec ion:
i
appea s
o
be
qui e
s able
as
an
exp ession
o
ei he
possibili y
o
necessi y
(c .
Van
de
Auwe a
&
Plungian
1998:
100-103).
LITHUANIAN
ba@ i
WITH
THE
INFINITIVE
|
77
meaning
o
he
e b
ba
in
[BUT]
+
NEG-INF
is
hus
ha
o
necessi y,
no
possibili y.
(1)
and
(2)
hus
ep esen
he
same
cons uc ion
exp essing
necessi y
as
(3):
(3)
Ai ecais
balelin,
Te
manam
é am
bi ,
Te
manim
zi gu
pi k ,
Te
a es
liga inu.
‘Oh,
my
elde
b o he ,
you
should
be
a
a he
o
me, you
should
buy
me
a
ho se
and
b ing
[me]
a
b ide’
(BW
13796)
When
he
nega ion
is
added
o
he
modal
e b
bi ,
one
may
heo e ically
expec
wo
meanings:
‘—Up’
o ,
wi h
lowe ed
o
in e nalised
nega ion,
‘O—p’.
Ac ually
only
he
la e
meaning
is
a es ed
in
hose
cases
whe e
he
e b
is
o e ly
exp essed:
(4)
Ne ienam
s esam
nebij
edzé ,
ka
Slaucéja
madama
éda
{...}
(R.
Blaumanis)
‘No
s ange
was
supposed
o
see
Madame
Slaucéja
ea ing
[...]’.
(5)
Te
nebiis
auda ,
mila
d esle...
(Rainis,
ci ed
a e
Be gmane
e.a.,
1959:
623)
“You
mus /should
no
weep,
my
dea
soul’.
Fo mal
a achmen
o
he
nega ion
o
he
in ini i e
(* e
bis
ne auda )
does
no
seem
o
be
a es ed,
and
i
is
e iden ly
supe luous
because
he
co esponding
meaning
is
exp es-
sed
by
he
cons uc ion
wi h
nega ion
lowe ing
( e
nebiis
auda ).
On
he
o he
hand,
he
la e
cons uc ion
does
no
ha e
any
equi alen
in
he
p esen
ense:
he e
is
no
cons uc ion
*Te
na
auda
‘you
should
no
weep’.
In
he
meaning
we
should
expec
such
a
cons uc ion
o
ha e
(nega ion
lowe ing
being
aken
in o
accoun ),
he
u u e
o
bii
is
always
used.
He e,
e iden ly,
a
specialisa ion
o
ense
o ms
has
aken
place
in
he
nega ed
cons uc-
ions.
The
use
o
he
u u e
biis
in
cons uc ions
like
Te
nebiis
zag
“Thou
shal
no
s eal’
is
no
mo i a ed
seman ically,
because
in
a
gene alising
s a emen
o
deon ic
necessi y
his
modali y
is
no
no mally
iewed
as
a ising
in
he
u u e
while
no
encompassing
he
momen
o
speaking;
he
u u e
o
he
modal
e b
is
no mally
used
only
in
he
case
o
dynamic
modali y
(c .
You
should
no
s eal
s
You
will
ha e
o
s eal).
E iden ly
he
nega ed
u u e
has
supe seded
he
nega ed
p esen
ense
in
deon ic
meaning.
In
he
p esen
ense,
only
an
in e nal
nega ion
is
possible.
This
dis ibu ion
o
deon ic
and
dynamic
meanings
has
e iden ly
sp ead
om
he
nega ed
cons uc ions
o
he
a i ma i e
ones,
so
ha
Je
bis
s ada
“You
should
wo k’
is
now
also
specialised
in
deon ic
meaning?.
Li huanian
seems
o
ha e
he
same
pa e n
wi h
nega ion.
The
nega ion
can
be
added
o
bii i
as
a
modal
e b,
he
meaning
is
deon ic
and
ecei es
he
in e p e a ion
‘O—p’
(wi h
nega ion
lowe ing).
The whole
cons uc ion
is
no
p oduc i e
any
mo e
in
mode n
Li hu-
anian.
