Die grense van grammatikale kategorieë wat op skoolvlak gebruik word, soos bysinne, behoort nie wollerig te wees nie. Tog wys Van der Merwe (1966) op die onduidelike grense van bysinkategorieë en op die feit dat Afrikaansonderwysers “skei waar nie geskei behoort te word nie”. Dit wil voorkom of hierdie verwarring steeds bestaan, soos blyk uit ’n meningsopname in 2024 op sosiale media. Omdat bysinne gesofistikeerde sinskonstruksies skep en taalgebruikers bemagtig om hul ervarings van die wêreld uit te druk, wou ons vasstel tot watter mate die bysinkategorieë wat in skoolhandboeke van die afgelope 100 jaar onderskei word, deur wollerigheid gekenmerk word. Hiervoor is ’n bestekopname gedoen, beginnende by Afrikaanse Spraakkuns (Bouman & Pienaar, 1924) tot by die resentste KABV-belynde handboeke vir graad 12. Daar is bevind dat die grense van die bysinkategorieë wat in hierdie bronne onderskei word, inderdaad “wollerig” is. Wollerige gre nse ontstaan, omdat: (i) bysinne wat dieselfde inleier het en logieserwys as dieselfde soort bysin beskou behoort te word, op grond van hul funksie as verskillende soorte bysinne gekategoriseer word; (ii) die kombinasie van verskillende kriteria vir kategorisering gebruik word; en (iii) die grense van verskillende subkategorieë van veral die bywoordelike bysin wollerig is, omdat betekenis as kategoriseringskriterium gebruik word en dieselfde bysin gevolglik as meer as een soort gekategoriseer kan word. Meerduidige kategorisering en ’n skeiding tussen bepalings en bysinne (wat as bepalings gebruik word) is ook aan die orde van die dag.
Categorisation is an inherent characteristic of all approaches to grammar. Although the boundaries of grammatical categories at school level should not be fuzzy, it seems that this is in fact the case. Fuzzy boundaries of grammatical categories are problematic within the context of school instruction, as any pedagogical school grammar, according to Bourke (2005:85) and Thornbury (1999:32), should meet specific criteria. Among such criteria are restrictedness and simplicity, which implies that a pedagogical grammar must clearly indicate what the boundaries of a given grammatical unit are. Van der Merwe (1966) refers to, for example, the unclear boundaries of subordinate clauses, stating that teachers of Afrikaans separate where separation should not occur, and this confusion seems to persist, as evidenced by a survey conducted on social media in 2024. Since subordinate clauses create sophisticated sentence constructions and enable language users to express their experiences of the world, it was important to determine to what extent subordinate clauses at school level are characterised by fuzziness. To this end, a survey was conducted, beginning with Afrikaanse Spraakkuns (Bouman & Pienaar, 1924) and extending the survey to the most recent CAPS-aligned textbooks for Grade 12 learners. From the conducted survey it is clear that the boundaries of subordinate clauses at school level have indeed been fuzzy over the past 100 years. Fuzzy boundaries arise because subordinate clauses that are introduced by the same word class and should therefore logically be considered the same type of subordinate clause, are assigned to different types of subordinate clauses, based on their function. Although functional criteria are primarily used to distinguish the main types of subordinate clauses, this criterion is often associated with formal, syntactic, or semantic properties. This combination of different criteria for categorisation also indicates the poorly defined nature of subordinate clauses, or their fuzzy boundaries. Furthermore, the boundaries of various subcategories, particularly adverbial subordinate clauses, are fuzzy, as meaning is most often used as a criterion. This means that the same subordinate clause (e.g., Jason het te veel kersies geëet, [sodat] hy nou maagpyn het) can be classified in multiple ways, either as an adverbial clause of degree (answering the question “To what extent?”) or as an adverbial clause of contingency (answering the question “What happened then?”). Certain gaps regarding clause categorisation were also noted. For instance, in the sentence Die berig dat hy dood is, is onwaar, the subordinate clause dat hy dood is, is regarded as a subject clause, even though the main clause (Die berig is onwaar) has a subject (die berig). This is in contrast with the guideline that a clause can only be an object/ subject clause if such a part of the main clause is not present. Therefore, dat hy dood is should rather be classified as another type of clause. Den Hertog (1904:70) classifies such clauses that fulfil the function of modifiers of nouns, limiting the representation of the noun or providing secondary information about the noun, as adjective clauses. We also noted a shift from viewing the modifier as a syntactic function that can be performed by different grammatical elements (including subordinate clauses), to viewing it as a formal-functional grammatical unit that is limited solely to the simple sentence and contrasted with subordinate clauses. For instance, in Den Hertog (1904:67) it is acknowledged that a subordinate clause can have the function of an adverbial when used as a modifier that restricts the action/state of the verb, therefore rendering it an adverbial clause. Likewise a subordinate clause can fulfil the function of a modifier of a noun within a noun phrase, leading to its classification as an adjective clause (compare Den Hertog, 1904:67). Modifiers have since been understood as a group of words without a predicate that is used for modification and is contrasted with subordinate clauses (compare Vermaak et al., 2013).
