North Sámi adjectives show a particularly complex morphological pattern. In certain contexts, they show case and number suffixes (1-2). In others, they don’t, but even here they’re not bare: they surface in a special “attributive” form (3).
| (1) | a. | biila | lea | ruoksat |
| | car.nom.sg | is | red.nom.sg |
| | "The car is red" |
| b. | biilla-t | leat | ruoksad-at |
| | car-nom.pl | are | red-nom.pl |
| | "The cars are red" |
| (2) | sii | orro-t | ruoksad-is |
| they | live-3pl | red-loc.sg |
| "They live in the red one" (e.g. house) |
| (3) | rukses | {biila | / | biilla-t | / | biilla-in | / | ...} |
| red.attr | car.nom.sg | | car-nom.pl | | car-loc.pl | | |
| "A/the car; (the) red cars; in (the) red cars; ..." |
This alternation barely scratches the surface: the empirical data is quite complex. I’m trying to develop a full account of the North Sámi system within current generative theories of morphology. This will then show us constraints on what an adequate morphological theory will have to look like.