The
ollowing
example
has
nebu o
‘one
shouldn’
ha e’:
(6)
Kaip
nusédau
nuo
Zi gelio,
nebu o
adin i,
kaip
jéjau
j
seklycig,
nebu o
sodin i.
(Amb azas
1995:
94)
=
In
olde
La ian
g amma s,
bis
wi h
he
in ini i e
is
some imes
desc ibed
as
a
u u e
impe a i e,
e .
And ono
(1999:
161-163).
78
|
AXEL
HOLVOET
‘When
I
dismoun ed
om
my
ho se,
you
shouldn’
ha e
in i ed
me
o
come
in,
when
I
en e ed
you
house,
you
shouldn’
ha e
gi en
me
a
sea ’.
When
he
nega ion
is
in e nal,
he
meaning
is
dynamic,
and
‘O—p’
is
used
o
con ey
impossibili y
(“0p”).
The
nega ion
is
hen
added
o
he
in ini i e,
and
he
e b
‘be’
has
no
su ace
ealisa ion:
he e
is
he e o e
no
possibili y
o
ense
a ia ion.
The
cons uc ion
is
desc ibed
as
one
wi h
a
p edica i e
in ini i e
in
Li huanian
g amma
(c .
Ul ydas,
ed.,
1976:
323):
(7)
Sese ys
a soko:
jo
(Zalcio)
nebis a y !
(Saloméja
Né is,
ci ed a e
Ul ydas,
ed.,
1976:
323)
‘The
sis e s
sp ang
back:
he
g ass-snake
was
no
o
be
d i en
away
any
mo e’.
(8)
A kliui
nepa ilk i
ezimo.
‘The ho se
is
unable
o
d aw
he
ca ’.
The
impossibili y
eading
can,
o
cou se,
apply
only
i
he
nega ion
is
a
VERB
PHRASE
NEGATION,
no
i
i s
scope
does
no
encompass
he
e b
ph ase
a
all.
In
(9),
o
ins ance,
he
nega ion
is
in e nal
in
he
sense
o
no
belonging
o
he
modal
e b
( he
unexp essed
e b
bii i),
bu
i
has
only
he
ad e bial
isuome
wi hin
i s
scope,
no
he
e b:
(9)
Ne
isuome
gi
i
mums
jungg
ilk i.
(A.
Vienuolis,
ci ed
a e
Ul ydas,
ed.,
1976:
616)
‘And
we
will
no
always
ha e
o
pull
he
yoke
ei he ’.
While
he
meaning
o
impossibili y
can
be
de i ed
om
ha
o
necessi y
in
he
way
desc ibed
abo e,
possibili y
ob iously
canno .
The e o e
he
modal
e b
ba (i)
‘mus ’
canno
be
con ained
in
cons uc ions
o
he
ype
namai
ma y i,
o
else
his
e b
mus
also
be
asc ibed
ano he
meaning,
iz.
possibili y.
Bu
he e
is
p obably
no
eason
o
do
so.
The
ype
namai
ma y i
has
i s
syn ac ic
as
well
as
i s
seman ic
peculia i ies:
he
mos
impo an
o
hese
is
he
consis en
e en ion
o
he
nomina i e
(Amb azas
1995:
90).
As
he
nomina i e
ep esen s
an
olde
s age
in
he
de elopmen
o
he
cons uc ions
we
a e
dealing
wi h
he e,
i
seems
legi ima e
o
assume
ha
he
modal
meaning
o
his
cons uc-
ion
also
ep esen s
an
olde
s age
in
he
p ocess
o
i s
g amma icalisa ion
as
a
modal
exp ession.
Vagueness
be ween
possibili y
and
necessi y
is
no
equen
in
specialised
lexical
exp es-
sions
o
modali y,
i.e.,
in
modal
e bs.
Bu
i
does
seem
o
be
cha ac e is ic
o
a ious
cons uc ions
in
he
ini ial
s age
o
hei
g amma icalisa ion
in
modal
unc ion.
I
was
s a ed
abo e
ha
he
ype
y a
namai
ma y i
/
s a y i
mus
ha e
o igina ed
as
an
exis en ial
cons uc ion:
‘The e
is
a
house
o
building/seeing’.
