?tekauer Pavol, Lieber Rochelle. Handbook of Word-Formation
Подождите немного. Документ загружается.
ONOMASIOLOGICAL APPROACH TO WOR
D
-
FORMATION
209
D
riven and Verspoor (1998) review the names for the concept of
C
ELL
U
LA
R
PHONE
,
demonstrating that AmE cellular (phone)
i
s a new coinage based on the
partiality relation in the same way as BrE mobile phon
e
,
while carphon
e
i
sb
a
sed o
n
t
he conti
g
uit
y
relation. On the other hand, Germa
n
Hand
y
i
s a partialit
y
-based
lo
an
wo
r
d.
I
n his stud
y
of Camito-Semitic and
A
frican lan
g
ua
g
es, Ta
g
liavini (1949)
analyzes various names for the concept of (eye)
P
U
PIL
,
indicating nine basic ways of
m
otivation underlying the naming of this concept, including BALL/EGG/APPL
E
,
B
LACK
,
C
ENTER
,
STAR/LIGH
T
,
N
U
T/PIP/PEARL
,
MIRROR
,
SEE/LOOK
,
LITTLE
MAN/GIRL/BOY/P
U
PPE
T
,
and redu
p
lication
.
4
Koch demonstrates – b
y
usin
g
a sampl
e
of expressions from 27 lan
g
ua
g
es for the
c
once
p
ts o
f
TREE an
d
F
R
U
I
T
– that these desi
g
nations displa
y
certain re
g
ula
r
“patterns sub
j
ect to co
g
nitive constants that characterize two different t
y
pes of
TREE-FRUIT frames”
(
1999a: 345
)
.
B
rammesberger and Grzega (2001) discuss the manifold ways of naming
Y
O
U
N
G
FEMALE PERSON in the history of English, and Grzega (2001) examines the names
for
W
EDNE
S
DAY in Germanic dialects. Some papers in Blank and Koch (2003)
demonstrate the applicability of an
o
nomasiological approach to syntax.
3. THE
O
RETI
C
AL APPR
O
A
C
HE
S
3
.1 M
iloš
D
okulil
W
hile little known in the English-speaking countries, Dokulil’s position in the
c
ountries of Central and Eastern Europe may be compared to that of Marchand in
W
estern Europe. No wonder, his onomasiological theory of word-formation,
published as early as 1962 (and furthe
r
developed in Dokulil 1964, 1968 a-d, 1997),
i
s a pioneering, highly seminal work, presenting a unique, comprehensive theory of
word-formation in which he – lon
g
befo
r
e the
g
enerativists – discussed a
m
ultiplicit
y
of essential word-formation issues, such as the
p
lace of word-formation
i
n the s
y
stem of lin
g
uistics, the differences between morpholo
g
ical and word-
formation analyses, word-formation motivation, productivity, internal form of
a word, lexicalization, word-formation paradigms, and the notion of word-formation
t
ype. In terms of its significance for the development of word-formation theory, this
i
ngenious book is on a par with Marchand’s Categories (1960, 1969). Although his
bar
‘public house’ is enabled by our knowledge of
r
c
ontiguity between public houses and counters.
T
he concept of PUBLIC HOUSE constitutes a fra
m
eo
n
eo
f
w
h
ose s
ali
e
nt
e
l
e
m
e
nt
s
i
s
th
e
C
OUNTER. Since a protot
y
pical public house has a counter (i.e., bar), we call it
ba
r.
C
onti
g
uit
y
also
c
haracterizes some cases of word-formation, e.
g
.
lemon
ĺ
lemon t
r
ee
,
where the concept expressed
b
y
N
1
(lemon fruit) and the concept expressed b
y
N1 + N2 (lemon tree) are conti
g
uous (Koch 1999b:
158-159
)
.
4
Cf., for exam
p
le, A. Blank
(
2001
)
for a detailed a
n
alysis of Tagliavini’s study as well as those by
Koch (1999a) – the concepts of TREE and FRUIT, and Krefeld (1999) – the concept of HUMAN
BO
DY
.
