by Leon Stassen
This map shows the possible relationships between the encoding of nominal and locational predicates. In particular, the map shows for each language whether nominal predications (such as John is a tailor) and locational predications (such as John is in Paris) can or cannot be encoded by the same strategy. In the terminology of Stassen (1997), a language is called a share-language if the encoding strategy for locational predications is (or can be) used for nominal predications, and a split-language if the encoding strategies for the two constructions must be different.
An obvious example of a share-language is English. As the above example sentences demonstrate, this language can use the lexical item be both as a nominal copula and as a locational support verb. In contrast to this, Mandarin is a split-language, as the copula and the locational verb are not the same.
(1) Mandarin (Li and Thompson 1977: 422; Li and Thompson 1981: 365)
a.
nei-ge
rén
shi
xuésheng
that-clf
person
cop
student
‘That man is a student.’
b.
Lısì
zái
hai-bian
Lisi
be.at
ocean-side
‘Lisi is by the ocean.’
| Go to map | ||
|---|---|---|
| Value | Representation | |
 |
Split (i.e. different) encoding of nominal and locational predication | 269 |
 |
Shared (i.e. identical) encoding of nominal and locational predication | 117 |
| Total: | 386 | |
As was shown in example (1), Mandarin is a split-language by virtue of the difference between the lexical items involved in nominal and locational predication. This type of split encoding is rather common. Spanish, with its difference between the copula ser and the locational verb estar, is another well-known case in point, as is Irish.
(2) Spanish (Max Kerkhof, p.c.)
a.
Julia
es
enfermera
Julia
cop.3sg.pres
nurse
‘Julia is a nurse.’
b.
Julia
está
en
Barcelona
Julia
be.3sg.pres
in
Barcelona
‘Julia is in Barcelona.’
(3) Irish (Greene 1966: 40, 43)
a.
is
múinteoir
é
cop
teacher
he
‘He is a teacher.’
b.
tá
sé
sa
tseomra
be.pres
he
in.the
room
‘He is in the room.’
However, this “lexical” form of split encoding is not the only way in which a language can achieve split-status. A second, also fairly frequent type of split encoding involves a contrast between a full supporting verb for locational predication and the absence of any overt linking item (a “zero copula”) for nominal predication. Examples of split languages in which this situation holds are Mokilese (Oceanic; Micronesia) and Waskia (Madang; Papua New Guinea).
(4) Mokilese (Harrison 1976: 142, 209)
a.
John
johnpadahk-men
John
teacher-indef
‘John is a teacher.’
b.
ih
mine
Hawaii
he
be
Hawaii
‘He is in Hawaii.’
(5) Waskia (Ross and Paol 1978: 11, 12)
a.
aga
bawa
taleng-duap
my
brother
police-man
‘My brother is a policeman.’
b.
kadi
mu
kawam
se
bage-so
man
art
house
in
stay-3sg.pres
‘The man is in the house.’
Finally, a third variant of split encoding is based on the difference between a full support verb for locative predicates and a verbal encoding for nominal predicates. Since there are not that many languages in which predicate nominals are treated as verbs anyway, it will be clear that this variant of split encoding will be less frequent than the other two. An example of this encoding option is the Philippine language Kapampangan: as is shown by sentences (6a-b), predicate nouns in this language have the same morphosyntactic properties as predicate verbs.
(6) Kapampangan (Mirikitani 1972: 137, 44, 72)
a.
tinerak
ya
ing
anak
ku
dance
3sg
art
child
my
‘My child danced.’
b.
mestro
ya
ing
lalaki
teacher
3sg
art
boy
‘The boy is a teacher.’
c.
ati
ya
ing
lalaki
king
eskwela
be
3sg
art
boy
at
school
‘The boy is in school.’
For the purposes of this map, the three possible forms which split encoding can take have been ignored. Thus, a language is rated as a member of type 1 if there is split encoding of any sort, regardless of whether this involves a lexical contrast, a zero-verb contrast, or a contrast between verbal and nonverbal encoding.
Parallel to split encoding, shared encoding of nominal and locational predication can be attested in three variants. Of these variants, the “lexical” form, which involves the use of the same lexical item for nominal copula and locational support verb, is by far the most frequent. In addition to English, some other examples of this variant are Miskito (Misumalpan; Nicaragua) and Luganda (Bantu; Uganda).
(7) Miskito (Anonymous 1985: 213; Conzemius 1929: 110)
a.
Giovanni
tuktan
sirpi
kum
sa
Giovanni
child
small
one
cop.3sg.pres
‘Giovanni is a small child.’
b.
aisi-kam
bara
sa
father-your
here
be.3sg.pres
‘Your father is here.’
(8) Luganda (Ashton et al. 1954: 434, 82)
a.
Mukasa
n-ange
tu-li
babazzi
Mukasa
and-1sg
1pl.pres-cop
carpenters
‘Mukasa and I are carpenters.’
b.
omugaati
gu-li
mu
kabada
loaf
3sg.pres-be
in
cupboard
‘The loaf is in the cupboard.’