As
long
as
he
cons uc ion
con inued
o
be
exis en ial,
i
was
also
ague
wi h
ega d
o
modal meaning:
y a
namai
s a y i
could
mean
‘ he e
is
a
house
o
be
buil ’,
whe eas
y a
namai
ma y i
could
a
he
same
ime
be
in e p e ed
as
‘ he e
is
a
house
o
be
seen’,
i.e.,
“a
house
can
be
seen’.
A
a
la e
s age,
when
he
modal
meaning
p e ailed
and
namai
s a y i
was
in e p e ed
as
an
embedded
in ini i al
clause
(wi h
a
nomina i e
objec ),
his
ambigui y
was
elimina ed
in
a ou
o
he
meaning
o
necessi y.
As
a
pa allel,
we
may
ci e
he
La ian
cons uc ions
wi h
p esen
passi e
LITHUANIAN
bi i
WITH
THE
INFINITIVE
|
79
pa iciples
men ioned
abo e:
in
hei
o iginal
shape
(as
copula
cons uc ions)
hey
a e
ague
be ween
possibili y
and
necessi y,
bu
a
a
mo e
ad anced
s age
o
g amma icalisa-
ion
hey
become
specialised
in
he
meaning
o
necessi y.
The
ype
namai
ma y i
is
hus
a
esidual
one,
and
his
ac
is
also
e lec ed
in
he
lexical
es ic ions
o
which
i
is
subjec :
he
ype
now
seems
o
be
es ic ed
o
a
small
g oup
o
e bs
o
senso y
pe cep ion®.
Fo
La ian,
Endzelin
(1951:
992)
ci es
one
example
ha
can
be
compa ed
o
he
Li huanian
ype
namai
ma y i:
(10)
pazi
man(i)
b alu
masas:
pilni
pi ks i
g edzen inu
BW
6259,
4
‘I
can
ecognise
my
kinsmen’s
sis e s:
hei
inge s
a e
ull
o
ings’.
This
ype,
like
he
co esponding
Li huanian
cons uc ion,
is
a
elic
om
he
unc io-
nal
poin
o
iew:
he
o iginal
meaning
o
possibili y
is
e ained
in
he
case
o
one
(o
pe haps
a
ew)
e bs
exp essing
senso y
pe cep ion
o
cogni ion.
In
La ian,
he
cons -
uc ion
has
comple ely disappea ed
om
he
mode n
language,
whe eas
i s
Li huanian
coun e pa ,
hough
now
subjec
o
lexical
es ic ions,
is
s ill
ully
ali e.
Syn ac ically,
he
La ian
cons uc ion
has
been
assimila ed
o
he
p oduc i e
ype
exp essing
necessi-
y,
and
he
nomina i e
has
been
eplaced
wi h
he
accusa i e,
which
canno
be
seen
in
(10)
bu
is
e ealed
by
he
o he
a ian s
o
BW
6259
(e.g.,
pazi
bija
kunga
iju
‘one
could
ecognise
he
squi e’s
ba n’;
pazi
man
b aJa
masu
‘I
can
ecognise
my
kinsman’s
sis e ’)
as
well
as
by
o he
con ex s
(pazi
bija
po a
egli
melnajam
skujinam
‘one
could
ecognise
he
i - ee
g owing
on
ma shy
soil
by
i s
black
needles’).
This
assimila ion
has no
aken
place
in
Li huanian,
whe e
he
nomina i e
is
e ained
in
he
esidual
ype
exp essing
possibili y.
4.
PECULIAR
DEVELOPMENTS
IN
LATVIAN
Example
(10)
ci ed
abo e
is
an
exac
coun e pa
o
he
esidual
Li huanian
ype
exp es-
sing
possibili y.
Bu
Miihlenbach
(Endzelin
&
Mii hlenbach
1928:
194)
and
Endzelin
(1951:
992)
ci e
a
whole
se ies
o
examples
allegedly
co esponding
o
he
ype
namai
ma y i.
