3.1 Lexicon: Names of the Days and Months

The names of the days and months as well as the system used in dividing a year into months are quite similar in Kinnauri to that of the names and the calendar system found in Kinnauri Pahari and also in the IA languages of the plains (i.e., outside the Himalayan region).³

Table 60 shows that the names of the days of the week in Kinnauri⁴ have similar counterparts in IA languages and that the names in Kinnauri are very different from those of Navakat.

As was the case with the days of the week, the terms for months in Kinnauri are also very similar to the terms used in those IA languages where the Hindu religion is prevalent (see Table 61). Here we find not only similarities in the forms of the names of the months, but also in the way in which the year is divided into months. The first column (“Period”) describes how a year is divided into months in both Kinnauri and in Kinnauri Pahari; the second column provides the Kinnauri terms and the third column provides corresponding month names in some IA languages.

Similar borrowing of the Hindu calendar system and names for the week-days is also found in some other West Himalayish languages, e.g., Kanashi (own fieldwork data), Darma (Willis Oko 2019: 467), and marginally also in Tinani (see below).

Distinct from this, two other ST languages of Himachal Pradesh for which we have the relevant data—Navakat and Tinani⁶—exhibit both a different division of the year into months (“Period”) and naming of the months (“Navakat” and “Tinani”), as shown in Table 62.⁷ The Navakat naming system, where the months are simply numbered, is also found in Tibetan. Interestingly, while Tinani has not borrowed the IA calendar system (Table 62), it has borrowed the names of the weekdays (Table 63). For further details, see Saxena and Borin (2022b).

To summarize, the terms for the days of the week and months as well as the calendar system in Kinnauri are very similar to that found in many IA languages. Singh (1990: 248) describes how the village gods were claimed to have more Hindu affinities in the Lower Kinnaur region, and more Buddhist affinities in Upper Kinnaur. He suggests that the Hindu and Buddhist characteristics that we see today in modern Kinnaur are secondary developments, which are superimposed on the earlier—pre-Hindu and pre-Buddhist religion of the ethnic population in Kinnaur. Keeping in view the socio-cultural factors involved, it is very likely that, in this case, the IA influence on Kinnauri comes either through religion or through some other channel, and not directly from Kinnauri Pahari.

3.2 Lexicon: Words for Past and Future Time Adverbs

ST languages tend to have distinct words for past and future time adverbs (i.e., for terms corresponding to the English yesterday and tomorrow; day before yesterday and day after tomorrow). This is illustrated in Table 64 with examples from some West Himalayish languages, including Kinnauri.⁹

Distinct from this, in many IA languages the same term is used for both past and future time adverbs (e.g. Hindi kal, Assamese kali, Punjabi kala and Rajasthani kyāla are all used in these languages for both ‘yesterday’ and ‘tomorrow’). However, Kinnauri Pahari has separate sets of terms for past and future time adverbs (e.g., hiːdz ‘yesterday’, kaːle ‘tomorrow’; see also Table 65),¹⁰ just as in Kinnauri—though the terms are different in the two languages.

At first glance, one might be tempted to conclude that Kinnauri Pahari has borrowed this feature from Kinnauri, but this is not borne out by the distribution of this feature across IA. There are several Western Pahari languages as well as some languages in other subfamilies of IA, which exhibit this pattern (e.g., Marathi kaːl ‘yesterday’, ud̪jaː ‘tomorrow’; Kashmiri yēwa, kāl ‘yesterday’, pagāh ‘tomorrow’) (see the emphasized items in Table 65).

Further, Sanskrit, which represents the older stage of the contemporary IA languages, had this distinction; terms such as hīdz ‘yesterday’ and shūī ‘tomorrow’ (see Table 65) are related to the Sanskrit forms hyas ‘yesterday’ and śvas ‘tomorrow’, which have disappeared from IA languages such as Hindi, but are retained in some modern IA languages.

An overview of past and future time adverbs in IA (and ST) languages is presented in Figure 19. It shows that among the IA languages outside the Himalayan region the normal system is the use of the same form for both, while the use of separate forms for ‘yesterday’ and ‘tomorrow’ among IA languages is more frequent in the Himalayan region, where they are in contact with ST languages.

One plausible conclusion could be that the contact with ST languages has favored a preservation of the older system in a number of Western Pahari languages, as seen in Table 65 (the boldfaced items). Once again, this seems to be an areal feature, and not a phenomenon exclusive to Kinnauri Pahari.