The other two possible forms of shared encoding are rather uncommon. This is due to the fact that, for locational predication, the use of a full locational support item is the overwhelmingly preferred option (see Stassen 1997: 55-61). Thus, we only rarely find that a language has share-status on the basis of a zero-zero encoding. One such case is Pitjantjatjara (Pama-Nyungan; South Australia).
(9) Pitjantjatjara (Douglas 1959: 55, 81)
a.
wait
ngalyayala
man
doctor
‘The man is a doctor.’
b.
tjitji
kutjara
ngura-ka
child
two
camp-at
‘The two children are at camp.’
Finally, share-status for a language is also possible on the basis of a verbal encoding for both nominal and locational predicates. Since verbal encoding is definitely a minor typological option for both of these predicate types, it follows that a verbal-verbal shared encoding will be very uncommon as well. This variant can be illustrated by Korku (Munda; central India).
(10) Korku (Drake 1903: 149, 132, 80)
a.
ing
shene-ba
1sg
go-nonpst
‘I go/will go.’
b.
di
dhega
kad
ojha-ba
that
stone
heavy
load-nonpst
‘That stone is a heavy load.’
c.
di
ura-gen-ba
it
house-at-nonpst
‘It is at home.’
As was the case with split encoding, the variation among languages with share-status has been ignored for the purposes of this map.
In the above discussion, the split-share distinction has been defined as a binary parameter, in a yes/no fashion, and it will be represented as such on the map. It must be noted, however, that this binary definition is a simplification in some respects. For one thing, many languages have not just one encoding item for nominal predicates and locational predicates; commonly, copulas and locational support items come in sets, and these sets usually coincide only partially, if they coincide at all. This situation can be illustrated by Dutch. This language has a set of copular items (such as zijn ‘to be’, worden ‘to become’, lijken ‘to appear’), as well as a set of locational verbs (such as zijn ‘to be’, liggen ‘to lie’, hangen ‘to hang’, staan ‘to stand’ and zitten ‘to sit’). Now, the only overlap between these two sets are the items zijn ‘to be’ and blijven ‘to stay’, which can be used for both nominal and locational predication; all the other items are specialized for one or the other of the two predicational functions. Furthermore, since the use of zijn in locational function is much more limited in Dutch than is the use of be in that function in English, one may well ask whether Dutch should not be considered as a split-language rather than as a share-language.
A second factor which tends to blur the distinction between split-languages and share-languages is the phenomenon of copularization of the locational support verb. In some languages, the locational support verb has (or has attained) a limited possibility to act as the copula in nominal predication, in addition to the “real” copula that the language has. This leads to the possibility of a double encoding possibility for nominal predications. Examples are from Spanish and from Tamil (Dravidian; southern India).
(11) Spanish (Max Kerkhof, p.c.)
a.
Julia
es
enfermera
Julia
cop.3sg.pres
nurse
‘Julia is a nurse.’
b.
Julia
está
de
enfermera
(en Madrid)
Julia
be.3sg.pres
prep
nurse
(in Madrid)
‘Julia is a nurse (in Madrid).’
(12) Tamil (Asher 1982: 49, 50, 51)
a.
avaru
(oru)
daktar
he
(one)
doctor
‘He is a doctor.’
b.
ippo
oru
daktar-aa
taan
irukkaraaru
now
one
doctor-adv
emph
be.3sg.hon.pres
‘Now he is a doctor.’
c.
Raaman
tootta-ille
irukkaraan
Raaman
garden-in
be.3sg.m.pres
‘Raaman is in the garden.’
In the large majority of relevant cases, this double encoding of nominal predications is connected with a clear semantic difference, which can be described in terms of the notion of Time Stability (see Givón 1984) or Permanency (see Stassen 1997). For example, in the Spanish examples given above, it must be understood that the (a) sentence (which has a form of the “real” copula ser ) indicates permanent class membership, whereas the (b) sentence (which contains the “copularized” locational verb estar ) must be interpreted as stating that the class membership is only temporary. In this latter case, the sentence might well be translated as “Julia works/acts as a nurse in Madrid”.
In view of the possible indeterminacy created by the phenomena of partial overlap and copularization, the map has been constructed along the following guideline. A language will be called a share-language if at least one of its locational items can be used for copula function, unless this use is governed by conditions of Permanency. In all other circumstances, the language will be rated as a split-language. As a result, Dutch is rated as a share-language, while Spanish has been included among the split-languages.
As the frequency numbers given above demonstrate, shared encoding is definitely the less frequent option among the world’s languages. Nonetheless, there are a number of areas in which this encoding appears to be the rule. First, shared encoding is encountered in what might be called the Eurasian landmass, comprising Europe, central and northern Asia, the Middle East, Pakistan, and at least the northern part of India; notable exceptions here are some languages of the Caucasus and a number of languages on the western fringe of this mega-area (Celtic, Spanish). Secondly, shared encoding is prominent in Australia and New Guinea. Thirdly, we can note a concentration of shared encoding in an area which comprises the southern part of Middle America and the northern part of South America. Finally, shared encoding is found in parts of eastern Africa, mainly due to the progress of copularization in the Bantu languages. Apart from these areas, however, split encoding seems to have a firm foothold.