Chapter 5 Linguistic Relationships in Kinnaur I: Sino-Tibetan

In: The Linguistic Landscape of the Indian Himalayas
Author:
Anju Saxena
Search for other papers by Anju Saxena in
Current site
Google Scholar
PubMed
Close
Open Access

1 Introduction

There has been a general lack of systematic, comparative linguistic studies of the Sino-Tibetan language varieties of Kinnaur (referred to as “KST varieties” in this book). Some comparative data are found in older works (e.g. Gerard 1842; Cunningham 1844; Bailey 1909). More recent works on the languages of this region (e.g., Neethivanan 1976; D.D. Sharma 1988; Saxena 1992, 1995b, 1997b, 2002, 2007, 2017; Takahashi 2001, 2007, 2012; Negi and Negi 2015; Negi 2017) have generally focused their attention on the linguistic analysis of one specific KST variety, the speech of Lower Kinnaur (Sangla, Pangi, Kalpa), the main exceptions being some work on Chhitkuli (Martinez 2019, 2021), and on the Middle Kinnaur variety Shumcho (Huber 2007, 2014a, 2014b) and a very brief “language snapshot” (descriptions of genealogy and sociolinguistic status) of Sunam by Negi (2020). Consequently, we have had no good grounds for examining how the various KST varieties relate to one another. The closest thing to such a study that I am aware of are the sociolinguistic surveys by Webster (1991) and Chamberlain et al. (1998).

This chapter presents such an investigation based on data collected in a questionnaire-based study carried out in Kinnaur. The KST varieties examined here represent the speech of nine villages located in different parts of Kinnaur. The results of the study are then compared with existing accounts of Sino-Tibetan languages in Kinnaur and their classification.

Summarizing briefly the results that are presented in detail below, the investigated KST varieties can be classified into three (or possibly four) groups, where the KST varieties spoken in Sangla, Nichar, Ropa and Kalpa (referred to below as the Sangla group or Kinnauri; see Chapter 2) form one externally distinct and internally cohesive group. The KST varieties spoken in Poo, Kuno and Nako (referred to as the Nako group or Navakat; see Chapter 3) form another clear grouping. The KST varieties of Chitkul and Labrang fall somewhere in between, where Chitkul and/or Labrang are more similar to one or the other group concerning some linguistic features, but with regard to other linguistic features Chitkul and/or Labrang behave distinctly from both Kinnauri and Navakat. At the same time, Chitkul and Labrang are not close enough to each other that we could say that they jointly make up a third grouping.

d155303993e212044

Figure 15

Location of the villages in Kinnaur for which data was collected

Robinson Projection. Map design: Ljuba Veselinova

2 Data Collection

Data was collected representing the speech of the following villages in Kinnaur: Nichar, Sangla, Chitkul, Kalpa, Kuno, Labrang, Poo, Ropa and Nako (shown on the map in Figure 15).1 The main motivation for selecting the speech of these villages was to include a representative range of language data from as diverse geographical regions as possible. Table 36 contains basic information on these villages. In general, Kinnaur is multilingual (see Chapter 1), and in several places, different traditional social groups in a village are known to use separate languages (Huber 2014b: 194 f.). In such cases, the KST variety discussed here reflects the speech of the majority group in that village. For the purposes of the study presented in this chapter the investigated KST varieties will consistently be referred to by the names of the villages where the corresponding KST varieties are spoken: for example, “Sangla” rather than “(Sangla) Kinnauri” and “Nako” rather than “Navakat” or “Bhoti Kinnauri”.

Table 36

Basic information on the villages (ordered south to north) for which data was collected

Village (tahsil; coordinates)

Some information about the village

Chitkul (Sangla; 31° 21′ N, 78° 26′ E)

Located in Sangla valley on the right bank of the Baspa river. It is the highest village in the Sangla valley (3,450 m).

Sangla (Sangla; 31° 25′ N, 78° 15′ E)

Located in Sangla valley on the right bank of the Baspa river.

Kalpa (Kalpa; 31° 32′ N, 78° 15′ E)

Located in Satluj valley. The Kalpa village was earlier the district capital of Kinnaur.

Nichar (Nichar; 31° 33′ N, 77° 59′ E)

Located in Satluj valley between Taranda and Wangtu, on the right bank of the Satluj river.

Kuno (Morang; 31° 38′ N, 78° 22′ E)

Located in Satluj valley.

Labrang (Poo; 31° 41′ N, 78° 26′ E)

Located in Satluj valley.

Poo (Poo; 31° 46′ N, 78° 35′ E)

Located in Satluj valley.

Ropa (Poo; 31° 48′ N, 78° 25′ E)

Located in Ropa valley.

Nako (Poo; 31° 53′ N, 78° 37′ E)

Located in Hangrang valley. It is the highest village in the valley (3,600 m).

Since the comparison of the KST varieties will be based primarily on a lexicostatistical investigation of basic vocabulary, the longer version of the Swadesh basic vocabulary list (207 entries; Swadesh 1950, 1952, 1955) was used as the point of departure for preparing our primary questionnaire. The Swadesh list was, however, modified extensively. This included both removing almost a third of the entries in the Swadesh list as expressing concepts not suitable for this region for pragmatic reasons (e.g., some entries expressing concepts connected with the ocean), and instead adding a number of entries important for the present study (e.g., numerals, the honorific–non-honorific distinction in pronouns, reflexive pronouns). The length of the list increased somewhat, resulting in a concept list for the primary questionnaire with 237 entries. The complete list can be found in Appendix 5A to this chapter. Some items designed to elicit noun phrases and some sentence types were also included in the questionnaire, to examine, for example, the order of constituents at the phrase and clause levels, and also to examine the reflexive construction. In addition to the entries in the questionnaire, some additional data were also collected in each case, e.g., data on pronominal possessive constructions, example sentences to understand the linguistic status of a lexical item, as well as other lexical items, to understand the relationship of the lexical item in question to other words in the same semantic field. In the case of Kinnauri and Navakat (as well as Indo-Aryan Kinnauri Pahari), we also collected lexical data based on the longer (1,460 entries) loanword typology list (Haspelmath and Tadmor 2009; Borin et al. 2013). These lists are provided in the chapters on Kinnauri (Chapter 2, Appendix 2A; 1,348 items), Navakat (Chapter 3, Appendix 3B; 1,135 items), and Kinnauri Pahari (Chapter 4, Appendix 4B; 1,215 items). All data items were trancribed in a broad phonetic transcription.

3 Methodology

The present investigation falls under the heading of lexicostatistics, a long tradition of describing and (implicitly or explicitly) quantifying similarities and differences among language varieties using basic vocabulary lists. For an overview, see the chapters in Borin and Saxena (2013), especially Borin (2013). A revised Swadesh list has been the main basis for comparison of the KST varieties examined here (see Section 2). Using such concept lists presents its own methodological challenges (Borin 2012; Borin et al. 2021). A fundamental decision in this context is whether a particular concept is represented by the same item (word) in two language varieties.

Here we must first define what we mean by “the same item”. In Swadesh-style lexicostatistics, this is normally interpreted as cognacy, i.e., whether the items are reflexes of the same proto-language item. Finer points of (derivational) morphological structure are often disregarded in this context, and only cognacy of roots or stems is important. Even in this case, determining that two items are cognate is far from straightforward and requires expert knowledge, especially if the languages are only distantly related.

This arguably means that the information about genealogical grouping sought by these methods, to a large extent is already known by other means, e.g., the classical comparative method. The requisite expert knowledge is a scarce resource, and if we would like to conduct larger-scale genealogical linguistic investigations encompassing also poorly documented languages, we need some other way of doing this. When the expert knowledge is available, it serves as a valuable yardstick, a known gold standard against which less knowledge-intensive methods can be judged before being applied to those cases where less is known beforehand.

Lexicostatistical investigations such as that presented by Holman et al. (2008) rely on a mechanical procedure—automatically computed Levenshtein distance (also called edit distance) between strings transcribed using a standardized coarse phonetic transcription—for determining cognacy. This has the advantage of being totally consistent, and the disadvantage of both missing some cognates and misclassifying some non-cognate pairs as cognates. However, the primary, most important requirement on such a method is that it is repeatable and objectively verifiable.

In dialect studies, the judgement of sameness may include also the sound shape and morphological structure of obviously cognate items in the sense of the preceding paragraph. This is the method chosen here when comparing the Kinnauri basic vocabulary lists: Certain—but not all—sound correspondences, and certain—but not all—morphological structures, are considered equal for the purpose of comparing lexical items among KST varieties.

A frequent presupposition in Swadesh-style lexicostatistics is that only one word from each language will represent each concept in the list. Here, we do not impose this restriction, however. Generally, with larger-scale investigations involving poorly documented language varieties that the researcher may not know well, this seems to be the only feasible alternative. In working with secondary sources and language consultants, presumably we will end up with one or several common expressions of the concept sought, regardless of their genealogical relationship to the corresponding expressions in related language varieties.