Figure 19

Words for past and future time adverbs (blue/darker = IA; red/lighter = ST; EQ25B2 = same; ■ = different)

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Figure 19

Words for past and future time adverbs (blue/darker = IA; red/lighter = ST; EQ25B2 = same; ■ = different)

• Download Figure
• Download figure as PowerPoint slide

Figure 19

Words for past and future time adverbs (blue/darker = IA; red/lighter = ST; EQ25B2 = same; ■ = different)

• Download Figure
• Download figure as PowerPoint slide

3.3 Lexicon: Words for ‘mouth’ and ‘face’

Many IA languages have a lexical item which is used for both ‘face’ and ‘mouth’ (Table 66).¹¹ Table 66 includes IA languages from different sub-branches. It shows that the majority of these languages (21 languages) exhibit a polysemous item expressing both ‘face’ and ‘mouth’. The six languages where this polysemy is not attested all belong to the Western Pahari branch of IA (see the Western Pahari section at the bottom of Table 66).

Unlike IA languages, ST languages (both inside and outside Kinnaur) typically have two separate terms for ‘face’ and ‘mouth’ (Tables 67 and 68). In our sample of 25 ST language varieties, only three—Tabo, Tibetan and Zeme—show evidence of this polysemy, reflecting two reconstructed Proto-Sino-Tibetan items *zyal ‘face, mouth’ and *s-muːr ‘mouth, face’, both of which have reflexes with both meanings at least in Written Tibetan.

In general in ST languages the reflexes of *zyal typically mean ‘face’, ‘cheek’, etc., while those of *s-muːr tend to mean ‘mouth’, ‘lip(s)’ or the like. It is worth keeping in mind here that the meaning of the proto-item has been assigned on the basis of the sum of attested meanings in the daughter languages. Thus, it is far from certain that the ‘mouth’–‘face’ polysemy is original to Sino-Tibetan.

Semantically, the meaning extension from ‘mouth’ to ‘face’ is not surprising. According to Wilkins (1996) this is the expected direction of semantic shift. With body-part terms, the semantic development is always from the part to the whole, and never the other way around (i.e., from ‘face’ to ‘mouth’ in this case). In Wilkins’s data, this particular semantic change is attested only in Sino-Tibetan (Wilkins 1996: 276). Still, it does not happen in languages as a matter of course; most languages seem not to have this particular polysemy. But it is widespread among the IA languages.¹²

This semantic shift is extremely rare among ST languages. The IA language Kinnauri Pahari is similar to Kinnauri and other ST languages in this respect (Tables 67 and 68).¹³

Note that while the term for ‘face’ in Kinnauri Pahari (mu) is etymologically related to the IA term for ‘face’ (see Table 66), the term for ‘mouth’ (kʰak) is a borrowing, most probably from Kinnauri. kʰa ‘mouth’ is found in many ST languages.

The non-polysemy that we observe here between ‘face’ and ‘mouth’ in Kinnauri Pahari distinguishes Kinnauri Pahari from the IA pattern, where ‘mouth’ and ‘face’ are usually the same.¹⁴ At the same time, note that several other IA languages (spoken outside Kinnaur), too, exhibit the Kinnauri Pahari/ST pattern (see Table 66)—most of them concentrated in the Himalayan region (see Figure 20).

To summarize this linguistic feature, the data presented here suggest that IA and ST languages typically display two separate patterns in this regard. The typical IA pattern is to have the same form used for ‘mouth’ and ‘face’, whereas the typical ST pattern is to have two separate terms for ‘mouth’ and ‘face’. The IA Kinnauri Pahari (and also some other Western Pahari languages) are similar to the ST languages in this regard, where Kinnauri Pahari has borrowed kʰak ‘mouth’ from ST and has restricted the use of its own lexical item (mukʰ) for ‘face’. As this development is also found in some other Western Pahari languages, once again, this is not a case of an isolated loanword in Kinnauri Pahari, rather the influence is more pervasive.

Figure 20

Words for ‘mouth’ and ‘face’ (blue/darker = IA; red/lighter = ST; EQ25B2 = same; ■ = different)

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Figure 20

Words for ‘mouth’ and ‘face’ (blue/darker = IA; red/lighter = ST; EQ25B2 = same; ■ = different)

• Download Figure
• Download figure as PowerPoint slide

Figure 20

Words for ‘mouth’ and ‘face’ (blue/darker = IA; red/lighter = ST; EQ25B2 = same; ■ = different)

• Download Figure
• Download figure as PowerPoint slide

3.4 Lexicon: Convergence¹⁵ in the Numeral System

It is a well-established fact that in the late stages of Proto-Indo-European the numeral system was a consistent decimal system, where higher decades (e.g. 20, 30, 40, 50, 100) were derived etymologically from the word for 10 by the principle 2 × 10=20, 3 × 10=30, 10 × 10=100 etc. (Winter 1992). This late PIE decimal system was inherited into Proto-Indo-Iranian, and it has carried on in the modern IA languages. The decimal system is found in many modern IA languages. But there are some modern IA languages which display a modified version of the vigesimal counting system (a vigesimal-decimal system where 50, for example, is derived by 2 × 20 + 10).¹⁶

In the Himalayan region, one finds occasional instances of the vigesimal numeral system.¹⁷ Both Kinnauri and Kinnauri Pahari display this pattern, as shown in Table 69.

