This chapter investigates the form of the verb in purpose clauses. Following a common practice in typological research (see among others Stassen 1985 and Croft 1990: ch.1), purpose clauses are defined in functional rather than morphosyntactic terms, as the latter turn out to be of restricted applicability in cross-linguistic comparison. A purpose construction is regarded as one encoding a particular relation between events. This relation is such that one of the linked events (the one coded by the main clause, or the main event) is performed with the goal of obtaining the realization of another one (the one coded by the purpose clause, or the dependent event). Typical cases of purpose relations involve motion predicates in the main event, as in (1). In this case, main and dependent event share a participant. However, non-motion predicates may also be involved, in which case main and dependent event need not share a participant, as is shown by (2) (the purpose clause is indicated by square brackets in the examples):
(1)
I went downtown [to buy books].
(2)
I printed out a copy of this chapter [in order for you to look at it].
In Cristofaro (2003: ch. 2, 6), the theoretical implications of the proposed functional definition of purpose clauses are discussed at length, and some pragmatic criteria are provided to identify purpose clauses under this definition.
The proposed definition of purpose clauses encompasses all of the traditional cases of purpose clauses (non-finite clauses, possibly introduced by specific purpose conjunctions, as exemplified in (1) and (2), as well as finite clauses introduced by specific purpose conjunctions), plus some other cases that might not be regarded as such under traditional morphosyntactic criteria. One such case is provided by the serial verb construction in Mandarin Chinese. This construction involves a number of juxtaposed verbs, and can be translated as a purpose sentence (as in (3a)) or a coordinate sentence (as in (3b)) depending on the context:
(3) Mandarin Chinese (Li and Thompson 1973: 98)
a.
nĭ
gui ̀-xialai
[qiú Zháng-san]
you
kneel.down
beg Zhang-san
‘You knelt down in order to beg Zhang-san.’
b.
nĭ
gui ̀-xialai
qiú Zháng-san
you
kneel.down
beg Zhang-san
‘You knelt down and begged Zhang-san.’
In (3a) begging Zhang-san is the goal of the person's kneeling, but the relevant clause does not show any of the morphosyntactic hallmarks that differentiate purpose clauses from main clauses in other languages. However, under the proposed functional definition of purpose clauses, this clause can be regarded as a purpose clause, because this is the way Mandarin Chinese expresses the conceptual situation associated with purpose clauses in other languages.
Following a distinction introduced by Stassen (1985), verb forms in purpose clauses are classified as either balanced or deranked, and the corresponding clauses will be called balanced and deranked purpose clauses (for some differences between the notion of deranking adopted here and Stassen's original formulation, see Cristofaro 1998 and 2003: ch. 3). A balanced verb form is one that can occur in an independent declarative clause, such as for instance an indicative form. Example (3a) above provides another instance of a balanced verb form: the linked verbs show the same morphological form, and each of them could occur in an independent declarative clause.
A deranked verb form is one that is structurally different from those used in independent declarative clauses. Deranking may take a variety of forms, such as lack of the categorial distinctions normally relevant to verbs in the language (e.g. tense, aspect, mood or person agreement distinctions), or use of special elements not relevant to verbs in independent declarative clauses, e.g. nominalizers, case marking or adpositions.
In example (4), from Lango (Nilotic; Uganda), the relevant verb form bears an infinitive suffix, has no aspect, mood or person affixes, and is accompanied by the adposition mɛ̂:
(4) Lango (Noonan 1992: 245)
lócə
òbínô
[mɛ̂
càmmò
gwɛ̀nò]
man
3sg.come.perf
for
eat.inf
chicken
‘The man came to eat chicken.’
Deranked verb forms may also display special tense, aspect, mood or person markers not used in independent declarative clauses. This is usually the case with subjunctives and the so-called dependent moods found in West Greenlandic (Fortescue 1984) or Abkhaz (Hewitt 1987). Example (5), from Italian, provides an instance of a subjunctive purpose clause. Tense, aspect, mood, and person are expressed by means of special affixes not used in independent declarative clauses (as is shown by the contrast between scriv-e ‘write-3sg.ind’ vs. accorg-a ‘realize-3sg.subj’):
(5) Italian
Glie-lo
scriv-e
[affinché
se
ne
accorg-a].
to.him/her-it
write-3sg.ind
in.order.for
refl
of.it
realize-3sg.sbjv
‘S/he writes that to him/her in order for him/her to realize that.’
In (6), from Finnish, person agreement on the deranked verb ‘buy’ is expressed by means of possessive suffixes, instead of the agreement suffixes found in independent declarative clauses:
(6) Finnish (Sulkala and Karjalainen 1992: 55)
Men-en
kauppa-an
[osta-a-kse-ni
ruoka-a].
go-1sg
shop-ill
buy-inf-trasl-1poss
food-prtv
‘I am going to the shop in order to buy food.’
Dependent moods and subjunctives display the same categorial distinctions as the verb forms used in independent declarative clauses (this is for instance the case in the Italian example in (5) above). Stassen (1985: 338-339) observes that in this case one might want to regard these forms as balanced (as is done for instance in Hengeveld 1998). However, since these forms cannot occur in independent declarative clauses, they will be regarded as deranked here.