In our investigation, every correspondence gets one point, but multiple correspondences for the same concept still count as only one correspondence. Let us assume that a particular concept is expressed in the following way in four languages (capital letters represent forms/words):

Language 1

Language 2

Language 3

Language 4

A

A, B

A, B

B

With this way of calculating similarity, languages 2 and 3 are as similar to each other as each of them is to language 1 and 4, although languages 2 and 3 share two items in this concept slot. This solution is not completely arbitrary, but not very strongly motivated either. However, it can easily be reconsidered—e.g., if more information becomes available on these language varieties—and the results recalculated. The main point to be made here is that the calculation is completely deterministic and repeatable.

As has perhaps become clear from the preceding, compiling comparable systematic linguistic data for the present investigation has presented something of a challenge. One complicating factor here is that a language can have more than one word for a concept, and it is largely fortuitous which alternative or alternatives the language consultants provide (Slaska 2005). Thus the data presented in the present work cannot be seen as complete. There may exist terms in a KST variety which have cognates in other varieties, which however do not happen to show up in our material. On the other hand, the terms provided by a consultant may say something about terms which are more neutral or more frequent or more basic than the other possible alternative forms which were not provided.

4 Towards Linguistically Informed Computational Lexicostatistics

The following proceedure was used in this investigation, developed in collaboration between a computational linguist (Lars Borin) and the author (see also Saxena and Borin 2011, 2013):

  • After the data collection and initial processing of the data,

  • a list of observations of relationships among varieties was made by the author.

  • This list formed the basis for developing a set of principles for comparing the linguistic correspondences in these KST varieties. These were formulated by the author and the computational linguist together and their purpose was to determine which segmental differences to disregard for the purpose of considering items in different varietes the same.

  • The principles were encoded by the computational linguist as context-sensitive phonetic segment transformation and equivalence rules in a small computer program for comparing items fully automatically in order to achieve consistency.2

  • The program was then applied to the data, the result inspected, the rules revised, and the modified program run again on the data. This process went through a few iterations.

The procedure is a variant of automated lexicostatistics, a methodology that has seen a strong revival in recent years (see Borin and Saxena 2013), but in our case with a clear qualitative element (somewhat in the spirit of Grant 2010). Rather than adopting the standard solution of designing a completely automated method applying a similarity metric to orthographic words, we have endeavored to include linguistic information into the process at an early stage. The results from the comparison come in the form of two kinds of tables:

  • tables of individual concepts and lexical items expressing them, where each language variety gets a numerical index (1–9), and each concept/language variety combination is provided with a list of indices showing which varieties share one or more expressions of this concept (see Appendix 5A to this chapter);

  • summary tables, where similarities among all lexical items of a particular grammatical or semantic category (nouns, kinship terms, etc.) are shown as ratios and percentages (see Section 5).

In the present investigation, the following principles were used in comparing word list items among varieties (in the list below, the following symbols are used: C: consonant; V: vowel; T: stop; Ø: zero/no segment).

Vowels: The following vowels appear to be in free variation in many of these varieties, and consequently the two members of each pair are considered equal for the purposes of our comparison, in any position:

a ~ ǝ; a ~ ɔ; i ~ ɪ; u ~ ʊ; e ~ ɛ; o ~ ɔ; o ~ ø

Note however that the similarities are not to be construed as transitive: e.g., ǝ and ɔ do not count as the same.

Vowel length: Long and short vowels are not distinguished for the purposes of the comparison.

Vowel nasalization and phonemic tone: Nasalization is disregarded in the comparison, as is tone (orthographically marked on vowels in the transcription).

Consonants: The following consonants appear to be in free variation in many of these varieties, and consequently the members of each group are considered equal, in any position:

ʤ ~ ʣ ~ z ~ ʒ; p ~ pʰ ~ f; ʧ ~ ʧʰ ~ ʈʃ; s ~ ʦ ~ ʦʰ

Consonant gemination: Short and long consonants are treated as one and the same, in any position:

C1ː ~ C1

In the preliminary analysis of the sound systems of these varieties, there has been no indication that geminates are phonemic in any of them.

Prenasalization: Prenasalization of consonants is disregarded in the comparison.

Unreleased stops: Unreleased stops are treated as equal to the corresponding fully released stops in the comparison, ignoring voicing.

Sound sequences: The following sequences will be treated as equal for the purposes of the comparison, in any position:

ɖr ~ ɖ; ʈr ~ ʈ; V1jV2 ~ V1V2

Word endings: The following word ending alternants will be treated as equal for the purposes of the comparison:

-h ~ -Ø; -ʦ ~ -Ø; -j ~ -Ø;
-pa ~ -ba ~ -ʋa; -po ~ -bo ~ -ʋo;
-V1T ~ -V1

Illustrating with a concrete example, the last item in this list states that word-final stops are counted equal to Ø following a vowel, as there is dialect-internal variation in this respect. Different stops are considered as separate, however. Thus, ja counts as the same as both jag and jak, but the latter two count as different forms (see yak in Table 54 in Appendix 5A).

Phrases: For terms such as older brother, younger brother, maternal aunt, paternal aunt, if the term consists of more than one word, e.g., ‘old sister’, then the modifier is disregarded; only the noun is used for the comparison.

In order to achieve consistency of judgement, the above principles were encoded in a small computer program which then was used to compare items fully automatically. In practice, the principles were initially manually developed and then successively refined by an iterative process where the program was applied to the data and the results subsequently inspected. Typically during such a round we would find that the program had missed some correspondence that should have been found. Because the principles tended to be fairly conservative, the opposite almost never occurred. The great advantage of having automated the application of the principles emerged in these situations, since a revision of the principles made on the basis of one or a few correspondences could be immediately tested on all the data in order to check that it would not introduce errors elsewhere.

This methodology is similar to recent work in dialectometry (e.g., Nerbonne and Heeringa 2009) and lexicostatistics (e.g., Holman et al. 2008; Wichmann et al. 2010) in relying on a completely automatic comparison of the items in the word lists. However, it differs from most of this work—a notable exception being the work reported on by McMahon et al. (2007)—in its usage of rules tailored to the particular linguistic configuration under investigation, rather than a general method for string comparison. In this respect, it falls somewhere in between traditional lexicostatistics—where expert statements are required about the cognacy of items—and these modern approaches—which rely entirely on surface clues for determining identity of items—although closer to the latter than the former.

The main methodological advantage of the approach used here is its consistency, and not as claimed for the work just referred to, that it should be language-independent. Instead, in our work we have tried to apply a principle sometimes formulated in computational linguistics as “Don’t guess if you know” (Tapanainen and Voutilainen 1994: 47), which inevitably leads us to include language-specific knowledge in the form of correspondence rules among dialects.

5 Results: Vocabulary

In this section we will examine how much of basic vocabulary the investigated KST varieties share. We will look at the following kinds of basic vocabulary: a set of open-class words (nouns and adjectives),3 some adverbs of time, numerals and numeral systems, question words, and personal pronouns. Among the nouns, kinship and body-part terms are investigated separately.

In the vocabulary correspondence tables presented in this section we use the following notational conventions. Abbreviations (italicized in the tables) are used for the village names: Sangla (Sa), Nichar (Ni), Kalpa (Ka), Ropa (Ro), Chitkul (Ch), Labrang (La), Poo (Po), Kuno (Ku), Nako (Na). The full correspondence tables are found in Appendix 5A at the end of this chapter (Tables 52–59). Vocabulary items refer to concepts and are identified by English words (or phrases on a few occasions) in small caps, both in the text and in the tables in Appendix 5A. Swadesh list items are further identified by their Swadesh list number added to the end of the English word and separated from the word by a slash: laugh/100. Items without a number do not appear in the Swadesh list. There are 88 Swadesh list concepts in the questionnaire (see Appendix 5A). If a Swadesh list item is marked with an asterisk, this means that the item is in the subset of 40 Swadesh list items found to be the most stable globally by Holman et al. (2008). There are altogether 25 out of these 40 items in the questionnaire (see Section 5.7).

The longer noun and adjective tables (Tables 54 and 55 in Appendix 5A) are arranged with the English words in alphabetical order. The other tables are arranged according to other principles (semantically or by Swadesh number). In the correspondence tables, numerical indices in square brackets appear in each cell to identify the language varieties which share a form for this concept, i.e. items considered the same according to the formal principles presented above in Section 4. Multiple indices in the same cell are separated by slashes.

Each subsection below is structured in a similar way. One or more tables are presented containing summary statistics on shared vocabulary between all pairs of varieties, calculated from the full correspondences presented in Tables 52–59 in Appendix 5A. Two figures are provided for each pairwise comparison: a fraction and a percentage (rounded to an integer). In the fraction, the denominator represents the total number of concepts where some form is recorded for both varieties (for a number of reasons, sometimes a particular concept has not been recorded for some variety), and the numerator indicates how many of these forms that have been computed to be the same by the automatic procedure. Finally, we discuss some salient linguistic points of the comparison.

5.1 Basic Nouns

5.1.1 Kinship Terms

Table 52 shows the investigated kinship terms and the automatically computed correspondences among varieties, and Table 37 contains the summary statistics extracted from Table 52.