2
1
0
PA
VO
L
Š
TEKA
U
ER
t
heory is illustrated with
C
z
ech exam
p
les its theoretical principles are of general
v
alidity.
W
ord-formation is conceived by Dokulil a
s
an ‘a
u
t
o
n
o
m
ous do
main
w
ithin th
e
system of linguistics’ (1997: 185). The cornerstone of his onomasiological theory of
wo
r
d
-f
o
rmati
o
n i
s
th
e
i
de
a
o
f
o
nomasiological categor
y
. Any act of naming an
o
b
j
ect is based on its reflection and processing in human consciousness.
O
nomasiolo
g
ical cate
g
ories are thus defined b
y
Dokulil as different t
y
pes of
structurin
g
the concept in view of its expression in the
g
iven lan
g
ua
g
e, i.e., the
e
ssential conceptual struc
t
ures establishing the basis for the act of naming. In
t
t
principle, they consist of two elements. The phenomenon to be named is first classed
with a certain conceptual group and functions as
o
nomasiological bas
e
. Then
,
within
t
he limits of this group, it is determined by an onomasiological mar
k
.
For exam
p
le,
t
he onomasiological base of
b
lackberr
y
i
s
b
err
y
(because the concept of
B
ERRY i
s
c
ommon to the whole conceptual group of various
be
rr
ies
). Its onomasiological
m
ark i
s
black.
While one can trace an analogy
w
ith Mar
c
han
d
’
swo
r
d
-f
o
rmati
o
n
s
y
nta
g
ma, anal
y
sed as
dete
r
minant
-
dete
r
minatum
, Dokulil’s terms
p
ut em
p
hasis on
t
he level of conceptual processin
g
.
W
hile base is alwa
y
s simple (an
y
differences concern the level of abstraction),
m
ark may be either simple or compound. A simple mark within the limits of the
c
onceptual cate
g
or
y
of
SU
B
S
TAN
C
E is Qualit
y
(
black
berry)
k
k
5
o
r A
c
ti
o
n
co
n
ce
i
ved
without regard to its Ob
j
ect (
wo
r
k
er). Examples of a compound mark include
kk
woodcutte
r
,
where the Ob
j
ect of Action is specified, and policeman illustratin
g
a non-Actional relation. The previous examples also indicate that the two elements
o
f mark
,
i.e.
,
the
determining
an
d
th
e
g
determined
elements, may but need not be
d
e
xplicitly expressed. In Dokulil’s view, the basic types of onomas
i
ological structure
c
an be determined according
t
o the categorial nature (
S
U
BTANCES
,
A
CTION
,
Q
UALITY
,
C
IRC
U
MSTANCE) of its polar members, i.e., according to the base and the
determining element of mark, called
motive
.
For example, a concept of the category
of
SU
BSTANC
E
is determined by its relation to a concept of the category of (a)
SU
BSTANC
E
(
policeman
(
(
)
,
(
b
)
Q
UALITY
(
b
lackberr
y
)
,
(
c
)
A
CTION
(
teache
r
)
,
(
d
)
CO
N
CO
MITANT
C
IR
CU
M
S
TAN
CE
(
evenin
g
paper). Other onomasiolo
g
ical structure
ty
pes are determined analo
g
icall
y
. These t
y
pes can stand for the multiplicit
y
of
semantic
r
elations
,
includin
g
the Bearer of Qualit
y
(
blackboard
), Agent (
d
d
teache
r
)
,
Instrument of Action
(
excavato
r
), Patient (
prisoner
(
(
), Result of Action (
print-out
(
(
),
t
e
t
c.
6
A certain structure ma
y
be realised b
y
several namin
g
units (NUs),
e
mphasizing its different aspects (compare
hot
-
house
,
glass-house
,
green-hous
e
)
.
Dokulil distinguishes three
O
nomasiological categories
.
The basic type
d
i
scussed
a
bove
i
sc
all
ed
Mutational
(
or, Relational). In this case, an ‘ob
j
ect’ of one
c
onceptual cate
g
or
y
is characterized (and n
a
m
ed) accordin
g
to its direct or mediated
r
elation to an ‘object’ of the same or some other conceptual category.
