Denumerals are complex morphological units formed on numerals, that is on linguistic expressions used to count objects. Ordinals (e.g. four-th ⟵ four) constitute the commonest kind of denumerals (Stump 2010). But several other varieties exist, among which fractional numbers e.g. hun harminc-ad ‘1/30’, distributives e.g. eus hiru-na N ‘3 apiece’; multiplicatives e.g. deu zehn-mal ‘10 times’; collectives e.g. pol czworo N ‘group of 4 N (persons)’, approximatives e.g. fra cinquant-aine; ‘about 50’, appellatives e.g. rus troj-ka ‘set of 3 elements (working together)’, element counting denumerals e.g. eng bin-ary ‘with 2 elements’, age specifying denumerals e.g. ita vent-enne ’20-year old person’. To these formations, one needs to add compounds whose base is a numeral as nld drie-daag-s 3-day-azr ‘three day lasting’. Detailed information about the aforementioned terms can be found in the Terminology section.
The issue of denumerals is very rarely addressed as such, if at all, in grammars or reference books about languages. The Handbook of Word-Formation (2015) is the first work that includes a chapter dedicated to this topic (if we except Fradin & Saulnier 2009). In general, only ordinal numerals are tackled but their inventory often stops short before reaching interesting questions, notably those involving morphology. For instance, the WALS chapter on ordinal numerals (Stolz & Urdze 2005) misses important phenomena because only the first numerals are dealt with.
The project “Denumerals across languages” was precisely intended to identify existing types of denumerals across languages and to undertake their description from a typological viewpoint. Its goal was to devise descriptive, conceptual and possibly formal tools allowing one to account for the variety of existing denumerals. Apart from publishing articles in specialized journals, the main goal of the project was to build an online accessible database.
While ordinal and fractional denumerals are the most common denumerals that you came across in the majority of languages with cardinal numerals, the situation is not the same for other denumerals. When they exist, they very often form fragmentary series. This is probably one of the reasons why denumerals remain poorly dealt with in grammars, even though this category is worth studying from a typological point of view. However, there are also languages where the series of cardinals is very small, almost nonexistent. There are others where numerals are borrowed from other languages. In such cases, it becomes a real challenge to account for denumerals and this task may turn out to be impossible sometimes. But unexpected findings may also happen, even for well-described languages (Fradin 2019).
The sample of languages described in the project is not typologically well-balanced. It is the result of the interest in the project expressed by colleagues working on various languages. This sample contains a large amount of European languages because the present database re-uses data that were used to document Fradin (2015). These data were obtained through questionnaires that colleagues who were native speakers of these language kindly completed. Many unclear or thorny issues were also discussed by email. About twenty languages have been described. The empirical covering is variable and highly depends on the number and quality of answers that were given to the questionnaire. Explanations about the metalanguage adopted in the database is given in the Terminology section. Editing the data in a coherent format inevitably leads to make mistakes and typos, especially when one knows that this format kept changing as the improvements were made. We hope that only very few of them remain and apologize in advance for the problems they could raise for our readership. Any feedback on this point will be welcome. However small this database, we hope it will help other linguists to describe denumerals in other language and will be useful for typological comparison.
The project was accepted by TUL federation in 2013 and started in January 2014. It ended in December 2018. The architecture of the website was taken, with some modifications, from the Graz Database on Reduplication.
Members of the project listed below contributed to the database for the language facing their name:
| Karla J. | AVILÉS GONZÁLEZ | Nahuatl, nahuatl |
| Isabelle | BRIL | nêlêmwa |
| Hélène | de PENANROS | Lithuanian, lithuanien |
| Aimée | LAHAUSSOIS | Thulung, thulung |
| Akiko | NAKAJIMA | Japanese, japonais |
| Tulio | ROJAS CURIEUX | Páez / nasa yuwe |
| Françoise | ROSE | Mojeño Trinitario |
| Anna | SÖRES DORSCH | Hungarian, Hongrois |
| Yvonne | TREIS | Kambaata, kambaata |
| Alice | VITRANT | Burmese, birman |
Alexandre Roulois (LLF) developed the website and upgrades it in function of coming new data.
Informants who provided data for the languages documented on the website or who answered to questions raised by the description at various moments are thanked on the homepage of the respective languages.
The project was financially supported by the Fédération TUL (Typologie et Universaux Linguistiques, CNRS). We thank the Laboratoire de Linguistique Formelle (UMR 7110, CNRS & Université Paris Diderot) for its human and material support.
Bernard Fradin
Manager of the projet
Directeur de recherche émérite au LLF