Figure 21

Numeral systems (blue/darker = IA; red/lighter = ST; EQ25B2 = base 10; EQ25BC = base 20)

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Figure 21

Numeral systems (blue/darker = IA; red/lighter = ST; EQ25B2 = base 10; EQ25BC = base 20)

• Download Figure
• Download figure as PowerPoint slide

Figure 21

Numeral systems (blue/darker = IA; red/lighter = ST; EQ25B2 = base 10; EQ25BC = base 20)

• Download Figure
• Download figure as PowerPoint slide

Some observations can be made here. First, both Kinnauri and Kinnauri Pahari exhibit the vigesimal system. However, while the basic system is the same in both these languages, the forms are not borrowed, only the constructions. Second, among the Western Pahari (IA) languages included in Figure 21, it seems that the numerals and the numeral system in Baghati, Kiunthali, Koṭgarhi and Inner Siraji are very similar to that of Hindi (Bailey 1908, 1920). Koṭgarhi (Hendriksen 1986) and Chinali (D.D. Sharma 1989) are the only languages in my material which show traces of a vigesimal system, even if the forms are built on IA material (Chinali: bῑ ‘twenty’, dui bi ‘forty’, dui bio das ‘fifty’, trāi bi ‘sixty’, trāi bio daš ‘seventy’), even though the default system in Koṭgarhi seems to be the decimal system.

According to Mazaudon (2010), in the Sino-Tibetan language family, the vigesimal system is found in languages only in or close to the Himalayas.²⁰ Among the IA/Iranian languages, the vigesimal system is found not only in the Himalayan region, but it is also found in Central Asia; it is also found in many Iranian languages, in Caucasian languages (Edelman 1999). Both Mazaudon (2010) and Edelman (1999) suggest contact as a possible origin for the vigesimal system in these languages. Thus, to summarize, there is some contact factor involved, but it seems to extend beyond Kinnaur, and also beyond the Himalayas (so far as IA languages are concerned).

3.5 Lexicon/Grammar: the Agentive Nominalizer

Apart from the clear cases of contact-induced changes where the direction of influence is clear, there are also some examples of language change where the two languages have become more similar to each other than they are to their genealogically related languages.

The two languages have a very similar way of forming deverbal agent nouns, as illustrated in Table 71. Further, both languages make a gender distinction here, which is otherwise very uncharacteristic of ST languages.

There is at least one other ST language (Pattani) where -ʦa is used as the agentive nominalizer. In Navakat, the nominalizer -(k)an occurs in similar constructions instead (see Chapter 3 for details). Similarly, Western Pahari languages such as Jaunsari (Satish 2000), too, use a different marker: gɪt-ärɪ ‘singer’ (cf. git ‘song’, gitɪänä ‘to sing’).

This is a clear case of borrowing, but the direction of borrowing is unclear. Note that the Kinnauri Pahari agentive forms contain the element -dɔ/-ndɔ : -di/-ndi. This is the habitual-aspect form, originating in a present participial marker (see Chapter 4). This seems to suggest that the agentive nominalizer in Kinnauri Pahari is a later addition, suffixing to the already participial IA form.

Furthermore, the agentive nominalizer in both languages makes a gender distinction, where -ʦjaː/-sjaː occurs with masculine head nouns and -ʦeː/-seː occurs with feminine head nouns. While there are instances of systematic gender distinctions being made in ST languages, at least in the derivational system (e.g. -pa/-po for male referents vs. -ma/-mo for female referents, found in Navakat and to some extent in Kinnauri), the particular formal means used here are telling. Many IA languages express the masculine–feminine distinction through the use of forms ending in -a/-o in the masculine, contrasting with forms ending in -i/-e in the feminine.²¹ It is possible, that even if the agentive nominalizer itself is the result of ST influence on Kinnauri Pahari, the gender distinction in the agentive nominalization in Kinnauri is due to IA influence.