Table 37

Summary statistics for kinship terms

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

12/18 (66 %)

13/18 (72 %)

9/18 (50 %)

10/18 (55 %)

6/18 (33 %)

1/18 (5 %)

1/18 (5 %)

1/18 (5 %)

Ni

9/18 (50 %)

6/18 (33 %)

7/18 (38 %)

3/18 (16 %)

0/18 (0 %)

0/18 (0 %)

0/18 (0 %)

Ka

9/18 (50 %)

8/18 (44 %)

5/18 (27 %)

1/18 (5 %)

1/18 (5 %)

1/18 (5 %)

Ro

7/18 (38 %)

7/18 (38 %)

5/18 (27 %)

4/18 (22 %)

3/18 (16 %)

Ch

4/18 (22 %)

1/18 (5 %)

1/18 (5 %)

1/18 (5 %)

La

6/18 (33 %)

6/18 (33 %)

5/18 (27 %)

Po

13/18 (72 %)

13/18 (72 %)

Ku

13/18 (72 %)

We will now look more closely at some of the individual kinship terms.

grandfather: Nako, Poo and Kuno use the term meme for grandfather, while the other varieties use another term, tete. A modifier is added to specify maternal relationship in some varieties. Nichar, Kalpa, Ropa and Chitkul add this additional component. In all varieties where it appears it precedes the base form, and seemingly related forms (maperɔŋ, mapɔ and maʧa) are used. This modifier occurs also in the terms for maternal grandmother in the same varieties.

grandmother: It is plausible that the terms for grandmother in all these varieties has the same origin: In Sangla, Kalpa, Ropa, Chitkul and Labrang it is api, in Nichar it is ai, and in Nako, Poo and Kuno it is aʋi.

mother: The same term occurs in all varieties for mother. It is ama, except in Nichar, where it is .

father: It is plausible that the terms for father in all these varieties have the same origin, but has developed in three different ways, classifying these varieties in three groups. The term for father in Sangla, Nichar and Kalpa is boʋa/baba/bɔba. In Ropa, Labrang, Poo, Kuno and Nako it is apa/aʋa and in Chitkul we find au, presumably related to aʋa. au also occurs as an alternate form in Kuno. The terms for mother and father in all KST varieties are etymologically related. They are: ama and (b)aba/aʋa (with the possible exception of Chitkul au if unrelated to aʋa).

husband and wife: Except for some similarities in the terms for husband in some varieties, the terms for husband and wife do not exhibit a consistent pattern. This may be partly due to the fact that there are several different ways of referring to the person who is a husband/wife, thus it is possible that different language consultants have provided different terms.

brother and sister: The terms for older brother classify these varieties into two groups. In Sangla, Nichar, Kalpa and Chitkul it is ate, while in Ropa, Labrang, Poo, Kuno and Nako it is aʧo/aʒo. It seems that there are several terms for younger brother in each variety, with different social functions. Some of these terms are borrowed from Indo-Aryan languages (e.g baja and other related terms in Table 52). The same is true also about the terms for younger sister and older sister (including the use in many varieties of an Indo-Aryan term).

son and daughter: The terms for son and daughter classify these varieties into three groups. Sangla, Nichar, Kalpa, and Labrang have the terms ʧʰaŋ and ʧimɛd; Nako, Poo and Kuno have the terms ʈuː and pomo and Chitkul has the terms ɖe aʧi and ɖju aʧi, son, daughter, respectively.

uncle and aunt: In Sangla, Nichar, Kalpa, Chitkul and Labrang, an Indo-Aryan loanword is used for maternal uncle, viz. mɔma, whereas in Nako, Poo, Kuno and Ropa, the term is aʒaŋ. The words for paternal uncle at least in some cases are probably related to the terms for father. It seems that the terms for paternal aunt in most of the varieties have the same origin, which has developed in three different ways: nane in Sangla, Nichar and Kalpa, ane in Labrang, Poo, Kuno and Nako, and ene in Chitkul. Only Ropa exhibits a divergent term: ʦima.

To summarize, looking at the kinship terms we can clearly differentiate a core Sangla group (Sangla, Nichar and Kalpa) from a core Nako group (Nako, Poo and Kuno), where these groups differ from each other regularly and consistently in all cases when the same term is not used in all varieties. With regard to the kinship terms Chitkul is generally similar to the Sangla group. Labrang and Ropa present interesting cases. In some cases (though not in identical cases) Labrang, for instance, has terms which are similar to the terms found in the Sangla group (e.g., the terms for grandmother, son, daughter, maternal uncle), but with regard to other terms (e.g., the terms for grandfather, father and brother) it has terms which are similar to the terms found in the Nako group.

5.1.2 Body Parts

Table 53 shows the investigated basic body part words and the automatically computed correspondences among varieties, and Table 38 contains the summary statistics extracted from Table 53.

Table 38

Summary statistics for body part terms

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

8/11 (72 %)

10/11 (90 %)

10/11 (90 %)

2/11 (18 %)

3/11 (27 %)

1/11 (9 %)

1/11 (9 %)

1/11 (9 %)

Ni

8/11 (72 %)

8/11 (72 %)

2/11 (18 %)

3/11 (27 %)

1/11 (9 %)

1/11 (9 %)

1/11 (9 %)

Ka

10/11 (90 %)

2/11 (18 %)

3/11 (27 %)

1/11 (9 %)

1/11 (9 %)

1/11 (9 %)

Ro

2/11 (18 %)

3/11 (27 %)

1/11 (9 %)

1/11 (9 %)

1/11 (9 %)

Ch

2/11 (18 %)

1/11 (9 %)

1/11 (9 %)

1/11 (9 %)

La

1/11 (9 %)

1/11 (9 %)

1/11 (9 %)

Po

8/11 (72 %)

10/11 (90 %)

Ku

8/11 (72 %)

Generally, these KST varieties display the same Sino-Tibetan cognate forms for the terms for eye, mouth and hair. Concerning the term for hair in these varieties, all of them exhibit reflexes of the same proto-item, reconstructed as *kra for Proto-Sino-Tibetan. This item is realized in two different ways, however: kra and ʈa, the latter occurring in Poo, Kuno and Nako, while the former occurs in all the other varieties. The correspondence kr ~ ʈ reflects a deeper (in time) sound change than what the automatic correspondence rules used here are meant to capture. Hence, in Table 53, the varieties are classified into two groups with respect to the item hair.

Perusing Table 53, it is quite clear that in those cases where the KST varieties do not share a body part vocabulary item, the Sangla group and the Nako group consistently use different sets of terms.

Labrang and Chitkul fall somewhere in the middle, where they sometimes show more similarities to the forms in the Sangla group (e.g., foot and hand), while in other cases they show more similarities with the forms in the Nako group (e.g., tooth). Chitkul and Labrang form a separate group with regard to the terms used for head, eye, tail and face. There are also cases where Labrang and Chitkul use terms which they neither share with each other nor with any of the other two groups (e.g. nose). Apart from this, there are some terms either in Labrang (e.g. hand, foot) or in Chitkul (e.g., hand) which are unique.

On the whole, the pattern which emerges here is similar to the one as observed above, where Sangla, Nichar and Kalpa form a group—but now clearly with Ropa, too, belonging in the Sangla group—and Poo, Kuno and Nako form another group, with Chitkul and Labrang standing out as different from both the Sangla and Nako group and from each other.

5.1.3 Other Basic Nouns

Table 54 shows the investigated other basic nouns—i.e., other than kinship terms and body parts—and the automatically computed correspondences among varieties, and Table 39 contains the summary statistics extracted from Table 54.

Looking at the larger data set of Table 39, we again find the earlier two clear groupings: (1) Sangla, Nichar, and Kalpa; and (2) Poo, Kuno, and Nako. Ropa appears as slightly closer to the Sangla group than Chitkul is, whereas Labrang emerges as distinct from both the Sangla and Nako groups, although closer to the former.

Table 39

Summary statistics for basic nouns

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

47/58 (81 %)

49/59 (83 %)

39/59 (66 %)

35/59 (59 %)

22/59 (37 %)

9/59 (15 %)

12/58 (20 %)

11/58 (18 %)

Ni

44/58 (75 %)

34/58 (58 %)

29/58 (50 %)

20/58 (34 %)

9/58 (15 %)

11/57 (19 %)

11/57 (19 %)

Ka

38/59 (64 %)

34/59 (57 %)

21/59 (35 %)

8/59 (13 %)

11/58 (18 %)

10/58 (17 %)

Ro

29/59 (49 %)

27/59 (45 %)

11/59 (18 %)

14/58 (24 %)

13/58 (22 %)

Ch

22/59 (37 %)

9/59 (15 %)

11/58 (18 %)

10/58 (17 %)

La

15/59 (25 %)

18/58 (31 %)

16/58 (27 %)

Po

35/58 (60 %)

39/58 (67 %)

Ku

36/58 (62 %)

Again we find cases where the simple automatic word comparison seems to miss obviously related words (e.g., egg, star, winter) but this does not in itself mean that we need to revise the comparison rules (see Section 5.8).

