<p><em>In Afrikaans, plurality is indicated phonetically in several ways. The large number of pluralisation rules and the many exceptions to these rules cause acquirers of Afrikaans to make some use of rote learning. The question arises as to how, if at all, the knowledge of pluralisation of Afrikaans-speaking children with specific language impairment (SLI) differs from that of typically developing children: if even typically developing Afrikaans-speaking children have to learn the correct phonetic realisation of the plural forms of nouns (to a certain extent) and if this learning is not yet completed by 6 years of age (Southwood, 2006), can knowledge of pluralisation then be used to differentiate between Afrikaans-speaking children with and without SLI (seeing that SLI is characterised by a deficit in grammatical morphology)? This paper attempts to answer this question by examining the comprehension and production of plural forms by 10 6-year-olds with SLI and 10 without. It was found that </em>some<em> selected measures of comprehension and production of pluralisation are sufficiently sensitive to differentiate between the two groups. It was also found that neither of two prominent accounts of SLI, namely the Feature Deficit Hypothesis (Gopnik, 1994a) and the Surface Hypothesis (Leonard, 1989 and others), offers an adequate explanation for the problems with pluralisation experienced by Afrikaans-speaking children.</em></p>