I
n th
e
Transpositional typ
e
,
the phenomenon, usually conceived as a mark,
de
p
endent on a
S
U
BSTANCE, is abstracted from all the
p
henomena u
p
on which it
5
My examples, if no suitable English equival
e
n
ts to the Czech exam
p
les are available.
6
Dokulil (1962)
g
ives a hi
g
hl
y
fine-
g
rai
n
ed c
la
ss
ifi
c
ati
o
n
o
f th
ese
r
e
lati
o
n
s.
ONOMASIOLOGICAL APPROACH TO WORD
-
F
ORMATION
211
o
b
j
ectively depends, and is viewed as a
n
independently existing phenomenon, fo
r
e
xample, the ob
j
ectification of Quality
(
r
apid – rapidit
y
) and the ob
j
ectification of
Action
(
fall
((
V
l
l
–
fall
N
l
l
).
Th
e
M
odi
f
icational t
y
pe
i
s based on addin
g
a modif
y
in
g
feature, for example,
diminutives
(
d
o
g
– do
ggy
), au
g
mentatives (a bi
g
do
g
), chan
g
e of
g
ender (
waite
r
–
wait
r
ess
), names of the young (
fox-cub
(
(
)
, collectiveness
(
mankind
), measure/degree
d
d
(
the tallest
).
t
3.2 Ján Horecký
A ma
j
or step in the development of onomasiological theory of word-formation is
Horecký’s
multi
-
level model
o
f word-formation (1983, 1989), including an ob
j
ect of
e
xtra-lin
g
uistic realit
y
, the pre-semantic (conceptual), semantic, and formal levels.
Th
e
pre-semantic level
i
s constituted b
y
l
o
g
ical predicates. Some of the lo
g
ical
predicates are expressed as
semantic ma
r
ke
r
s
.
Horeck
ý
’s semantic level is carefull
y
e
laborate. He provides an inventory of sema
n
ti
cd
i
s
tin
c
ti
ve
f
e
a
t
ures, analyzes their
t
t
r
elations, and proposes their hierarchical organization.
A
t the to
p
of the semantic
m
arker hierarchy, there are categorial
ma
r
ke
r
s
(e.g.,
Substance, Quality, Agent names, names of Relations), which, due to thei
r
g
rammatical nature, are part of the formal onomasiological level and represent the
o
nomasiologial base. At a lower level, the
i
dentification
ma
r
ke
r
s
(
or, archisemes
)
r
epresent
g
enus proximum. The
y
capture a propert
y
common to all of the meanin
g
s
o
f a particular namin
g
unit. The next lower level is constituted b
y
s
peci
f
ication
ma
r
ke
r
s
.
The formal facet of linguistic sign
i
s composed of the onomasiological,
o
nomatological, and phonological structures. The onomasiological structure consists
o
f a base and a mark. The base also expresses relevant grammatical categories,
i
ncluding a word-class. The onomatological level functions as both inventory of
m
orphemes, and at the same time, it linguistically expresses the base and mark.
Finall
y
, the phonolo
g
ical level determines the specific form of morphemes and othe
r
phonolo
g
ical features.
An important part of Horeck
ý
’s onomasiolo
g
ical theor
y
is his classification of
m
eanings. In (1994), he dis
t
inguishes four types of
m
eaning of a naming unit: (i)
c
ategorial meaning; (ii) invariant meaning, (iii) specific meaning, and (iv) lexical
m
eaning. The first three meanings as a whole are labeled as the ‘structural’ meaning
(given by the interrelation between onomasiological base and mark), and underlie
t
he lexical meaning. For illustration, the respective meanings of Sl.
t
r
etina
(third
N
d
d
)
are as follows:
(
i
)
desubstantival noun;
(
ii
)
is defined as ‘abst
r
act quality defined by
the strin
g
of semantic features –HUM
–
CONCR –QUAL’;
(
iii
)
‘a third
p
art of
s
omethin
g
’, (iv) ‘one part of hocke
y
match’ (as one of its lexical meanin
g
s).