Among
hese,
only
(10)
is
a i ma i e
and
can
be
plausibly
in e p e ed
as
exp es-
sing
possibili y.
The
emaining
examples
a e
all
nega i e,
so
ha
he e
is
no
eason
o
se
hem
apa
om
cons uc ions
like
(1-2),
which
ac ually
ep esen
he
meaning
o
neces-
si y.
The
eason
why
hey
a e
men ioned
oge he
wi h
he
use
o
pazi
in
(10)
is
ha
hey
also
con ain
e bs
o
pe cep ion.
On
close
sc u iny,
howe e ,
he
La ian
cons uc ions,
:
In
Sla onic,
ce ain
aces
o
he
es ic ion
o
by i
+
in ini i e
as
an
exp ession
o
possibili y
o
e bs
o
senso y
pe cep ion
can
be
ound
as
well.
Fo
Old
Russian,
c .
Bo ko skij,
ed.
(1978:
280).
In
Polish,
he
o iginal
in ini i es
widaé
‘one can
see’,
s ychaé
‘one
can
hea ’,
czué
‘one
can
smell’,
znaé
‘one
can no ice’
a e
s ill
used
wi h
byé
‘be’
(omi ed
in
he
p esen
ense),
as
in
widaé
bylo
‘one
could
see’
e c.
They
show
he
eplacemen
o
he
o iginal
nomina i e
wi h
he
accusa i e,
e.g.,
s ychaé
muzyke
‘music
is
hea d,
can
be
hea d’.
Synch onically,
howe e ,
hey
a e
no
ea ed
as
in ini i es
any
mo e,
pa ly
because
he
e bs
om
which
hey
de i e
(widac,
s ychac)
ha e
allen
in o
disuse.
In
con empo a y
Polish
g amma ,
widaé
and
s ychaé
a e
desc ibed
as
non-in lec ing
e bs using
byé
as
an
auxilia y
o
ense
and
mood.
86
|
AXEL
HOLVOET
(b)
zi gs
ezumu
ne
pa ilk
(wi h
non-p ocli ic
nega ion)
‘The
ho se
canno
d aw
he
ca ’.
I
he e o e
assume
(a)
and
(b)
o
be
gene ically
ela ed,
i-e.,
I
assume
(b)
o
ha e
a isen
om
(a),
a
p ocess
o
which
ce ain
o he
cons uc ions,
no
gene ically
ela ed
o
(a),
could
ha e
con ibu ed
as
a
as
he
o mal
make-up
o
he
cons uc ion
and
i s
unexpec ed
oli ional
eading
a e
conce ned.
One
u he
poin
should
be
made
he e
wi h
espec
o
ela i e
ch onology.
No e
ha
in
o de
o
he
p ocess
o
subs i u ion
o
@,,,,,
o
@,,,
o
occu ,
he
syn ac ic
posi ion
o igi-
nally
ese ed
o
he
nomina i e
exp essing
he
objec
o
he
ac ion
(as
in
Li h.
namai
ma y i)
had
o
be
eed
i s .
In
o he
wo ds,
he
eplacemen
o
he
o iginal
nomina i e
esidually
e ained
in
he
Li huanian
ype
man
(y a)
namai
s a y i
wi h
an
accusa i e
(as
in
Li h.
man
(y a)
namus
s a y i)
had
been
comple ed
be o e
a
new
nomina i e
noun
ph ase
was
in oduced.
This
means
ha ,
syn ac ically,
no
cons uc ion
o
he
ype
namai
ma y i
as
a
dis inc
esidual
ype
alongside
he
g amma icalised
exp essions
o
necessi y
could
ha e
been
e ained
a
he
ime
when
he
eanalysis
@,,,
as
@,,,,
occu ed.
Seman ically,
some
aces
o
he
o iginal
agueness
o
he
ype
ep esen ed
by
Li h.
namai
ma y i
seem
o
ha e
been
e ained,
i
pazi
in
(10)
exp esses
possibili y.
As
no ed
abo e,
he
o iginal
cons uc ion
wi h
he
da i e
was
e ained
alongside
he
new
one.