Table 40

Summary statistics for basic adjectives

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

14/19 (73 %)

11/19 (57 %)

11/19 (57 %)

2/19 (10 %)

2/19 (10 %)

1/19 (5 %)

1/19 (5 %)

1/19 (5 %)

Ni

12/19 (63 %)

10/19 (52 %)

1/19 (5 %)

1/19 (5 %)

0/19 (0 %)

0/19 (0 %)

0/19 (0 %)

Ka

12/19 (63 %)

1/19 (5 %)

1/19 (5 %)

0/19 (0 %)

0/19 (0 %)

0/19 (0 %)

Ro

2/19 (10 %)

3/19 (15 %)

1/19 (5 %)

0/19 (0 %)

1/19 (5 %)

Ch

9/19 (47 %)

1/19 (5 %)

0/19 (0 %)

1/19 (5 %)

La

1/19 (5 %)

0/19 (0 %)

1/19 (5 %)

Po

12/19 (63 %)

14/19 (73 %)

Ku

11/19 (57 %)

5.2 Basic Adjectives

Table 55 shows the investigated basic adjectives and the automatically computed correspondences among varieties, and Table 40 contains the summary statistics extracted from Table 55.

The adjectives, too, confirm the grouping that we have observed above. Even though the data set is small, the trend is obvious: Poo, Kuno and Nako form one group, and Sangla, Nichar, Kalpa and Ropa form another group. This is very clear for the majority of the adjectives in Table 55. Again, Chitkul and Labrang stand apart: In some cases a similar form occurs in both languages (e.g. fɔsi dry in both Labrang and Chitkul and also some of the color terms). But there are also cases (e.g., good, wet) where separate forms occur in Labrang and Chitkul. If the forms in Labrang and Chitkul show similarity with any of the two clearer groupings, it is rather with the Sangla group than the Nako group; see, e.g., the terms for beautiful, old and new.

5.3 Some Adverbs of Time

Table 56 shows the investigated adverbs of time and the automatically computed correspondences among varieties, and Table 41 contains the summary statistics extracted from Table 56.

Table 41

Summary statistics for time adverbs

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

4/10 (40 %)

8/10 (80 %)

4/10 (40 %)

0/8 (0 %)

2/9 (22 %)

0/9 (0 %)

0/8 (0 %)

0/10 (0 %)

Ni

5/10 (50 %)

5/10 (50 %)

0/8 (0 %)

3/9 (33 %)

0/9 (0 %)

0/8 (0 %)

0/10 (0 %)

Ka

5/10 (50 %)

0/8 (0 %)

2/9 (22 %)

0/9 (0 %)

0/8 (0 %)

0/10 (0 %)

Ro

0/8 (0 %)

2/9 (22 %)

0/9 (0 %)

0/8 (0 %)

0/10 (0 %)

Ch

0/7 (0 %)

0/8 (0 %)

0/7 (0 %)

0/8 (0 %)

La

0/8 (0 %)

0/8 (0 %)

0/9 (0 %)

Po

4/8 (50 %)

6/9 (66 %)

Ku

4/8 (50 %)

This material is too small to draw any conclusions beyond the fact that it supports the same groupings of the language varieties as the previously presented vocabulary subsets. The time expressions are a bit too complex for the simple mechanical comparison used here to work well. Manual inspection of the expressions shows some fairly obvious connections which are not captured by the rules, e.g., Sangla rɪgʦɔmja versus Nichar/ Kalpa/ Ropa rɪkʦɔmja/rikʦɔmja 2 days before tomorrow.

This set of terms seems to classify the KST varieties into roughly the same groups as other lexical-semantic fields discussed in this chapter, Sangla, Nichar, Ropa, Kalpa form one group. Concerning the terms for future time points too, these languages are similar to one another. They form one group. All languages in this group make (at least) a five-way distinction in the future (tomorrow, 1–4 days after tomorrow) and the terms used to express these concepts in these languages are also very similar.

Generally speaking, Poo and Nako form another group, though they also differ slightly from each other—both in terms of the number of distinctions made lexically in referring to the past and to the future, as well as the forms used. Nako has a more detailed system, with separate lexical terms for up to 4 days before yesterday and 4 days after tomorrow, whereas Poo has distinct terms for up to 2 days before yesterday and 2 days after tomorrow. Despite this difference, the forms (when the distinction is there in both languages) are quite similar in Nako and Poo. The manual and automatic analysis agree with respect to the positions of Labrang and Chitkul: If Labrang displays any similarity with any of the other groups, it is with the terms found in the Sangla group, e.g., in the terms for today, 1 day after tomorrow and 3 days after tomorrow. Chitkul, which exhibits a detailed system in this regard, does not show similarities with any of the other varieties.

5.4 Numerals and Numeral Systems

Table 57 shows the investigated numerals and the automatically computed correspondences among varieties, and Table 42 contains the summary statistics extracted from Table 57.

Table 42

Summary statistics for KST numerals

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

17/25 (68 %)

16/25 (64 %)

16/25 (64 %)

13/25 (52 %)

7/25 (28 %)

1/25 (4 %)

2/25 (8 %)

2/25 (8 %)

Ni

17/25 (68 %)

16/25 (64 %)

12/25 (48 %)

6/25 (24 %)

1/25 (4 %)

2/25 (8 %)

2/25 (8 %)

Ka

18/25 (72 %)

15/25 (60 %)

7/25 (28 %)

1/25 (4 %)

2/25 (8 %)

2/25 (8 %)

Ro

14/25 (56 %)

7/25 (28 %)

1/25 (4 %)

2/25 (8 %)

2/25 (8 %)

Ch

5/25 (20 %)

1/25 (4 %)

2/25 (8 %)

2/25 (8 %)

La

3/25 (12 %)

5/25 (20 %)

4/25 (16 %)

Po

18/25 (72 %)

16/25 (64 %)

Ku

17/25 (68 %)

The examination of the numerals 1–10 suggests a similar grouping of the KST varieties as observed above, where Sangla, Nichar, Kalpa, and Ropa constitute one group and Poo, Kuno and Nako constitute another group. Except for gɛt eight in Labrang, Chitkul and Labrang numerals are similar to the forms found in the Sangla group. The numerals 1–10 in the KST varieties are cognate to a very large extent (see Table 43 below). They are consistent with the Sino-Tibetan numeral forms noted by Hodson (1913).

For the numerals two, three, five, six and nine the same cognates are found in all varieties (with some phonological modifications). The case of the numeral three is interesting: Even though the same cognate occurs in all varieties, it is realized in three different ways: Sangla, Nichar, Kalpa and Ropa form one group (ʃum/ʂum), Chitkul and Labrang form another group (homo/hʊm) and Nako and Poo form a third group (sum). For the numerals one, four, seven, eight and ten these varieties use two distinct cognate forms: Poo, Kuno and Nako agree among themselves and use the same form as is noted by Hodson (1913) for Central Tibetan (namely, ʧɪk, ji/ʒik, dʊn, get/gjat, respectively), Nichar, Sangla, Kalpa, Ropa, Chitkul and Labrang use another set of forms (namely, ɪd, pa, (s)tɪʃ, rɛ/rajɛ, sɛ/sajɛ, respectively). This set of forms, too, is noted by Hodson (1913). In Table 43, the forms for the numerals 1–10 in the KST varieties are shown together with the reconstructed Proto-Sino-Tibetan (PST) forms for these numerals (Matisoff 2003).

A similar subgrouping pattern emerges also concerning the formation of higher numerals in the KST varieties. Generally speaking, two different systems for forming the numerals 20–99 are found in these varieties. Sangla, Nichar and Kalpa form one group. They exhibit a vigesimal system, i.e., one where the basic units are multiples of twenty. Multiples of ten which are not also multiples of twenty (thirty, fifty, seventy, ninety) are indicated as ‘plus ten’, with one exception: The term for fifty in Ropa is is nɪʃ nɪʣʊ adʰaŋ (‘two twenty half’). Concerning all other higher numerals, Ropa is consistent with the pattern (and forms) of the Sangla group. Nako and Poo, on the other hand, exhibit a consistent decimal system. Labrang is interesting in this regard. It shows a decimal system for 30, but for higher multiples of ten it exhibits the same kind of vigesimal system as in the Sangla group.

Table 43

Numerals 1–10 in KST varieties in comparison with reconstructed Proto-Sino-Tibetan (PST)

Sa

Ni

Ka

Ro

Ch

La

Po

Ku

Na

PST

1

ɪd

ɪd/i

i

ʧikEQ031A

*t(y)ak ~ *gt(y)ik; *ʔit

2

nɪʃ

niʃi

niʃ

ɲiː

*gnis

3

ʂʊm

homo

hʊm

sum

*gsum

4

pa/pǝ

ʒi

*blǝy

5

ŋa

*bŋa, lŋa

6

ʈug

ʈukEQ031A

ʈu

ʈʰok

ʈukEQ031A

*druk, *kruk

7

(s)tɪʃ

tiʃ

ʃiniʃ

dun

*snis

8

rajɛ/raje

rea

gɛtEQ031A

gjet

*brgyat ~ *bgryat

9

gui

sgui

gui

gu

gu

*dgǝw, *skǝw

10

sajɛ/saje

sja

sa

ʧu

*g(y)ip; *ʦ(y)i(y) ~ *ʦyay

The numeral system in Kuno distinguishes itself remarkably from the systems found in the other varieties. First, Kuno has both a decimal and a vigesimal system side by side.4 In the vigesimal system there are important differences between the patterns exhibited in Kuno and in the Sangla group. This concerns both the ordering of smaller numerals in forming higher numerals (e.g 2 × 20 in Sangla, but 20 × 2 in Kuno) as well as the structure of higher numerals in Kuno and in the Sangla group. In Kuno ʋa and naŋ occur in higher numerals, ʋa indicating multiplication and naŋ addition, so that, e.g., fifty is literally expressed as ‘twenty times (ʋa) two plus (naŋ) ten’ in Kuno, whereas it is ‘two twenty ten’ in the Sangla group. However, in the decimal system also used in Kuno, the order is multiplier–base–addend, as in all the other varieties: dunʧu ʧik [seven.ten one] seventy-one (also ɲiʃuʋa sumnaŋ ʧugʃɪk [twenty.ʋa three.naŋ eleven]).