3.3 Pavol
Š
tekauer
Štekauer’s cognitive onomasiological
t
heory (Štekauer 1996, 1998, 2001b) was
inspired by Dokulil’s idea of
onomasiological structure and, primarily, by Horecký’s
f
2
12 PA
VO
L
Š
TEKA
U
ER
m
ultilevel model of linguistic sign (1983, and Horecký et al., 1989). At the same
t
ime, it responds to the one-sided formalism of the mainstream generative word-
f
o
rmati
o
n
.
7
The general linguistic background is that of the functional-structural
approach of the Prague School of Linguistics. Therefore, the form-meaning unit
y
,
i
.e., the bilateral nature of morphemes
i
s regarded as the fundamental principle.
3.3.1 Word-formation as an independent componen
t
The basic scope and principles of word
-
f
o
rmati
o
n
c
an
be de
fin
ed
a
s
f
o
ll
ows:
(1) Word-formation deals with productive and rule-
g
overned patterns
(
word-formation t
y
pes and rules, and morpholo
g
ical t
y
pes) used to
generate motivated naming unit
s
8
in response to the specific naming
n
eeds of a
p
articular s
p
eech co
m
munity by making use of word-
formation bases of bilateral naming units and affixes stored in the
Lexical Component.
The individual aspects of this definition are discussed below. The co
g
nitive
o
nomasiolo
g
ical theor
y
identifies word-formation as an
i
nde
p
endent com
p
onen
t
of
l
inguistics, as illustrated in Figure 1. The scheme re
p
resents a crucial triad o
f
relations
between extra-linguistic reality (object to be named), a speech community
(
represented by a ‘coiner’), and the wo
r
d-formation component, thus emphasising
the fact, ignored by the vast ma
j
ority of the mainstream word-formation theories,
that each act of naming responds
t
o a very real and specific
n
aming demand
o
n th
e
part of a member (members) of
s
peech communit
y
.
The notion of speech communit
y
s
hould not be taken absolutel
y
, i.e., there is hardl
y
an
y
word-formation process
which responds to a namin
g
demand of all the speakers of a particular lan
g
ua
g
e.
Rather, such a demand is closely connected with a limited number of ‘first-contact’
users; a coinage may or may not subsequently find a wider use.
T
he above-mentioned triad reflects the following principles:
(
a) It lays emphasis on the
active
r
ole
o
f language users in the process of giving
n
ames to objects instead of presenting word-formation as an impersonal syste
m
o
f rules detached from the objects named and from language users.
(
b) The naming act is not a purely linguistic act. Naming units do not come into
existence in isolation from factors
,
s
uch as human knowledge, human cognitive
abilities, experiences, discoveries of new things, processes, and qualities, human
i
ma
g
ination, etc. This position is in a
cco
r
d
an
ce w
ith K
oc
h’
s
i
de
a that th
e
o
nomasiolo
g
ical viewpoint is closer to that of the speaker as a lin
g
uistic
i
nnovator than the semasiolo
g
ical viewpoint (2001: 17). An ob
j
ect to be named
7
An important and most valuable e
x
c
eption to this formalism is Beard’s Lexeme-Morpheme Base
M
orpholog
y
(
1995
)(
cf. cha
p
ter... in thi
s
volume) which is, in effect, a variant of an onomasiological
approach to word-formation.
8
This term was first introduced by Mathesius (1975
)
.
In my approach, it sub
s
t
it
u
t
es
f
o
r th
e
t
e
rm
s
lik
e
w
ord
,
lexem
e
,
lexical unit
,
etc.
,
because of their inconsistent u
s
e
and various connotations in linguistic
l
iterature. “Naming unit” refers here to a compl
e
x unit generated by the Word-Formation Component.
ONOMASIOLOGICAL APPROACH TO WORD
-
F
ORMATION
213
Lexical Component
A
ctual naming units
A
ffix
es
(includin
g
all relevant
s
pecifications)
W
ord-formation Component
S
emantic level
Onomasiological level
O
nomatolo
g
ical level
Phonological level
Syntactic
C
omponent
i
s not named in isolation but is envisaged in relation to the existing objects. By
i
mplication, any naming act is necessar
i
ly preceded (or dominated) by a network
o
f ‘objectively’ existing relationships. By implication, the naming act is
a
co
g
nitive phenomenon
r
el
y
in
g
on the intellectual capacities of a coiner.
