Search

In this paper, we investigate the value of derivational information in predicting the inflectional behavior of lexemes. We focus on Latin, for which large-scale data on both inflection and derivation are easily available. We train boosting tree classifiers to predict the inflection class of verbs and nouns with and without different pieces of derivational information. For verbs, we also model inflectional behavior in a word-based fashion, training the same type of classifier to predict wordforms given knowledge of other wordforms of the same lexemes. We find that derivational information is indeed helpful, and document an asymmetry between the beginning and the end of words, in that the final element in a word is highly predictive, while prefixes prove to be uninformative. The results obtained with the word-based methodology also allow for a finer-grained description of the behavior of different pairs of cells.

This paper deals with the restrictive (limitative) marker kə̄n ‘only’ in Ulcha (Southern Tungusic). This marker has nontrivial positional features: it can attach before inflectional suffixes (as a derivational affix) or after them (as an enclitic). One might see the process of affix reordering described in Haspelmath (1993) as “externalization of inflection”, when a former clitic becomes a derivational affix. However, there is evidence that the uses of kə̄n after inflection are innovative as compared to those before inflection, not vice versa, and this direction of diachronic development is very unexpected. In this paper, I propose an explanation for this nonstandard reordering pattern and show that in fact it has the same motivation and the same mechanisms as previously reported types of affix reordering.

The morphological expression of non-verbal predication is a geographically widespread, although not very frequent, typological feature. This paper highlights the existence of two radically contrasting types of non-verbal predicative inflection. Construction A has already been described in the literature. It consists of attaching person-sensitive inflection markers to non-verbal predicates, possibly extending this treatment to adverbs and adverbial phrases (locational and temporal), pronouns and quantifiers. This type is well attested in Uralic, Turkic, and Paleosiberian, as well as in some Amazonian language families (most notably Chicham), but it has also been pointed out for some sparse languages of Oceania and Africa. Such non-verbal person inflections diachronically stem from incorporation of conjugated copula elements. Construction B, by contrast, is much rarer and is described here for the first time. It also consists of a dedicated morphological form of the non-verbal predicate (limited, however, to nouns and adjectives), but such form stands out as morphologically lighter than any other form to be found in nouns or adjectives in argument or attribute position. While the latter forms carry some kind of case marker, the noun/adjective predicate merely consists (or historically did) of the word’s root. This type of construction can be found in the small Zamucoan family and still survives in some Tupí-Guaraní languages. Diachronic inspection of Semitic indicates, however, that this predicative strategy was possibly adopted in some ancient varieties, although at later stages it intertwined with the expression of referential specificity. The paper compares the two construction types, highlighting similarities and differences.

Regularity and irregularity are among the most widely invoked notions in linguistics. The terms are backed up by a long and venerable tradition, and yet (or maybe therefore) different disciplines and authors seem to be using them for very different phenomena and in very different ways. The most frequent usage conflates or replaces other notions such as type frequency, productivity, (non-)concatenative morphology, storage vs. computation, predictability, etc. An assessment of these and other variables in Icelandic verbal inflection reveals that most of them are in practice strongly correlated. I conclude, however, that this is largely unsurprising by virtue of the definitional dependencies holding between those notions. It is empirically doubtful whether there exists a single underlying phenomenon or category which the terms designate. In addition, given their multiple and overlapping senses, and the existence of separate, unambiguous labels for the relevant underlying notions, I contend that the terms ‘regular’ and ‘irregular’ should be ideally abandoned in scientific literature in order to avoid ambiguity, sloppy reasoning and misunderstandings and to facilitate cross-linguistic comparison and interdisciplinary dialogue.

The paper discusses the morphological status and the function of Spanish nominal endings -o and -a (ciel+o ‘sky’ vs. caj+a ‘box’); it is shown that both endings, plus the endings -e and -Ø, are inflectional suffixes that mark, however, not the values of an inflectional category (like nominal number or verbal tense), but the values of a feature of the syntactics of the noun — the nominal gender. The ‘nominal gender’ is defined as a cluster concept based on eight properties; it is a particular case of ‘agreement class’ opposed to ‘noun class.’ Some particularities of Spanish nominal gender are examined: its interaction with diminutive suffixes, gender conversion, and its “non-prototypical” character (a parallel is drawn between Spanish nominal genders and noun classes in Fula).

This paper displays examples of inconsistencies in spontaneous speech. It refers to cases in which the speaker changes his manner of expression while speaking, even though there generally seems to be no objective reason for doing so. I demonstrate the phenomenon in the use of verb tense, of person inflection, of singular/plural form, and of direct/indirect speech. I suggest that these phenomena be viewed as cases in which the speaker tries (not necessarily consciously) to make his speech less monotonous and more attractive to the listener by providing various ways of expression differing mainly in the degree of closeness they convey between the reported event and the addressee.

In Palenquero, an Afro-Iberian creole spoken in Colombia, the -ba marker appears in a wider range of environments than other TMA markers, which are commonly pre-verbal. While most creole scholars have assumed that -ba is a past anterior, I argue that this marker is in fact a past imperfect. The crucial cases for demonstrating this involve examples in which -ba occurs with non-stative verbs. The significance of this finding counters the alleged general uniformity of the prototypical creole TMA system which does not include past imperfect marking.