It should be stressed that the distinction between balancing and deranking is not equivalent to that between finiteness and non-finiteness (for which see Joseph 1983: ch. 2 and Koptjevskaja-Tamm 1994). The latter distinction is chiefly based on morphological criteria, i.e. non-finite verb forms have reduced inflectional potential, while finite verb forms display full inflectional potential. The balancing/deranking distinction, on the other hand, is based exclusively on the ability of a verb form to occur in independent declarative clauses. Many verb forms with reduced inflectional potential (infinitives, participles and the like, such as the verb càmmò in (4)) cannot occur in independent declarative clauses, and thus count as deranked forms. However, verb forms with reduced inflectional potential do count as balanced if they can occur in independent declarative clauses. This is for instance the case of nominalized verb forms in Kayardild, which are used in both purpose clauses and independent declarative clauses (where they express uncompleted action: Evans 1995: 470-474). On the other hand, verb forms with full inflectional potential, such as subjunctives or dependent moods, cannot occur in independent declarative clauses, and thus count as deranked.
An apparent problem for the balancing/deranking distinction is provided by verb forms bearing clitics or affixed conjunctions, such as the purpose conjunction –ghoo in the following Muna (Austronesian; Indonesia) example:
(7) Muna (van den Berg 1989b: 264)
no-kala
we
sangku
[nae-wei-ghoo]
3sg.real-go
loc
forest
3sg.irr-clear-purp
‘He goes to the forest to clear (it).’
One might want to regard such forms as deranked, because the complex verb + clitic/affixed conjunction could not occur in an independent declarative clause. However, the verb as such often displays the same categorial distinctions as the forms used in independent declarative clauses. This is for instance the case in Muna, where the verb in the purpose clause is inflected for the same categories (person and realis) appropriate for verbs in independent declarative clauses. Therefore, these forms will be regarded here as balanced, unless some proper deranking strategy (e.g. lack of inflectional distinctions, expression of inflectional distinctions by means of special forms not used in independent clauses) is also at work.
For any given language, a purpose clause can be coded by deranked verb forms only, by balanced verb forms only, or by either deranked or balanced verb forms. Map 125 shows the geographical distribution of these three situations in the world's languages.
| Go to map | ||
|---|---|---|
| Value | Representation | |
 |
Balanced | 38 |
 |
Balanced/deranked | 30 |
 |
Deranked | 102 |
| Total: | 170 | |
The vast majority of the world's languages have deranked purpose clauses, as can be seen from the feature value table at the end of the previous section. Languages with exclusively balanced purpose clauses are concentrated in Southeast Asia and the Pacific islands. Languages with deranked purpose clauses are common all over the world, except for mainland Southeast Asia and Indonesia.
Languages with exclusively deranked purpose clauses are dominant in the Americas, though languages with both deranked and balanced purpose clauses, or exclusively balanced purpose clauses, are also widespread there. Languages with exclusively balanced purpose clauses are relatively common in North and Central America, but rarer in South America. Languages with both deranked and balanced purpose clauses are more common in North and Central America than in South America.
Languages with exclusively deranked purpose clauses are also dominant in Africa. Languages with both deranked and balanced purpose clauses, or balanced purpose clauses only, are also relatively common, and the former are more frequent than the latter.
Europe is divided in two clear-cut areas, with languages with both balanced and deranked purpose clauses towards the north, and languages with exclusively deranked purpose clauses towards the south.
Most languages in central, southern, and eastern Asia display exclusively deranked purpose clauses.
Languages with exclusively deranked purpose clauses are dominant in Australia, although a few languages with exclusively balanced purpose clauses, or both balanced and deranked purpose clauses, are found in the north. In the Pacific islands, languages with exclusively deranked purpose clauses are as common as languages with exclusively balanced purpose clauses; languages with both deranked and balanced purpose clauses are marginal here (the only such language in the sample is Amele, in Papua New Guinea).
The most striking fact about the distribution of balanced and deranked purpose clauses across the world's languages is the dominance of languages with deranked purpose clauses. In Cristofaro (1998, 2003), it is argued that there is a functionally motivated connection between the structural features of deranked verb forms and the semantic features of purpose relations. In purpose relations, the time reference of the dependent event is predetermined as posterior to that of the main event, and the two events often share a participant (see example (1) above). Thus, the time reference of the dependent event, and the participant shared by main and dependent event, need not be expressed overtly. This leads to the use of verb forms not marked for time and person agreement, i.e. deranked forms. Thus, the use of deranked forms in purpose clauses reflects an economic motivation, i.e. the tendency not to express what is entailed or recoverable from the context (see on this point Haiman 1983 and 1985: 102-130). Furthermore, the dependent event in a purpose relation is unrealized, and unrealized events are often expressed by means of special verb forms not used for realized events, such as subjunctives and the like - these forms are also regarded as deranked here. Finally, purpose relations involve a high degree of semantic integration between main and dependent event (in the sense defined in Givón 1980 and 1990: ch. 13). Use of deranked verb forms often leads to syntactic integration between clauses, in that a number of categories relevant to the dependent verb (tense, aspect, mood, person) are not expressed, so that the information about these categories must be recovered from the main verb. Hence the use of deranked forms in purpose clauses reflects an iconic motivation, i.e. semantic integration between events is iconically reflected by syntactic integration between clauses.
The functional correspondence between the structural features of deranked verb forms and the semantic features of purpose relations explains why balanced purpose clauses seem to be disfavored across the world's languages. In fact, as is shown in Cristofaro (1998, 2003), an implicational hierarchy exists such that if a language uses balanced verb forms for purpose clauses, than it uses balanced verb forms for the other adverbial clause types, as well as for relative clauses and most types of complement clause.