Which languages have the most grammatical cases?
In the WALS sample of 261 languages, 24 have ten or more cases, the top category. Among them are Finnish, Hungarian, Estonian, Basque and Lezgian, a Nakh-Daghestanian language of the Caucasus. At the other end, 100 languages (38%) mark no case on nouns at all.
The WALS chapter adopts the definition of case as a system of marking dependent nouns for the type of relationship they bear to their heads. English is coded with two cases, German with four and Russian with six or seven.
What the counts include
The chapter counts a marker as case when it is bound closely enough to its noun, which can include clitics and adpositional markers as well as endings. It sets apart a group of 24 languages whose only case marking is borderline, and it uses bands (2, 3, 4, 5, 6–7, 8–9, 10 or more) rather than exact counts.
Where the large systems are
| Macro-area | Coded | No morphological case-marking | 6-7 case categories | 8-9 case categories | 10 or more case categories |
|---|---|---|---|---|---|
| Africa | 41 | 56% | 17% | 0% | 0% |
| Eurasia | 83 | 22% | 17% | 12% | 18% |
| Papunesia | 44 | 70% | 5% | 7% | 2% |
| Australia | 16 | 12% | 12% | 31% | 31% |
| North America | 45 | 38% | 16% | 7% | 2% |
| South America | 32 | 28% | 16% | 6% | 6% |
| All areas | 261 | 38% | 14% | 9% | 9% |
Rich case systems cluster in Eurasia and Australia. Eurasia has 15 of the 24 languages with ten or more cases, among them the Uralic languages Finnish, Estonian and Hungarian and several Nakh-Daghestanian languages of the Caucasus. Australia has 5 more, with 10 of its 16 coded languages at eight or above. Papunesia leans the other way: 31 of its 44 coded languages (70%) have no case marking, as do 23 of 41 in Africa.
Cases and word order
Languages with many cases tend to put the verb last. Of the 73 languages with six or more cases that are also coded for basic order, 47 (64%) are subject–object–verb. Among the 97 languages with no case marking, the most frequent order is subject–verb–object, with 50 (52%).
| Number of Cases | Subject-object-verb (SOV) | Subject-verb-object (SVO) | Verb-subject-object (VSO) | Lacking a dominant word order | All |
|---|---|---|---|---|---|
| No morphological case-marking | 18 | 50 | 10 | 11 | 97 |
| 2 case categories | 11 | 5 | 3 | 1 | 21 |
| 6-7 case categories | 22 | 4 | 3 | 4 | 33 |
| 8-9 case categories | 16 | 0 | 0 | 3 | 21 |
| 10 or more case categories | 9 | 4 | 0 | 6 | 19 |
Languages with ten or more cases also include a sizeable group without a dominant basic order: 6 of 19. The table shows an association, not a mechanism, and it rests on a sample of a few hundred languages.
Familiar languages
Several widely studied languages sit in the middle bands. German is coded with four cases, and Russian, Turkish and Korean with six or seven. Japanese falls in the eight-to-nine band. Spanish, French, Mandarin and Swahili have no case marking on nouns. Within families the picture varies: 23 of the 23 Austronesian languages coded lack case marking, while 5 of the 8 Uralic languages have ten or more cases.
One family, every band
A family does not fix the number of cases. The 32 Indo-European languages coded spread across the scale: 8 have no case marking on nouns, 14 have two to four cases, 10 have five to seven, and 0 have eight or more. French and Spanish sit at the low end, German has four, and Russian six or seven. In North America, 9 of the 45 languages coded fall in the borderline group, the highest share of any area, which is a reminder that the line between an ending and a separate word is not always sharp. Across the whole sample, 24 languages fall in that group.
How the chapter draws the lines
Two rules in the chapter explain many of the codes. In languages without morphological case, such as Vietnamese, it notes that grammatical relations are expressed by word order and by separate function words, mostly prepositions and postpositions, and partly by marking on the verb. And the smallest case system counts two members: whenever a language has one overtly marked case, the unmarked base form counts as a second case. That is why a language with a single case ending, such as Mapudungun in the chapter’s own example, is coded with two cases.
Languages with ten or more cases on this site
Of the 24 languages in the top category, 21 have a page here: Awa Pit, Basque, Brahui, Chukchi, Epena Pedee, Estonian, Evenki, Finnish, Gooniyandi, Hamtai, Hungarian, Hunzib, Ingush, Kayardild, Ket, Lak, Lezgian, Martuthunira, Nez Perce, Nunggubuyu and Pitjantjatjara. Each page lists the language’s other coded features, so the case count can be read next to its word order, its verb morphology and its closest profiles. The feature page for Number of Cases maps every coded language by band. In the contrast tool, Finnish against English makes a clear example of how a rich case system shows up across other features.
Sources. Dryer, Matthew S. & Haspelmath, Martin (eds.) 2013. The World Atlas of Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology. (Available online at https://wals.info) Values from the WALS CLDF dataset, release v2020.4 (doi:10.5281/zenodo.13950591), licensed CC BY 4.0. Counts and percentages computed by this site; see how to read these pages.
WALS chapters quoted or summarised (CC BY 4.0):
- Oliver A. Iggesen. 2013. Number of Cases. In: Dryer, Matthew S. & Haspelmath, Martin (eds.) 2013. The World Atlas of Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology. wals.info/chapter/49
More field notes