Two separate terms (ra and gja/gʰeja) occur for the numeral 100 in the KST varieties. ra occurs in Sangla, Nichar, Kalpa and Chitkul and gja/gʰeja occurs in Labrang, Poo, Kuno and Nako. According to Hodson (1913), both ra and gja are variations of the Central Tibetan form rgya. In Hodson’s view, this form cannot be analyzed as forming part of a decimal or vigesimal system, instead it is a separate distinct form.

Interesting differences are observable in the composition of the words for 500, 1,000 and 1,001 between the Sangla group (including in this case Labrang and Chitkul) and the Nako group. The order of constituents is 5 × 100 for 500 in all varieties. The term for 1,000 in the Sangla group is hǝzar (which is a loanword from Indo-Aryan), but it is tɔn in the Nako group.5

Despite these differences, all KST varieties (except the Kuno vigesimal system) examined here form their composite numerals in the same way. When higher numerals are made by multiplication, the multiplier precedes the base, regardless of whether the variety uses a decimal or vigesimal numeral system.6 For example, forty will be expressed as 2 × 20 or 4 × 10, and not 20 × 2/10 × 4. In the case of the formation of higher numbers by addition, the base precedes the (smaller) number which is being added to it. For example, so niʃ (10 + 2) twelve; soŋa (10 + 5) fifteen in Kinnauri, and ʧʊkɳi (10 + 2) twelve; ʧeŋga (10 + 5) fifteen in Navakat. In higher numbers formed by both multiplication and addition, the order becomes multiplier–base–addend, as expected. e.g.: 3 × 10 + 2=32. Further, all KST varieties use their ordinary numerals for forming higher numerals, although sometimes this is obscured by the result of phonological or morphophonological changes.

In all the varieties, except Poo, Kuno and Nako, no functional morpheme is added between the base (20 or 10) and the smaller numeral. In Poo and Nako more than one morpheme is found (see items thirty-one, forty-one, and seventy-one in Table 57).7 As seen above, in Kuno, there are additional morphemes for both multiplication (ʋa) and addition (naŋ).

On the whole, the numerals examined here are quite consistent with the observations made by Hodson (1913) for Sino-Tibetan languages. The forms of the numerals support the observations made above concerning the classification of KST varieties, where Sangla, Kalpa, Ropa, Chitkul and Labrang and Nichar form one group and Nako and Poo form another group—the former, for example, exhibiting a modified vigesimal system and the latter exhibiting a decimal system.

5.5 Basic Question Words

Table 58 shows the investigated basic question words and the automatically computed correspondences among varieties, and Table 44 contains the summary statistics extracted from Table 58.

In all cases, Poo, Kuno, and Nako exhibit the same cognate for the question words (even if the automated comparison does not always show this; see Section 4). On the whole, it seems that the forms in Sangla, Nichar, Kalpa, and Ropa are also etymologically related. As earlier, Chitkul and Labrang stand apart, sometimes siding with the Nako group (who in Chitkul), sometimes with the Sangla group (how in both Chitkul and Labrang), and sometimes exhibiting unique forms (who in Labrang; where in Chitkul).

Table 44

Summary statistics for basic question words

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

3/5 (60 %)

3/5 (60 %)

3/5 (60 %)

0/5 (0 %)

0/4 (0 %)

0/5 (0 %)

0/4 (0 %)

0/5 (0 %)

Ni

3/5 (60 %)

3/5 (60 %)

0/5 (0 %)

0/4 (0 %)

0/5 (0 %)

0/4 (0 %)

0/5 (0 %)

Ka

4/5 (80 %)

1/5 (20 %)

0/4 (0 %)

0/5 (0 %)

0/4 (0 %)

0/5 (0 %)

Ro

1/5 (20 %)

0/4 (0 %)

0/5 (0 %)

0/4 (0 %)

0/5 (0 %)

Ch

0/4 (0 %)

1/5 (20 %)

1/4 (25 %)

1/5 (20 %)

La

0/4 (0 %)

0/4 (0 %)

0/4 (0 %)

Po

4/4 (100 %)

4/5 (80 %)

Ku

4/4 (100 %)

5.6 Personal Pronouns

Table 59 shows the investigated personal pronouns and the automatically computed correspondences among varieties, and Table 45 contains the summary statistics extracted from Table 5A.8.

All the KST varieties examined here share some similarities with regard to their pronominal systems: First, in the second person the honorific–non-honorific distinction is made in all varieties (e.g., 2sg.nh, ki 2sg.h in Sangla; kʰóŋ 2sg.h, kʰøt 2sg.nh in Nako). Further, the plural pronominal forms are made by suffixing a plural marker to the corresponding singular pronominal form.

Apart from this, with regard to the pronominal forms, varieties fall into two groups: Sangla, Nichar, Kalpa, Ropa, Chitkul, and Labrang form one group and Poo, Kuno and Nako constitute another group. The two groups differ from each other consistently in this regard. Generally speaking, there is more homogeneity within the first group than within the second group regarding the pronominal forms. The same base forms for 2sg.h (ki) and 2sg.nh () occur in the Sangla, Kalpa, Nichar, Ropa, Chitkul and Labrang varieties. Poo, Kuno and Nako have the same base form for the 2sg non-honorific: kʰøtEQ031A, but they have three distinct forms for the 2sg honorific pronoun: ɲetEQ031A in Poo; rue in Kuno; and kʰóŋ in Nako. In all KST varieties, the 2pl is formed by affixing a plural marker to the 2sg pronoun (for example, ki 2sg.h and ki-nɔ 2sg.h-pl in Sangla). This is the case in both the second person honorific as well as non-honorific forms in all varieties. These varieties, however, do not use the same plural markers. If we concentrate our attention on the Sangla group it is, -nɔ in Sangla (e.g., kinɔ 2pl.h), but it is in Kalpa, -ʧaŋ in Chitkul (e.g. kaʧaŋ 2pl.nh) and -paŋ in Labrang (e.g. kɪnpaŋ).

Table 45

Summary statistics for personal pronouns

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

6/8 (75 %)

6/8 (75 %)

4/8 (50 %)

4/7 (57 %)

3/8 (37 %)

0/7 (0 %)

0/6 (0 %)

0/9 (0 %)

Ni

5/7 (71 %)

5/7 (71 %)

5/7 (71 %)

3/7 (42 %)

0/7 (0 %)

0/6 (0 %)

0/8 (0 %)

Ka

4/7 (57 %)

4/7 (57 %)

3/8 (37 %)

0/6 (0 %)

0/6 (0 %)

0/8 (0 %)

Ro

5/6 (83 %)

3/7 (42 %)

0/6 (0 %)

0/5 (0 %)

0/8 (0 %)

Ch

3/7 (42 %)

0/6 (0 %)

0/6 (0 %)

0/7 (0 %)

La

0/6 (0 %)

0/6 (0 %)

0/8 (0 %)

Po

3/6 (50 %)

3/7 (42 %)

Ku

4/6 (66 %)

The 3sg and 3pl forms in KST varieties, too, classify these varieties in two groups: Poo, Kuno, and Nako form one group. They have the same base form for 3sg. It is kʰɔ. This is distinct from the forms (e.g., ) found in the Sangla group (including Chitkul and Labrang).8 With the exception of Ropa which in our material has ʊno as the 3sg pronoun, all other varieties of this group (including Chitkul and Labrang) have forms which are also found in Sangla.9 The formation of the plural form in the third person is the same as that of the second person in these varieties—the plural marker is suffixed to the pronoun. But it seems that the plural markers are not necessarily the same in second and third person pronouns. Compare ki-nɔ 2sg.h-pl, but dɔ-gɔ 3sg-pl in Sangla, ki-ʃi 2sg.h-pl and nʊ-go 3sg-pl in Kalpa. This seems to be the case in all varieties, except Nako and Poo, where the same plural markers occur in all persons. See Chapters 2 and 3 for more detailed information on plural formation in Kinnauri and Navakat.

To summarize, the pronominal systems (including the pronominal forms) in these varieties classify Sangla, Nichar, Ropa, Kalpa, Chitkul and Labrang varieties in one group and Poo, Kuno and Nako as a separate group. The two groups differ from each other in all cases concerning their pronominal forms. The only similarities between these two groups are structural: Both groups make a honorific–non-honorific distinction in the second person, and the plural pronouns are formed in both groups by suffixing the plural marker to the corresponding singular pronouns.

5.7 Basic Vocabulary: Summary and Discussion

In Table 46 the combined statistics from a comparison of all nouns is presented, i.e., the figures from Tables 37 (kinship terms), 38 (body part terms) and 39 (other basic nouns) are combined into one in Table 46.

Since the individual comparisons of the noun subsets painted a unanimous picture of the classification of the KST varieties, it should come as no surprise that the combined noun statistics provides evidence for the same groupings.