(
c
)
It stresses a close interconnection between
l
in
g
uisti
c
an
d
extra-lin
g
uisti
c
p
henomena.
EXTRA-LIN
GU
I
S
TI
C
REALITY
S
PEE
C
H
CO
MM
U
NITY
Conceptual level
Figure 1 Word-Formation Component and its relation to other components
The model represented in Figure 1 also in
d
i
c
at
es
a
d
ir
ec
t
co
nn
ec
ti
o
n
be
t
wee
n th
e
W
ord-Formation and the Lexical com
p
onents, and a mediated connection between
t
he Word-Formation and the Syntactic components. This makes this model different
from those theories that consider Word-Formation as a part of the Lexicon or a part
of
syntax.
9
The relation between the Word-Formation and the Lexical components is
9
Cf. Dokulil (1964) for an insightful discussion suppo
r
ting the separation of word-formation from syntax.
2
1
4
PA
VO
L
Š
TEKA
U
ER
based on their close ‘co-operation’. On the one hand, the Lexicon stores all naming
units (monemes and complex words, borrowed words, clippings and acronyms) as
well as affixes, and feeds the Word-Formation com
p
onent with word-formation
b
a
ses
an
d
affix
es
in a
cco
r
d
an
ce w
ith it
sn
e
eds. On the other hand, all new naming
units formed in the Word-Formation component are stored in the Lexicon.
I
t should be noted that word-formation focuses on the process of formin
g
isolated
namin
g
units rather than on usin
g
them (this bein
g
the scope of s
y
ntax).
A
namin
g
unit which falls within the scope of word-formation
must
be a structurally
t
anal
y
sabl
e
l
in
g
uistic si
g
n
,
and the si
g
n nature must also be an inherent feature of its
co
n
s
tit
ue
nt
s.
Thi
sco
n
d
iti
o
n i
s
i
de
nti
c
al
t
o that proposed by Marchand (1960: 2).
I
t i
s
a
ssu
m
ed
that
e
a
c
h act of naming
i
s preceded by
scanning
the lexical
g
c
omponent by a coiner. The scanning operation determines the next procedure.
Either a completely new naming unit is coined by taking the path of the Word-
F
ormation component; or, if a naming unit is found in the lexical component that
c
an
se
r
ve
a
s
a
b
a
s
i
s
f
or
s
emantic formation
,
it is the path of the lexical component
which is preferred (hence, two downward arrows from the ‘Conceptual level’ in
F
i
g
ure 1). B
y
implication, no new namin
g
units, formed accordin
g
to productive and
r
e
g
ular rules of word-formation are
g
enera
t
ed in the Lexicon (however, an
y
and all
l
ater semantic shifts and/or formal modifications (clipping, acronymization) of
naming units, productively formed in the
W
ord-Formation com
p
onent, take
p
lace in
t
he Lexicon
)
.
3.3.2 The act of naming
The following theoretical account of the act of naming interprets the model
g
raphically represented in Figure 1. For ease of understanding, the theory is
i
llustrated with an example of giving a name to the class of ‘persons whose
j
ob is to
drive a vehicle designed for the transportation of goods’
E
xtra-lin
g
uistic realit
y
vs. speech communit
y
A
s mentioned above, a speech-com
m
unity, through its diverse cognitive
m
m
activities
,
selects what there is in extra-linguistic reality
t
hat
dese
r
ves
a nam
e.
Thi
s
i
nterrelation between extra-linguistic rea
l
ity and a speech community predetermines
all the subse
q
uent ste
p
s within the act of naming
.
One of thousands of ‘objects’ of
e
xtra-linguistic reality that were conside
r
ed as worth naming sometimes in the past
was ‘a person whose
j
ob is to drive a vehicle designed for the transportation of
g
oods’
.