Wha
we
obse e
is
he
p ocess
o
a
syn ac ic
s uc u e
spli ing
up
in o
wo
sepa a e
ones
a he
han
o
he
de elopmen
o
one
s uc u e
in o
ano he .
Fo
he
o iginal
cons -
uc ion
o
subsis ,
he
new
one
had
o
be
su icien ly
dis inc
om
i
o mally
and
unc io-
nally.
The
di e ences
could
be
summed
up
as
ollows:
(1)
case
ma king
(nomina i e
as
agains
da i e),
(2)
s a us
o
nega ion
(non-p ocli ic),
(3)
seman ic
agueness
(impossibi-
li y
o
nega i e
oli ion
as
opposed
o
impossibili y
as
he
only
possible
eading),
(4)
pe haps
he e
migh
be
di e ences,
wi hin
he
gene al
meaning
o
impossibili y,
be ween
se e al
ypes
o
modali y
ha
could
pe haps
be
exp essed
by
he
new
ype
bu
no
by
he old
one
o
ice- e sa;
his,
howe e ,
mus
be
in es iga ed
sepa a ely.
5.
CONCLUSIONS
The
cons uc ion
discussed
in
his
a icle
is
acommon
ea u e
o
Li huanian
and
La ian.
I s
oldes
shape
is
e ained
only
in
Li huanian:
i
is
a
cons uc ion
wi h
a
nomina i e
subjec
used
wi h
he
subs an i e
e b
and
an
in ini i e
o
pu pose:
namai
(y a)
s a y i
‘ he
house
is
o
be
buil ’.
This
cons uc ion
was
modally
ague
be ween
possibili y
and
necessi y,
a
ea u e
equen ly
obse ed
in
modal
cons uc ions
based
on
‘be’
wi h
he
in ini i e
in
se e al
Eu o-
pean
languages.
The
use
o
his
cons uc ion
as
an
exp ession
o
possibili y
is
a
syn ac ic
elic,
e ained
mainly
in
Li huanian,
and
subjec
o
lexical
es ic ions
( e bs
o
pe cep ion,
as
in
namai
ma y i
‘ he
house
can
be
seen’).
In
he
cou se
o
he
u he
de elopmen
o
his
cons uc ion,
bii i
began
o
unc ion
as
a
modal
e b
exp essing
exclusi ely
necessi y,
and
his
is
he
only
p oduc i e
use
in
his o ical
imes.
The e
is
no
longe
a
possibili y
eading,
only
an
impossibili y
eading de i ed
om
necessi y
(‘O—p’
=
‘“—0p’).
A
syn ac ic
concomi an
o
his
p ocess
was
he
eanalysis
o
he
nomina i e
as
an
objec
o
he
in ini i e
and
i s
eplace-
men
wi h
he
accusa i e
( hough
Li huanian
has
esidually
e ained
he
nomina i e
objec ).

LITHUANIAN
bi i
WITH
THE
INFINITIVE
|
87
Wi hin
he
domain
o
nega ed
cons uc ions
exp essing
impossibili y,
La ian
also
shows
a
peculia
de elopmen
no
a es ed
in
Li huanian,
iz.
he
ise
o
cons uc ions
wi h
a
nomina i e
subjec
ins ead
o
a
da i e
subjec ,
e.g.,
zi gs
ezumu
ne
pa ilk
ins ead
o
zi gam
ezumu
nepa ilk .
This
cons uc ion
is
also
cha ac e ised
by
ce ain
peculia i ies
in
he
unc ioning
o
he
nega ion.
The
mechanism
o
i s
ise
is
no
clea .
Mos
o
he
cons uc-
ions ci ed
in
he
li e a u e
as
equi alen s
o
he
Li huanian
cons uc ion
exp essing
possi-
bili y
(namai
ma y i)
ac ually
belong
o
his
ype,
and
should
be
se
apa
(as
an
inno a-
ion)
om
he
Li huanian
cons uc ions
om
which
hey
di e
s uc u ally.
