Voorwaardelike konstruksies met indien soos gebruik in wetenskaplike tydskrifte (deel 2: gemerkte konstruksies) Translated title: Conditional constructions using "indien" in scientific journals (Part 2: Marked constructions)

Abstract

Hierdie artikel vorm die tweede deel van 'n beskrywing van voorwaardelike konstruksies met indien soos in wetenskaplike tydskrifte aangetref. In die eerste deel is basiese konstruksies onder die loep geneem, terwyl hierdie deel oor gemerkte konstruksies handel. Konstruksies met verkorte bysinne of indien-uitdrukkings word as gemerk beskou. 'n Volledige konstruksie word as gemerk beskou wanneer dit nie met die protasis begin nie. Die agterplasing van die protasis word in terme van diskoersverwante, semantiese en sintaktiese faktore verklaar: posisie van die konstruksie in die groter diskoers, fokus in die sin, betekenisnuansering deur bywoorde, inlywing by sinsdele en inlywing as sinsdele. Dit blyk dat 'n skrywer volgens sy/haar kommunikatiewe doel 'n keuse ten opsigte van die protasis-apodosis-volgorde kan uitoefen, maar dat die keuse soms deur sintaktiese faktore beperk word. Die gebruike van gemerkte indien-konstruksies met hipotaktiese binding, kan soos basiese konstruksies geklassifiseer word volgens die mate van die waarskynlikheid dat 'n stand van sake gerealiseer sal word. Twee van die subkategorieë wat by die basiese volgorde voorkom, is egter nie in die gemerkte volgorde aangetref nie. In die beskrywing word na die rol van indien as ruimtebouer verwys en aan die einde van hierdie artikel word 'n oorsig oor die soorte indien-konstruksies in die korpus met verwysing na denkruimtes verskaf. Verdere navorsing is nodig om dieper op die aard en sterkte van die binding tussen protasis en apodosis in te gaan.

Translated abstract

This paper is the second part of a two-part description of the use of the Afrikaans conjunction "indien" (if) in scientific journals. The research framework, the data corpus and the nature of the conditional constructions are the same as in Part 1 and are not repeated here. This paper deals with non-basic, i.e. marked constructions and analyses the syntactic, semantic and pragmatic aspects of marked constructions. "Indien" constructions are considered to be marked when they contain incomplete clauses or the subordinate clause follows the main clause. In some sentences the clauses are not hypotactically bound. The paper starts with a table giving the frequencies of the placement of the "indien" clause in the sentence. The data show that there are different motives for using the marked (q,p) order in sentences. Some cases can be explained by the discourse where a connection with a preceding part is needed. Sometimes the usage is syntactically mandated. Complex sentence structure, deep embedding and a change in the condition type often lead to placement of the subordinate clause near the end of the sentence. Associated adverbs change the meaning of the subordinate clause and the q,p order is semantically motivated in those cases. In the data the following factors were identified that caused the q,p order. Cohesion with the preceding discourse through adverbs, pronouns and nouns was noted in the q,p order. The q,p order is also used in definitions and descriptions to draw focus to the term being described or defined. In complex sentences the conditional construction is not always dependent on the main verb. Deep embedding or dependency on noun phrases may cause the q,p ordering. "Indien" constructions, where the main clause contains a comparison, tend to have the marked order. When the"indien"-clause is part of an infinitive, it is placed after the main clause. When the "indien" construction is embedded in a relative clause the q,p order is generally used. "Indien" clauses are placed at the end of a sentence if they are embedded in a complement clause of a noun phrase. Using "dit" (this) as a preliminary subject of a copulative verb causes the q,p order. Several associated adverbs cause the q,p order. The adverb "slegs" (only) and its less formal synonyms "net", "alleen" and "alleenlik" are often associated with "indien". When a conjunction is preceded by an adverb, the combination can change the type of relationship between the apodosis and the protasis. The construct "slegs indien" (only if) implies multiple possible mental spaces, where only one is accepted. The construct implies a strong condition or restriction. It is possible to regard "slegs indien" as a composite conjunction with a meaning close to "mits" (provided that). Other adverbs in the data include "selfs" (even), "ook" (also), "veral" (especially) and "behalwe" (except). "Veral indien" (especially if) implies that more than one mental space exists, but that a particular one is preferred. The corpus contains eleven cases with "veral indien" of which ten exhibit the q,p order An "indien" clause can be placed in a main clause in such a way that the main clause is interrupted. The "indien" clause has a parenthetic character in those cases and is often placed between commas, parentheses or attention hyphens. The binding does not show the typical hypotactic firmness. Most ofthese internal clauses act as post-determiners ofnouns. The purpose of the internal "indien" clauses seems to be to indicate a restriction of the state of affairs in the main clause. The data contain sentences where the subordinate clause containing "indien" is shortened to a stock phrase. Examples are "indien nie" (if not), "indien wel" (if so), "indien nodig" (if necessary), "indien ter sake" (if relevant) and "indienprakties uitvoerbaar" (iffeasible). These stock phrases occur in all three possible positions, at the start of the sentence (p,q order), at the end of the sentence (q,p order) or internally embedded in the main clause. The shortening is observed in various degrees and is not always down to a stock phrase. Where the stock phrases occur between commas, the phrase can usually be extended to a complete phrase without changing the meaning of the sentence. There is, however, a set of shortened constructions that cannot be extended to complete clauses without changing the meaning of the sentence. In these cases the "indien" constructions are used as a strengthening or weakening of a term. They are incorporated as post-determiners of nouns, adjectives or adverbs. The q,p order is used in statements about the past where the "indien" construction describes the conditions under which the event in the main clause happened. Statements in the generic present tense tend to use the q, p order when restricting adverbs are present. Two subcategories that were described in part 1 were not found with the q, p order. These are "academic communications" and statements that make derivations from the state of affairs in the past. Predictions rarely use the q,p order, because of the iconicity in time or causality: p has to hold before q can be considered. No commands using the q,p order were found in the corpus. Questions do exist where the "indien" construction creates a hypothetical mental space to serve as background. Rhetorical questions used to criticise scientific argument are present in the q, p order. The paper concludes with an overview diagram ofthe marked "indien" constructions, showing the different types and their relationship in terms of the Theory of Mental Spaces. In conclusion the paper notes that the marked constructions can be categorised in the same broad usage categories than the basic constructions discussed in Part 1. Two sub-categories were only observed with p, q order. Additionally, the p, q order was observed in sentences with different types of clause linking. More research is necessary to investigate the type and strength of the binding between the protasis and the apodosis.