Conceptual level
The primar
y
task to be mast
e
red at the conceptual lev
e
l
is to anal
y
se the ob
j
ec
t
(in the broadest sense of the word) to be named; or better, a class of objects – a name
i
s not given to a single object but to a whole class of similar objects. This is the task
o
f th
e
conce
p
tual level which, based on the processes of generalisation and
abstraction, captures the prot
o
typical features of the class of ob
j
ects by means of
l
ogical predicates (simple declarative sentences, al
so c
all
ed
noemes
)
. A set of
O
NOMASIOLOGICAL APPROACH TO WOR
D
-
FORMATION
215
l
ogical predicates constitutes a
l
ogical spectrum
.
1
0
The logical spectrum is an
‘onomasiological answer’ to the gener
a
t
ion of complex words from a single ‘ill-
defined’ kernel sentence by transformationalists, and to the account of the internal
s
tructure of complex words b
y
a sin
g
le paraphrase b
y
lexicalists. The lo
g
ical
s
pectrum provides a more comprehensive view of the class of ob
j
ects to be named,
an
d
i
s
th
e
r
e
f
o
r
e
l
ess vo
l
u
ntari
s
ti
c.
I
n our example, the logical spectru
m
can be re
p
resented as follows:
(
2) The motivating Object 1 is
SU
B
S
TAN
C
E
1
.
A
S
U
BSTANC
E
1
is
H
u
man
.
The Human
p
erforms a
n
AC
TI
O
N
.
Th
e
AC
TI
O
N is the Human’s Profession (=A
g
ent).
The Human is an A
g
ent.
T
h
e
A
CTION
co
n
ce
rn
s
S
U
BSTANCE
2
(=Object
o
f Action
).
T
h
e
AC
TI
O
N
i
s based on an O
p
eration of
SU
B
S
TAN
CE
2
.
S
U
BSTANCE
2
i
s a class of
V
ehicles.
S
U
BSTANCE
2
i
s an Object of th
e
A
CTION performed b
y
S
U
BSTANCE
1
.
T
he Vehicles are designed for the Transportation of goods.
Et
c.
S
emantic level
The lo
g
ical spectrum is not a part of a lin
g
uistic si
g
n, and is lan
g
ua
g
e-
i
ndependent. Therefore, the individual lo
g
ical predicates of this
s
upralin
g
uistic level
m
ust be represented by
semes
1
1
c
onstituting the
semantic st
r
uctu
r
e
(meaning) of the
l
inguistic sign proper.
T
hus, the semantic level as the meaning facet of linguistic si
g
n maps the defining
s
pectrum, represented in (2), onto the semantic level of a new linguistic sign:
(
3
)
[+Material] [+Animate] [+Human] [+Adult] [+Profession] [+A
g
ent];
[
+Material
][
–Animate
][
+Vehicle
]
[+Transportation] [+Ob
j
ect of
O
p
eration] etc.
O
nomasiological level
At th
e
onomasiolo
g
ical level
,
one of the semes is selected to function as an
onomasiolo
g
ical
base
denotin
g
a class, to which the ob
j
ect belon
g
s, and one of the
m
is se
l
ec
t
ed
t
o
f
u
n
c
ti
o
n a
s
a
ma
r
k
t
hat s
p
ecifies the base. The mark can be, in
p
rinci
p
le, divided into the
d
etermining constituen
t
an
d
th
e
dete
r
mined constituent
.
T
he latter always stands for the category of Action in one of its three modifications
10
Cf. Horeck
ý
(1983).
1
1
Th
e
n
o
ti
o
n
o
f ‘
se
m
e
’ i
sco
n
ce
i
ved
h
e
r
e
in a
cco
r
d
a
n
ce w
ith th
e
n
o
ti
o
n
o
f ‘
se
ma
n
t
i
c
mark
e
r’
used
in th
e
theor
y
of componential anal
y
sis).