Table 46

Summary statistics for all nouns

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

67/87 (77 %)

72/88 (81 %)

58/88 (65 %)

47/88 (53 %)

31/88 (35 %)

11/88 (12 %)

14/87 (16 %)

13/87 (14 %)

Ni

61/87 (70 %)

48/87 (55 %)

38/87 (43 %)

26/87 (29 %)

10/87 (11 %)

12/86 (13 %)

12/86 (13 %)

Ka

57/88 (64 %)

44/88 (50 %)

29/88 (32 %)

10/88 (11 %)

13/87 (14 %)

12/87 (13 %)

Ro

38/88 (43 %)

37/88 (42 %)

17/88 (19 %)

19/87 (21 %)

17/87 (19 %)

Ch

28/88 (31 %)

11/88 (12 %)

13/87 (14 %)

12/87 (13 %)

La

22/88 (25 %)

25/87 (28 %)

22/87 (25 %)

Po

56/87 (64 %)

62/87 (71 %)

Ku

57/87 (65 %)

Table 47 summarizes the comparison statistics for the whole lexical questionnaire. As can be seen from the denominators in the fractions, there is no single pair of varieties where all the 157 questionnaire concepts have been recorded in both members of the pair. However, they share from 149 (e.g., Kuno–Nako) to 155 recorded concepts (e.g., Sangla–Nako).

Again, the same picture as before emerges (see Figure 16):

  • Sangla, Nichar, and Kalpa form a clear grouping,

  • with Ropa closely associated.

  • Poo, Kuno, and Nako form another grouping, possibly somewhat less close than the Sangla group.

  • Finally, Chitkul and Labrang show greater affinity to the Sangla group than to the Nako group, but are distant from both. At the same time, Chitkul and Labrang are equally—or in some instances more—distant from each other as they are individually from the Sangla group.

Since Swadesh lists are often used in this kind of lexicostatistical investigation, summary statistics for all Swadesh list items in the questionnaire (88 concepts) are shown in Table 48, and in Table 49 we show corresponding statistics for the 25 concepts used in the questionnaire from the 40-item globally most stable Swadesh subset defined by Holman et al. (2008). If anything, the Swadesh list comparison ties Ropa closer to the Sangla group. Otherwise, nothing substantial changes.

Table 47

Summary statistics for the full lexical questionnaire

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

111/154 (72 %)

116/155 (74 %)

96/155 (61 %)

66/152 (43 %)

45/153 (29 %)

13/153 (8 %)

16/149 (10 %)

16/155 (10 %)

Ni

103/153 (67 %)

87/153 (56 %)

56/151 (37 %)

39/151 (25 %)

11/152 (7 %)

14/148 (9 %)

14/153 (9 %)

Ka

100/154 (64 %)

65/152 (42 %)

42/153 (27 %)

11/152 (7 %)

15/149 (10 %)

14/154 (9 %)

Ro

60/151 (39 %)

52/152 (34 %)

19/152 (12 %)

21/148 (14 %)

20/154 (12 %)

Ch

45/150 (30 %)

14/151 (9 %)

16/148 (10 %)

16/151 (10 %)

La

26/150 (17 %)

31/149 (20 %)

27/152 (17 %)

Po

97/149 (65 %)

105/152 (69 %)

Ku

97/149 (65 %)

d155303993e219046

Figure 16

Preliminary grouping of the nine investigated KST varieties (branch lengths are not significant)

5.8 Reflections on the Methodology

In this chapter, we have made a systematic comparison of nine KST varieties in order to throw some light on the genealogical classification of these underdescribed linguistic systems. The comparison has focused on the lexicon, which was investigated using an automatic, computational and purely quantitative method inspired by recent work on lexicostatistics and dialectometry, combined with traditional linguistic analysis and reasoning.

Table 48

Summary statistics for all Swadesh list items

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

65/87 (74 %)

64/87 (73 %)

59/87 (67 %)

33/86 (38 %)

28/86 (32 %)

9/86 (10 %)

10/83 (12 %)

11/87 (12 %)

Ni

58/86 (67 %)

53/86 (61 %)

28/86 (32 %)

25/85 (29 %)

7/86 (8 %)

9/83 (10 %)

9/86 (10 %)

Ka

58/86 (67 %)

31/86 (36 %)

25/86 (29 %)

7/85 (8 %)

9/83 (10 %)

9/86 (10 %)

Ro

31/85 (36 %)

33/85 (38 %)

12/85 (14 %)

13/82 (15 %)

14/86 (16 %)

Ch

31/85 (36 %)

10/85 (11 %)

10/83 (12 %)

11/85 (12 %)

La

15/84 (17 %)

17/83 (20 %)

16/85 (18 %)

Po

57/83 (68 %)

62/85 (72 %)

Ku

58/83 (69 %)

In Figure 16 we show the subgrouping of these nine KST varieties resulting from applying the method to our lexical data.

As has been pointed out a number of times above, the automatic comparison of lexical items often failed to pick out lexical item identities among varieties which were glaringly obvious to the linguist. At this point we should remind ourselves that this kind of computer program is simply a tool among many others in the linguist’s toolbox. Correctly used, it can be very helpful and save a lot of effort. In the present investigation it has turned out to be quite helpful to have an automated way of quickly calculating similarities among the language varieties under scrutiny, not least as a “generator” of new research questions.

It has helped to provide some answers and in the process proved its worth. Given that one accepts lexicostatistics using Swadesh-style core vocabulary lists as producing valid results, the refinement of this method that we have presented here seems to be a step in the direction of making this methodology more useful for teasing out the relationships among closely related language varieties.

Table 49

Summary statistics for the 25 most stable Swadesh items

Ni

Ka

Ro

Ch

La

Po

Ku

Na

Sa

20/25 (80 %)

22/25 (88 %)

20/25 (80 %)

11/25 (44 %)

11/25 (44 %)

3/25 (12 %)

3/25 (12 %)

3/25 (12 %)

Ni

19/25 (76 %)

19/25 (76 %)

11/25 (44 %)

11/25 (44 %)

3/25 (12 %)

3/25 (12 %)

3/25 (12 %)

Ka

18/25 (72 %)

9/25 (36 %)

10/25 (40 %)

3/25 (12 %)

3/25 (12 %)

3/25 (12 %)

Ro

10/25 (40 %)

14/25 (56 %)

4/25 (16 %)

4/25 (16 %)

4/25 (16 %)

Ch

11/25 (44 %)

4/25 (16 %)

4/25 (16 %)

4/25 (16 %)

La

5/25 (20 %)

5/25 (20 %)

5/25 (20 %)

Po

19/25 (76 %)

21/25 (84 %)

Ku

19/25 (76 %)

6 Results: Grammatical Features

In this section some preliminary observations about grammatical phenomena in the investigated KST varieties will be made on the basis of the noun phrase and sentence items in the questionnaire, as well as some additional grammatical data on reflexive and possessive pronouns collected during the fieldwork.

6.1 Reflexive and Possessive Pronominal Forms

In this section we will examine the forms as well as the composition of the reflexive pronominals in the KST varieties. In all the KST varieties examined here, the reflexive pronouns inflect for number and person of their coreferential antecedents. This is illustrated below with data from Sangla and Nako.

Sa

maŋ-o gǝs aŋ-u sa-k

‘In the dream I killed myself.’

maŋ-o ka-s kan-u sa-n

‘In the dream you (nh) killed yourself.’

do-s an-u-i lo-kjo

‘He said to himself.’

do-goː-s ane-goː-n(u) taŋtaŋ

‘They looked at themselves.’

Na

mà=su mà-raŋ=la tá(e)

‘I observed myself.’

màʃak=su màʃak-raŋ=la táe

‘We observed ourselves.’

kʰóŋ=su kʰóŋ-raŋ=la táe-ʋãːk

‘You (sg) observed yourself.’ (indirect knowledge)

kʰó=su kʰráŋ=la táe-ʋãːk

‘He observed himself.’ (indirect knowledge)

kʰóʋat=su kʰóʋat-raŋ=la táe-ʋãːk

‘They observed themselves.’ (indirect knowledge)

kʰó kʰóŋ=la táe-ʋãːk

‘He observed you.’ (indirect knowledge)

The composition of the reflexive pronoun is, however, not the same in all KST varieties. In Sangla, Kalpa, Nichar, and Ropa the reflexive form is the same as the non-nominative personal pronominal forms in the first and second persons (for example, ‘my/me’, kan ‘your/to you’ in Sangla), to which the dative case marker is suffixed. In the third person the third person non-nominative anaphoric pronoun10 an, functions as the reflexive pronoun. This can be seen by comparing the examples of Sangla reflexives, provided above, with the examples of possessive pronouns in Sangla, provided below (see also Chapter 2):

Sa

laː

my shadow’

kin bapu

your father’

an gas-oː

his (own) clothes’

do-goː-n gas-oː

their (someone else’s) clothes’

The reflexive pronominal formation in Nichar, Kalpa, and Ropa is the same as described here for Sangla, and the forms , kan, kin, an for 1sg, 2sg.nh, 2sg.h, 3sg, respectively, are also the ones used in Nichar, Kalpa, and Ropa.