I
is
impo -
an
o
no e
his
in
connec ion
wi h
he
Li huanian
ype
namai
ma y i,
because
schola s
elying
o
hei
La ian
da a
on
he
publica ions
o
Mihlenbach
and
Endzelin
(c .,
e.g.,
Amb azas
1995:
90)
ha e
been
misled
in o
belie ing
ha
he e
is
no
undamen al
di e-
ence
be ween
he
Li huanian
and
La ian
cons uc ions
exp essing
(im)possibili y.
REFERENCES
AMBRAZAS,
V.
1995:
Lie u iy
kalbos
bend a ies
kons ukcijy
aida.
Lie u iy
kalbo y os
klausimai
33,
74-109.
AMBRAZAS,
V.
1997:
Lie u iy
kalbos
papildinio
a dininko
su
bend a imi
kilmé.
Lie u iy
kalbo y os
klausimai
38,
93-103.
ANDRONOV,
A.
1999:
,,Vajadzibas
iz eiksme“
la ieSu
alodas
g ama iskaja
adicija
(jeb
The
luc ua-
ing
o unes
o
he
La ian
debi i e).
Bal u
ilologija
8,
1998,
154-177.
AUWERA,
J.
VAN
DER,
PLUNGIAN,
V.
A.
1998:
Modali y’s
seman ic
map.
Linguis ic
Typology
2,
79-124.
BENVENISTE,
E.
1966:
‘E e’
e
‘a oi ’
dans
leu s
onc ions
linguis iques.
Bulle in
de
la
Socié é
de
Linguis ique
de
Pa is
LV,
1960;
he e
ci ed
a e :
P oblémes
de
linguis ique
géné ale,
Pa is:
Gallima d,
187-207.
BORKOVSKIJ,
V.
I.,
ed.
1978:
Hcmopuyecxan
2epammamuxa
pyccKozo
a3bika.
Cun axcuc.
Mpoc oe
npewioxenne,
Mocksa:
Hayka.
BRUGMANN,
K.
1916:
Ve gleichende
Lau -,
S ammbildungs-
und
Flexionsleh e
[...]
de
indoge mani-
schen
Sp achen.
I.
Leh e
on
den
Wo o men
und
ih em
Geb auch.
3/2,
S assbu g:
T iibne .
ENDZELINS,
J.
1951:
La iesu
alodas
g ama ika,
Riga:
La ijas
als s
izde nieciba.
ENDZELINS,
J.,
MUHLENBACHS,
K.
1928:
La iesu
g ama ika.
4.
iespiedums,
Riga:
Val e s
un
Rapa.
FRAENKEL,
E.
1928:
Syn ax
de
li auischen
Kasus,
Kaunas:
Vals ybés
spaus u é.
HOLVOET,
A.
1995:
Inde ini e
ze o
subjec s
in
La ian.
Linguis ica
Bal ica
4,
153-161.
MIKLOsICcH,
F.
1926:
Ve gleichende
G amma ik
de
sla ischen
Sp achen.
IV.
Syn ax.
Manuld uck
de
E s ausgabe
on
1868-1874,
Heidelbe g:
Ca l
Win e .
PALionIs,
J.
1972:
Ube
einen
Typ
des
In ini i sa zes
im
Li auischen.
Bal is ica,
1
p iedas,
125-129.
POTEBNJA,
A. A.
1958:
H3
zanucox
no
pyccxoil
zepammamuxe.
Tom
I-II.
O6maa
penaKuus...
B.
H.
Bopxoscxo o,
Mocksa:
Tocygapc seHHoe
yucOHO-nema o muccKoe
H3AaTCMBCTBO.
TIMBERLAKE,
A.
1974:
The
nomina i e
objec
in
Sla ic,
Bal ic
and
Wes
Finnic,
Miinchen:
Sagne .
(Sla is ische
Bei age,
82.)
U ypas,
K.
ed.
1976:
Lie u iy
kalbos
g ama ika
3.
Sin aksé,
Vilnius:
Mokslas.
Axel
Hol oe
Gau a
2001
05
26
Lie u iy
kalbos
ins i u as
An akalnio
g. 6,
2055
Vilnius,
Lie u a
[email p o ec ed]