The under-researched Hindu Kush-Karakoram region (NE Afghanistan, N Pakistan and N Kashmir) with its 50 densely situated language communities, representing 6 genera, provides unique opportunities for studying the many, often competing, forces at work in the typological fine-tuning of a selected geographical area. By applying Koptjevskaja-Tamm’s (2010:584) guidelines for areal-linguistic research, 5 linguistic domains (retroflexion, gender, alignment, numerals and spatial reference) were investigated – for a tight sample and partly by novel field data – and their respective micro-typologies were evaluated against wider distributions, as well as against Nichol’s (2003) stability predictions, each of the domains illustrating a different profile in terms of diversity, family-internal stability, cross-linguistic diffusion, sub-areality or inclusion in region-external configurations.

A number of languages, particularly in the central–northern parts of the region, are characterized by large retroflex inventories. While retroflex plosives have a wide distribution in South Asia, regional languages in 5 of the 6 genera include an additional set of fricative/affricate retroflex consonants. Grammatical gender is present in four of the six genera. In Indo-Aryan, the region’s largest phylogenetic component, an inherited sex-based system persists but is more pervasive in the East, i.e. contiguous with the main Indo-Aryan belt of the Subcontinent, whereas it overlaps with an animacy-based system in the SW and has been entirely “replaced” by such a system in the NW adjacent to similarly gender-deprived or gender-less languages. A combination of sex and animacy as a basis for gender characterizes Burushaski’s 4-gender system, the region’s only language isolate, thus hinting at possible substratal influences. Ergative alignment is evidenced in 5 of the 6 genera. The distribution of particular alignment patterns, however, illustrates how sub-areas participate in a few wide-spread configurations that conflate in the region, see Liljegren (2014). Overt case-marking of patients characterizes the West, which extends to a large, Persian-dominated, area. Overt, tense-aspect-independent, case-marking of agents characterizes an area in the East, linking it to Tibetan-dominated regions of the Himalayas. Patient agreement in the perfective is a strong feature of the South, i.e. contiguous with the influential Hindi-Urdu belt. As for numerals, there is a dominance of vigesimal systems across the region, including 5 of the 6 genera, but in addition there is a distinct sub-areal distribution of numeral composition, with a consistent 10+n/20+n structure along the northern fringe, continuing into the Pamirs, a consistent n+10/n+20 structure in the southeastern parts (thus contiguous with the dominant languages of the Subcontinent which, although decimal, share this compositional structure), while languages distributed in a central west-to-east belt display a mixed (n+10/20+n) structure, see Liljegren (2017:143–145). Finally, the study reveals the presence of a common geomorphic system of spatial reference in a subarea in the West, whereas it seems absent in other parts of the region. These languages encode the inclination of the mountain slope as well as the flow of the river. While this system reflects the topography of the surrounding landscape, the emergence and pervasiveness of it seems linked to language contact, clustering significantly with a few other features (Liljegren & Svärd 2017) and coincides with the boundaries of an area that only recently came under the influence of Islam (Klimburg 2008; Jettmar, Jones & Klimburg 1975:394).     

Finally, an attempt is made at characterizing the Hindu Kush in its entirety along a scale of diversity—homogeneity.