Distinct from this, in the Poo, Kuno and Nako varieties, the base of the reflexive forms is the nominative form of the pronouns. The reflexive pronoun is formed in all three varieties by adding the suffix -raŋ to the nominative forms of the personal pronouns. See the Nako examples above and Chapter 3.

Labrang exhibits some similarity to the Sangla group in the reflexive pronouns, in that the non-nominative pronominal form functions also as the reflexive pronoun in the first person. It is in Labrang, as in the Sangla group. However, the second and third person reflexive pronoun raŋ—not similar to the other pronouns in Labrang—is shared with neither the Sangla group nor the Nako group, although it could be related to the reflexivizing suffix -raŋ of the Nako group and/or reflexive raŋ ‘self’ of Modern Tibetan.

Chitkul is distinct from all the other varieties in its reflexive pronouns. In Chitkul the first person reflexive is the same as the nominative pronoun (). Like in Labrang, a special reflexive pronoun—e—is used in both second and third person, distinct from the non-reflexive second and third person pronouns in this variety.

To summarize this section, as in the case of the personal pronouns, also with regard to reflexive pronouns the KST varieties form two groups: Sangla, Nichar, Kalpa, and Ropa form one group and Poo, Kuno and Nako form the other group. The reflexive form in the Sangla group is the non-nominative forms of pronouns, but in the Nako group it is the nominative pronominal form which is the base form(s) for reflexives, to which a reflexive affix is suffixed. Labrang and Chitkul do not clearly belong to one or to the other group, but also do not form a third group together.

6.2 Adjective—Noun Order

The order of constituents within the noun phrase in most of the investigated varieties seems to be Adjective–Noun. The exception is Nako, where the normal constituent order is Noun–Adjective. This is illustrated here with examples from Kinnauri and Navakat (see also Chapters 2 and 3):

Kinnauri

Navakat

moʈʰes ʧʰaŋfat boy’

ʈúː ɖùmpofat boy’

uʃk kimold house’

kítaːp ʈápothick book’

ʃare ʦʰeʦʰaʦbeautiful girl’

ʧìʋa kítpuhappy child’

6.3 Some Preliminary Observations about the Grammatical Structure of KST Varieties

Based on the sentences provided in the questionnaire (see Appendix 5A to this chapter), some very preliminary observations on their grammatical structure are presented below. The groupings among the KST varieties which we observed above are less clear when we consider the linguistic features which we examine on the basis of these sentences, perhaps because the grammatical features that we investigate are more abstract and change more slowly than the lexicon. It is still worthy of note that many of these varieties show different and noncognate endings for the same grammatical feature.

Case markers in nouns: All KST varieties have ergative and dative markers, although different markers are used in different varieties. The ergative markers in all varieties has some form of -(ǝ)s or -ʧi or -su. At least -(ǝ)s and -su may be related. The dative markers are -la (sg)/-nu (pl) or -ra or -u.

Plural markers in nouns: All varieties seem to have -a as a nominal plural marker. The plural marker precedes the case marker. Personal pronouns have distinct plural markers (see above).

Constituent order: The order of sentence constituents in all varieties is SOV. As we saw above, the noun phrase constituent order is Adj–Noun except in Nako, where we find the reverse order.

Verbal morphology: It seems that future and past tense markers are suffixed to the verb. In the case of Kalpa (future), Nichar/Poo/Chitkul (past) tense markers are similar to those in Kinnauri. In some varieties an occurs as the 3.h marker on the verb, while Nako exhibits no person or number indexing.

7 KST Varieties and Their Classification

Gerard (1841) lists five Sino-Tibetan varieties spoken in Kinnaur (“Koonawur”): (1) “Milchan or common Koonawuree”; (2) “T,heburskud”; (3) the dialect spoken in “Lubrung” and “Kanum”; (4) the dialect spoken in “Leedung”; and (5) “B,hoteea or Tartar”. According to this account, while Milchan and B,hoteea and, possibly also, T,heburskud are distinct languages (“tongues”), the varieties spoken in Lubrung/Kanum and Leedung are “dialects” of Milchan.

Gerard (1842) provides a word list (containing approximately 1,190 entries11), 98 direct-elicited phrases and clauses, and short descriptive notes on nouns and verbs in three KST varieties: Milchan, T,heburskud and B,hoteea/Tartar. The word list contains primarily nouns, adjectives, numerals and infinitive forms of verbs. There is also a word list of “Shoomchoo” (246 entries).

Cunningham (1844) adds Kinnauri Pahari (speech of the “Kohlis or Chumars” to use Cunningham’s terminology) to the list of “tongues”/ “dialects” mentioned by Gerard (1842), and provides a short comparative word list of “Milcháng or common”, “Tibberkad”, “Chamangee” (Kinnauri Pahari) and “Bhotee of Pitti, Hangrang, Rungchung, &c”. In total there are 110 entries, most for Milchan and Bhotee and relatively fewer for the other two (Cunningham 1844: 225–228).

Bailey (1909: 661–662) classifies Kinnauri into four dialects: (i) “Kanauri proper”, (ii) “Lower Kanauri”, (iii) “Thĕbör skad‘ ” and (iv) the variety spoken in Rakcham and Chitkul. The only difference between Kanauri proper and Lower Kanauri, according to Bailey, is in the lexicon—where Lower Kanauri has borrowed many lexical items from the neighboring Indo-Aryan languages. He regards the variety spoken in Chitkul and Rakcham as a distinct dialect of Kanauri, and classifies the KST varieties of Upper Kinnaur as Tibetan (Bailey 1909: 662). This information is also provided in later work by Bailey (1910), and is also included in the 1981 Indian Census Handbook (p. 9).

More recent accounts of the linguistic situation in Kinnaur extend these older accounts and recognize approximately eight languages indigenous to the region. Common to these accounts—e.g., Chamberlain et al. (1998), Huber (2007), and Saxena (2011)—is that they essentially rely on the Ethnologue (Eberhard et al. 2021 and earlier editions) for this assessment.12

The seven Sino-Tibetan languages recognized by the Ethnologue and also other sources (e.g., Glottolog; Hammarström et al. 2020) as spoken in Kinnaur are described in Table 50. Genealogically, these languages are generally classified under two different subbranches of Sino-Tibetan, with Bhoti Kinnauri and Tukpa classified as Tibetic and the other five languages as West Himalayish.

Table 50

KST varieties according to the Ethnologue

Name (ISO 639-3 code)

Alternative names / village(s) (tahsil) where spoken in Kinnaur

Jangshung (jna)

Jangrami, Zangram, Zhang-Zhung, Jangiam, Thebor, Thebör Skadd, Thebarskad, Central Kinnauri / Jangi, Lippa, Asrang (Morang)

Kinnauri (kfk)

Kinnaura Yanuskad, Kanoreunu Skad, Kanorug Skadd, Lower Kinnauri, Kinori, Kinner, Kanauri, Kanawari, Kanawi, Kunawari, Kunawur, Tibas Skad, Kanorin Skad, Kanaury Anuskad, Koonawure, Malhesti, Milchanang, Milchan, Milchang / From Chaura to Sangla and north along Satluj River to Morang, upper Ropa valley villages.

Kinnauri, Bhoti (nes)

Nyamskad, Mnyam, Myamskad, Myamkat, Nyamkat, Bud-Kat, Bod-Skad, Sangyas, Sangs-Rgyas, Bhotea of Upper Kinnauri / Nisang [Nesang] and possibly also Kuno and Charang (Morang); Poo (Poo)

Kinnauri, Chitkuli (cik)

Chitkuli, Chitkhuli, Tsíhuli, Tsitkhuli, Kinnauri, Kanauri, Thebarskad / Rakcham, Chitkul (Sangla)

Shumcho (scu)

Sumchu, Sumtsu, Shumcu, Thebor, Thebör Skadd, Thebarskad, Central Kinnauri, Sumcho / Kanam, Labrang, Spilo, Shyaso, Taling, Rushkaling (Poo)

Sunam (ssk)

Sungam, Sungnam, Thebor, Thebör Skadd, Thebarshad, Central Kinnauri, Sangnaur / Sunam (Poo)

Tukpa (tpq)

Nesang / Nesang, Charang, Kunnu [Kuno] (Morang)

The Ethnologue places all seven languages under the subbranch Kinauri (earlier Kanauri), which in other respects corresponds to West Himalayish or Tibeto-Kanauri in more accepted classifications among experts on Sino-Tibetan languages (e.g., Bradley 1997, 2002; LaPolla 2006, 2017a; Thurgood 2017), which in their turn largely coincide with Benedict (1972). The placement of the Tibeto-Kanauri (or [West] Himalayish) subbranch among the Sino-Tibetan languages varies somewhat, on the other hand. In the most common classification, (West) Himalayish forms a sister branch of Bodic under Bodish (Benedict 1972; Bradley 1997, 2002; Hyslop 2014), whereas LaPolla (2006, 2017a) and Thurgood (1984, 1985) place West Himalayish and Tibetic further apart in the family tree, under different primary branches of Sino-Tibetan (see Figure 17).13

d155303993e221019

Figure 17

Placement of the West Himalayish (WH) and Tibetic subbranches among the Sino-Tibetan languages according to the most common view (left) and according to LaPolla (2006, 2017a) (right)