3.6 Grammar: Perfective and Imperfective Aspect Markers

ST and IA languages in general exhibit two different patterns with respect to the historical source of their modern perfective and imperfective aspect markers. In IA languages this is frequently the participial forms, where the present participial form is reanalyzed as the present/imperfective/habitual aspect marker and the past participial form is reanalyzed as the past/perfective aspect marker. Like a typical IA language, Kinnauri Pahari, too, has reanalyzed its participle forms as aspect markers: -indɛ functions as the perfective aspect marker and as the past participle marker, and -(n)dɔ/-(n)di functions both as the habitual aspect marker and as the present participial marker. This is also corroborated by the other Western Pahari languages presented in Tables 71–72: the neighboring IA varieties have past/perfective markers which are the same as the past participle forms (Table 71) and the present/imperfective aspect markers are the same as the present participle markers (Table 72).²²

Distinct from this, the modern past/perfective and present/imperfective/habitual aspect markers in most ST languages do not come from participles, but from other kinds of nominalization.

Additionally, those ST languages which do exhibit participle-based forms are predominantly spoken in geographical regions where they have been in contact with IA languages for a long time (Saxena 1997b); see Figure 22. This is also the case with Kinnauri. In Kinnauri the two perfective markers are a reduplicated form of the verb and -is, which coincide with the past participle forms (see Chapter 2, Section 4.5.2.2). The habitual (imperfective) aspect markers are -ts and -id, which are the same as the present participle forms (see Chapter 2, Section 4.5.2.3).

Based on these data, some generalizations can be made: While the IA languages consistently show one pattern, where the past participial form and past/perfective aspect markers are the same, among the ST languages, only a few languages (e.g. Thami, Rai, Kinnauri, Kanashi) show the “IA” pattern (i.e., where the perfective aspect marker is the same as the past participial form.); other ST languages retain their indigenous path of grammaticalization. Returning to Kinnauri and Kinnauri Pahari, once again, we find that while the two languages have become more similar with regard to the mechanism used, the forms are not borrowed. Further, once again, this contact-induced feature is not restricted to Kinnauri and Kinnauri Pahari, rather it displays a wider geographical footprint.

Figure 22

Past/perfective same as participle (blue/darker = IA; red/lighter = ST; EQ25B2 = yes; ■ = no)

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Figure 22

Past/perfective same as participle (blue/darker = IA; red/lighter = ST; EQ25B2 = yes; ■ = no)

• Download Figure
• Download figure as PowerPoint slide

Figure 22

Past/perfective same as participle (blue/darker = IA; red/lighter = ST; EQ25B2 = yes; ■ = no)

• Download Figure
• Download figure as PowerPoint slide

3.7 Grammar: the 1pl Inclusive–Exclusive Distinction

Both Kinnauri and Kinnauri Pahari makes the inclusive–exclusive distinction in the first person plural pronouns:

The inclusive–exclusive distinction is brought forth, at times, in discussions on “South Asia as a linguistic area” (e.g. Southworth 1974; Emeneau 1980; Masica 1991, 2001). Among the IA languages, at least the following languages have been mentioned in the literature as having this distinction: Marathi, Gujarati, Sindhi, some Rajasthani varieties, and the Tirupati dialect of Saurashtra (Southworth 1974; Emeneau 1980; Masica 1991, 2001; Osada 2004). In the same vein, it has been pointed out that all three varieties of Marathi, Kannada and Urdu spoken in the Kupwar village exhibit this distinction, where Marathi is suggested to have influenced Kannada and Urdu (Gumperz and Wilson 1971).²³ The presence of this distinction in IA is generally assumed to reflect an areal feature, with Dravidian as the most likely source (Masica 1991).²⁴ Further, all the IA languages with the inclusive–exclusive distinction discussed in the literature exhibit the same path in developing this distinction, where they are said to have reanalyzed the reflexive pronoun as the inclusive form (Masica 2001; Osada 2004).

LaPolla (2005) presents an overview of the inclusive–exclusive distinction in ST languages based on an examination of 170 languages. Out of these, 69 languages make this distinction in one way or another., and it is found in almost all sub-groups today. LaPolla (2005) claims that this distinction cannot be reconstructed for Proto-Sino-Tibetan or for the mid-level reconstruction, rather each individual sub-group seems to have developed this distinction independently.

Kinnauri Pahari seems to be unique among the Western Pahari languages in having this distinction in personal pronouns, a feature which it shares with the coterritorial but unrelated language Kinnauri (Chapter 2),²⁵ as well as with Navakat (Chapter 3). Further, unlike other IA languages, which have this distinction, in Kinnauri Pahari the reflexive form (ap sg, apori pl) shows resemblance, if any, with the 1ple pronoun (aːmɔri)—and not the 1pli pronoun (taːmɔri).²⁶

Once again, we see here that while Kinnauri Pahari and Kinnauri share a pattern, they use two different sets of forms.

3.8 Grammar: the Finite Verb System

Finally, the finite verb system in Kinnauri is structurally similar to the system typically found in IA languages, where the grammatical categories of tense and aspect generally are given separate expression. This is distinct from the system found, e.g. in Navakat, where tense and evidentiality are expressed by portmanteau morphs.²⁷