Prefix-Suffix Neutrality in Evaluative Morphology

Among studies on morphological typology, the issue of asymmetries in affix position has often drawn the attention of scholars. Broadly speaking, discontinuous affixes (infixes, circumfixes, etc.) seem to be fairly infrequent among world languages, while suffixes and prefixes show a great degree of occurrence in almost all linguistic families. Nonetheless, in a wide cross-linguistic perspective, there seems to be a broader frequency of suffixes than of prefixes. Typologists have often tried to explain this asymmetry between prefixes and suffixes, recurring both to internal and external explanations (for example, correlating affix position to word order or to psychological factors, such as processing of complex words, etc.). But what is really interesting from a typological point of view is that there seem to be some semantic categories which are insensitive to this prefix-suffix asymmetry, that is, which are always cross-linguistically expressed by suffixes or prefixes, independently of the general morphological outline of the language. For example, case for nouns and mood for verbs are nearly always expressed by suffixes, while negation and/or reversion of verbs and adjectives are rarely, if ever, expressed by suffixes. Of course, the formal behaviour of these semantic categories constrains all the typological generalizations that aim to link affix position to other linguistic constructions such as word order. But other data, that have rarely been taken into account by scholars, seem problematic from this point of view. In fact there are semantic categories that seem indifferent to the prefix-suffix asymmetry in the opposite way: they never choose prefixes or suffixes, but take up a neutral position, recurring both to prefixes and suffixes at the same time, even within the same language. One of the best examples of this situation is represented by the so-called ‘evaluative’ semantic functions (SMALL vs. BIG, GOOD vs. BAD, etc.), which are indifferently expressed both by prefixes and suffixes in many languages worldwide. The aim of this paper is to focus on evaluative morphology, in order to explain the possible reasons for this ‘prefix-suffix neutrality’. The paper is divided in three sections. The first section is devoted to the selection of a sample of languages: with the aid of some typological maps, we show the cross-linguistic spread of evaluative prefixes and suffixes. In the second section, we draw a comparative sketch of the properties of evaluative prefixes and suffixes, in order both to single out the features that can explain this neutrality and to understand whether in the single languages the choice between evaluative prefixes or suffixes is semantically/syntactically/morphologically/phonologically conditioned or not. In the third section, we try to establish the ‘typological weight’ of the prefix-suffix neutrality. In other words, we try to verify the possible existence of typological conditions that could foster or discourage this neutrality. In order to achieve this aim, we outline a general typological picture of the languages in the sample, making a correlation between the position of evaluative affixes and four relevant typological parameters: the prefixal vs. suffixal nature of the language, OV vs. VO word orders, the presence of prepositions vs. postpositions, and the position of the adjectival modifier in NP constructions.