Tibetan affixes in Wutun


Affix function number of borrowed affixes


Information and examples are mainly from Janhunen et al.’s (2008) detailed description, including etymologies for affixes. However, Lee‑Smith & Wurm (1996) give different etymologies for almost every one of the potentially borrowed items. The preference for Janhunen et al.’s (2008) etymologies is based on the authority of Juha Janhunen as a specialist of East Asian languages. The situation in Wutun has been summarized as follows: “Wutun has borrowed so extensively from Amdo Tibetan, both grammatically and lexically, that on first encounter, one hardly recognizes it as a Chinese language” (Li 1984: 320). Note also that according to Slater (2003: 329) it often “becomes impossible to trace the precise historical path of any given linguistic feature [in China’s Qinghai‑Gansu Sprachbund]”.


2 case markers (out of a total of 7 in Wutun, the remaining 5 are native Chinese: genitive, locative, superessive, sociative, comparative)

‑na ‘distributive case’ (Janhunen et al. 2008: 59), e.g. san nian‑na ‘for three years’, liang tian‑na ‘for two days’

‑la ~ ‑ra ‘ablative case’, e.g. gguan‑la ‘from the temple’, gunse‑la ‘from the company’. This marker is possibly identical to the verbal “serial marker” ‑la. Since Janhunen et al. (2008) are silent about the possible etymology of this marker, in this case I follow Lee‑Smith & Wurm (1996: 888) according to whom “directional ‑ra probably reflects the Tibetan directional, locative ‑la.


3 “verb complements” (out of a total of 20 in Wutun). All of these appear to be paradigmatically related, the subdivision in three sets given by Janhunen et al. (2008: 76–86) seems to be based on semantic criteria. With respect to their status as affixes, they “differ greatly in their degree of grammaticalization”, but all of these are written as suffixes. Some at least are “very close to grammatical markers”.

‑la ‘non‑completive’ “voice complements”, “‑la can also be attached to verbs of a Sinitic origin” (Janhunen et al. 2008: 78), e.g. qu‑la‑li ‘to live in it’ (from qu ‘live’). The other two “voice complements” are ‑gu ‘completive’ which is native Chinese, and ‑ge ‘causative’ which is possibly native Chinese (Janhunen et al. 2008: 76–80).

‑qai ‘to get broken’ “aspect complement”, no example of hybrid formation given. There are a total of 11 “aspect complements” in Wutun (Janhunen et al. 2008: 80–82).

‑dd‑ ‘to want’ “modal complement”, no example of hybrid formation given. There are a total of 11 “modal complements” in Wutun.


2 serial markers “that indicate the relationships between the complement events described by the serialized verbs” (Janhunen et al. 2008: 89–92). For the general serial marker ‑ma no etymology is given, only that it is potentially identical to the conjunction ma ‘and’. There are a total of 3 such serial markers in Wutun.

‑da ‘consequential’, e.g. ngan‑di‑di ‘as (he) was pressing’, jja‑la‑gu‑da ‘if you visit’

‑la ‘conditional’, e.g. kan‑gu‑de hong‑la ‘if you read’. This marker is possibly the same as the ablative case (see above).