2
1
6
PA
VO
L
Š
TEKA
U
ER
(
Action
p
ro
p
er, Process and State
)
. The semantic relations between the base and the
two
mark
co
n
s
tit
ue
nt
sco
n
s
tit
u
t
e
an onomasiological structur
e
.S
ince this structure
c
onsists of semes which reflect, at the semantic level of a linguistic sign, the
r
espective logical predicates of the con
c
eptual level, it may be concluded that
the
onomasiological structure is a conceptual-semantic basis for the act of naming
.
To return to our example, it follows from the conceptual level analysis that a
g
ood candidate for the act of namin
g
seems to be an onomasiolo
g
ical structure in
which the base stands for an A
g
ent
12
(a class of Humans performin
g
the Action as
t
heir profession) of Action (determined constituent of mark) aimed at its Ob
j
ect, i.e.,
t
he class of Vehicles (determining constituent of mark):
(4) (Logical) Object ĸ Action – Agent
O
nomatolo
g
ical level
A
t this level, the onomasiological structure is linguistically expressed in
a
cco
r
d
an
ce w
ith th
e
M
orpheme-to-Seme-Assignment Principle
(
MSAP
).
1
3
In
particular, the individual constituents of onomasiological structure (its semes) are
assigned morphemes, in particular, wo
r
d-formation bases of naming units and
affixes stored in the Lexicon. The operation is based on
matching
the meaning facet
g
o
f a potential morpheme with the respective seme of the onomasiolo
g
ical structure.
The MSAP operates both horizontall
y
and verticall
y
.
V
ertically, it scans the Lexicon with regard to the lexical and affixal morphemes
t
hat can be retrieved to represent the semes of the onomasiological structure.
Horizontally, it reflects the semantic compatibility and formal combinability/
r
estrictions of the individual lexical and affixal morphemes
.
1
4
In our example, there are several option
s
at this level. Thus, Agent can be
e
xpressed,
inte
r
alia
,
b
y
-er
,
-ist
,
-ant
,
-ian
,
-man
,
because the meaning facet of each
o
f these morphemes can be represented as ‘A
g
ent’. The Action of operatin
g
the
SU
B
S
TAN
CE
2
c
an be expressed, for example, b
y
word-formation bases of namin
g
u
nit
s
d
rive, steer, operate
, because the meaning facet of
each of them matches with
f
t
he seme ‘Operation’. Finally, the (logical) Object can be represented by
t
r
uck
,
l
orr
y
,
and possibly some other word-formation bases, the meaning of which is
Vehicle. The selected o
p
tions in our
p
articular case are as follows:
(5) Ob
j
ect
ĸ
Action – A
g
ent
t
r
uck
d
r
ive
-
er
1
2
The majority of logical and semantic categories
h
ave been taken over from Hansen et al.
(
1982
)
.
13
In Štekauer (1998) and all the subsequent publi
c
ations I use the term Form-to-Meanin
g
-Assi
g
nment
Principle. However, I find the present labeling mo
r
e accurate as it is mor
p
hemes
(
rather than
p
ure
formal elements) that are assigned to semes.
1
4
This concept of onomasiological structure differs from
that of Horecký in three points. First, Horecký’s
m
o
nomasiological level is a formal level; second, all morphemes in the present model are stored in the
l
exicon
;
and
,
third
,
the function of MSAP is elaborated.
O
NOMASIOLOGICAL APPROACH TO WOR
D
-
FORMATION
21
7
T
here are at least two other basic representation types of the selected
o
nomasiological structure. First,
SU
B
S
TAN
C
E
2
may be backgrounded, in which case
t
he resulting naming unit may be, for example,
d
r
ive
r; and second, Action may be
back
g
rounded, which ma
yy
ield somethin
g
like
truckist
or
t
t
r
uckman
.
The fact that all namin
g
units are based on assi
g
nin
g
lin
g
uistic units to semes,
c
onstitutin
g
an onomasiolo
g
ical
s
tructure, makes it
p
ossible to dis
p
ense with the
t
raditional notions of word-formation processes, including compounding,
prefixation, suffixation, back-formation, and blending. The traditional classification
o
f word-formation processes is based on purely formal criteria, i.e., on the external
form of naming units. Consequently, it does
no
t r
e
fl
ec
t th
e
‘int
e
ra
c
ti
o
n
s
’ a
bove
an
d
within the Word-Formation component. Therefore, it appears to be more appropriate
t
o classify the processes leading to new naming units by reflecting the mutual
i
nteraction between the concept-
g
rounded onomasiolo
g
ical level and the morpheme-
g
rounded onomatolo
g
ical level, i.e. b
y
interrelatin
g
the supra- and the intralin
g
uistic
l
evels. This makes it possible to view all new namin
g
units as resultin
g
from the
i
dentically grounded acts of coining. Put differently, the generation of all naming
u
nits is
p
ut on a uniform basis. This a
pp
roach makes it
p
ossible to show what is, fo
r
example, common to ‘compounding’ and ‘suffixation’.
F
or illustration, they may express the sa
me onomasiological st
ructure of ‘Action
tt
–
Agent’ (the common feature) of, for example, ‘a person who frequently smiles’,
with the difference bein
g
in assi
g
nin
g
different morpheme t
y
pes: WF base + -
er
(
smile
r
)
vs. WF base + WF base
(
s
mile
p
erson
)
.
Similarl
y
, blendin
g
is, in principle, viewed as the same process of word-
formation as compounding. It is accounted for as a regular act of naming taking
p
lace in the Word-Formation com
p
onent. During this process, a particular
o
nomasiological structure is assigned two word-formation bases (e.g.,
slang
+
g
l
anguage). Such a naming unit is then formally reduced in an unpredictable (and
h
ence, irregular) way which cannot be
c
aptured by any productive Word-Formation
Type/Rule. Such a change necessarily takes place in the lexical component.
1
5
Phonolo
g
ical level
The final step in the act of namin
g
consists in phonolo
g
ical shapin
g
the new
n
aming unit in accordance with relevant phonologi
c
al rules. In our exam
p
le, it is the
assignment of the corresponding stress pattern.
(
10
)‘
t
r
uck
,
d
r
ive
r
3
.3.3 Onomasiolo
g
ical T
y
pes
Onomasiolo
g
ical T
y
pes result from the interaction between the onomasiolo
g
ical
and the onomatolo
g
ical levels. There are five possible relations between the two
l
evels that identify five basic Onomasiological Types.
1
5
Cf. Štekauer (1998) on the onomasiological account of back-formation and exocentric compounds.
2
1
8
PA
VO
L
Š
TEKA
U
ER
In
O
nomasiological Type 1
,
illustrated in the above-
g
i
ven exam
p
le, all three
o
nomasiological structure constituents, i.e., the base, the determining and the
d
etermined constituents of the mark
,
are linguistically expressed at the
o
nomatological level. Two more examples will illustrate the point (it should be
n
oted that the following onomasiological structures are based on logical spectra that
are not specified here for space reasons):
(
11
)
house-keepin
g
(
the Process of performing some Action aimed at an
Object)
:
Ob
j
ect – Action – Process
h
ouse keep -ing
(
12
)
signal-generator
(Instrument for an Action producing some Result)
r
R
esu
lt – A
c
ti
o
n – In
s
tr
u
m
e
nt
s
ignal generate -or
The onomasiolo
g
ical structure of
O
nomasiolo
g
ical T
y
pe 2 is binar
y
: the
d
eterminin
g
constituent of the mark is absent. However, this T
y
pe is extendable to
O
nomasiological Type 1.
(13) Action – Agent
w
r
ite
-
e
r
(
14
)
Action – Instrument
spinning wheel
A
crucial feature of the first two t
y
pes is that the Actional seme (the determined
c
onstituent of the mark) is morphematicall
y
expressed, which facilitates the
i
nterpretation of namin
g
units.
The onomasiological structure of
O
nomasiological Type 3 is ternary as in
Onomasiological Type 1, but the determined constituent of the mark is left
unexpressed at the onomatological level:
(
15) Result – Action – A
g
ent
novel 0
-
ist
(
16
)
Patient – Stat
e
– Evaluation
(
Diminutive
)
d
og 0 -ie
(
17) Temporal Stative – State – Patient
summe
r
0
house