Based on our results, we could then classify Sangla, Nichar, Kalpa, and possibly Ropa as the language (Lower) Kinnauri (kfk), Chitkul as Chhitkuli Kinnauri (cik), and Labrang as Shumcho (scu). Overall, the lexical comparison made here shows Poo and Nako to be slightly closer to each other than either is to Kuno, but the differences are small and with some vocabulary subsets actually go the other way (e.g., Tables 42, 44, and 45). If we are to speak of languages rather than a dialect continuum, these results indicate that we should recognize three languages or one language, but not two. The Nako group is consistently different from the Sangla group by a large margin in all cases, and thus the results shown here suggest a classification of these three varieties—Poo, Kuno, and Nako—as Tibetic (rather than West Himalayish) languages or varieties, namely as Bhoti Kinnauri (nes), completely in agreement with the traditional view (see Figure 18).

d155303993e221037

Figure 18

Resulting lower-level classification of the investigated KST varieties (branch lengths are not significant)

The Nako group is certainly distant enough from the other varieties for this to be conceivable. Further, all three varieties of the Nako group exhibit the probative lexical features of Tibetic, namely the form of the personal pronouns for second person singular (Navakat kʰjǿt) and third person singular (kʰó) (see Table 59 in Appendix 5A), plus the numeral ‘seven’ (dùn, dỳn) (Thurgood 2017: 11). Further, the finite verb forms in Kinnauri and Navakat differ more or less along the lines discussed by DeLancey (2014), the former exhibiting an “archaic” inflectional system, conveying information about the argument structure of its clause, while in the latter we find a “creoloid” structure, which encodes only discourse-grounding information. In this sense, too, Navakat is a typical Tibetic language, and not a West Himalayish one (DeLancey 2014: 58 ff.).

Also, going back to the more detailed descriptions of Kinnauri and Navakat in Chapters 2 and 3, we note some striking differences between the respective linguistic systems (Table 51). In all these cases, as also mentioned in Chapter 3, Navakat is similar to Modern Tibetan, exemplified here by Lhasa Tibetan (Bell 1939; DeLancey 2017b).

Table 51

A comparison of Kinnauri and Navakat with (Lhasa) Tibetan

Lhasa Tibetan

Navakat

Kinnauri

Phonetics: Is there phonemic tone?

Yes

Yes

No

Case markers

erg = ins case marker?

Yes

No. -su [erg]; daŋ [ins]

Yes. It is -s

Is the dat marker la?

Yes

Yes

No. It is -u, -n(u), -pǝŋ

Is the poss marker (-)ki?

Yes

Yes

No. It is -n(u)

Are the loc markers ru, na?

No. loc = dat (C-la, V-r)

Yes

No. It is -o, -no, -r

Is the case marking system consistently ergative?

No

Insufficient data

Yes

Honorificity

Are there distinct honorific and non-honorific verb stems?

Yes, for a set of verbs

Yes, for a set of verbs

No

Is honorificity marked on the verb with an inflectional ending?

Yes

No, exception: some verbal categories (e.g. imperative) distinguish h/nh

Yes

Are there distinct honorific and non-honorific second person pronouns?

Yes.

Yes.

Yes.

khyedrang [2sg.h];

khyedranggnyis [2du.h];

khyedrangʦho [2pl.h];

khyodrang [2sg.nh];

khyodranggnyis [2du.nh];

khyodrangʦho [2pl.nh]

kʰóŋ [2sg.h];

kʰóŋʃak, kʰóŋʤak [2pl.h];

kʰjǿt [2sg.nh];

kʰjǿtʋat [2pl.nh]

ki [2sg.h];

kiʃi [2du.h];

kino [2pl.h];

ka [2sg.nh];

kaniʃ [2du.nh];

kano, kanego [2pl.nh]

Pronouns

Is there an inclexcl distinction in the first person pronoun?

Yes. ŋā=ʦho (excl); ŋa=rang=ʦho (incl)

Yes. màʃak, ɲèt (excl); òn (incl)

Yes. niŋɔ (excl); kiʃa (incl)

Are there distinct nominative and non-nominative pronominal forms?

No

No

Yes. It has distinct forms for 1sg & 3.ana pronouns: [1sg.nom]; [1sg.nnom]; do, no [3sg.nom]; an [3sg.ana]

How are reflexive pronouns formed?

One reflexive pronoun for all persons: raŋ ‘self’

Personal pronoun + -raŋ

The non-nominative pronoun

Constituent ordering: Adj, N

N–Adj

N–Adj

Adj–N

Verb inflection

Are there different verb stems to mark tense/aspect and/or imperative?

Yes

Yes, in some cases

Only in one case: the verb ‘come’ has a distinct imperative verb form (ʤi)

Are tense and aspect two distinct inflectional grammatical categories?

No

No. There are fusional grammatical morphemes signalling tense and evidentiality.

Yes

Is there a subject indexing marker?

No. There is an egophoric system combined with evidentiality

No. There is an egophoric system combined with evidentiality

Yes

Negation: Is the negative marker sensitive to tense/ aspect?

Yes. ma- (pfv, fut) and mi- (ipfv)

Yes. ma- (pst) and mi- (npst)

No. ma- occurs in all tenses

How are imperatives formed?

The basic imperative is equivalent to the present or perfect verb root, sometimes with vowel changes (e.g. a > o). To this can be added various endings reflecting degree of honorificity, e.g. -ronaŋ, -roʧe (h) and -ʃi (nh).

A small set of verbs have distinct h/nh forms, including the imperatives in this verb set.

Apart from this, the h.imp form is formed by adding the suffix -rɔʧì to the verb stem. The nh.imp forming strategies: (i) bare verb form; (ii) a change in the stem vowel (a or e > o); (iii) -i or -e is suffixed to the verb; (iv) lengthening of the stem vowel

Only in one case: the verb ‘come’ has a distinct imperative verb form (ʤi). In all other cases, one of the following suffixes is added to the verb:

-riɲ : -iɲ/-ɲ : -iʧ /-ʧ : -ra : -o : -u : Ø

How are prohibitives formed?

ma- is prefixed to the imperative form

nhon: ma- is prefixed to the bare verb stem.

hon: V-ro mapèt

tʰa- is prefixed to the imperative verb form

In conclusion, here we have seen that the two KST varieties examined in this monograph—Kinnauri and Navakat, differ from each other at the phonological, lexical as well as at the grammatical level. In almost all the cases where the two languages differ, Navakat shows affinity with Tibetan, confirming the conclusions of the vocabulary comparison described in Section 5 above.

Appendix 5A: Questionnaire Items and Vocabulary Comparison Tables

5A.1 Questionnaire Items14

5A.1.1 Lexical Items

In the following list, all 237 questionnaire concepts are listed, and the 157 items used for the lexicostatistical investigation reported on in Section 5 of this chapter are shown in italics. For the latter set, Swadesh list items (88 concepts) are marked by their Swadesh list number, and Swadesh items in the set of 40 globally most stable items identified by Holman et al. (2008) are marked by an asterisk after the number (25 concepts).

I/1*

you (sg h)/2

you (sg -h)/2*

(s)he/3

we (incl)/4*

we (excl)/4

you (pl h)/5

you (pl -h)/5

they/6

this

that

here

there

who/11

what/12

where/13

when/14

how/15

not

all

many

some

girl

boy

maternal grandfather

maternal grandmother

paternal grandfather

paternal grandmother

woman/36

man (adult male)/37

man (human being)/38

child/39

daughter

son

wife/40

husband/41

mother/42

father/43

older sister

younger sister

older brother

younger brother

maternal aunt

paternal aunt

maternal uncle

paternal uncle

yak

yak (female)

animal/44

goat

bird/46

dog (f, m)/47*

cat (f; m)

sheep

snake/49

lamb

tree/51*

forest/52

hen

fruit/54

seed/55

leaf/56*

root/57

bark

beautiful a.

grass/60

rope/61

cat (m, f)

meat/63

blood/64*

bone/65*

milk

egg/67

food

tail/69

sugar

face

hair (head)/71*

head/72

ear/73*

eye/74*

nose/75*

mouth/76

tooth/77*

foot/80

leg

hand/83*

butter

glacier

village

breast

heart

drink v.

eat v.

bite v.

suck v.

laugh v.

see v.

hear v.

know v.

think v.

smell v.

fear v.

sleep v.

live v.

die v.

kill v.

fight v.

hunt v.

hit v.

cut v.

dig v.

swim v.

fly v.

walk v.

come v.

lie v.

sit v.

stand v.

fall v.

give v.

hold v.

wash v.

wipe v.

pull v.

push v.

throw v.

tie v.

say v.

sing v.

play v.

flow v.

gold

silver

copper

sun/147*

moon/148

star/149*

water/150*

rain/151

river/152

pond; lake

iron

salt/155

stone/156*

summer

winter

earth

cloud/160

autumn

sky/162

wind/163

snow/164

ice

spring (season)

fire/167*

mountain/171*

red a./172

green a./173

yellow a./174

white a./175

black a./176

night/177*

day/178

year/179

warm a./180

cold a./181

small a./32

big a./27

long a./28

new a./183*

old a./184

good a./185

bad a./186

straight a./189

round a./190

wet a./194

dry a./195

near a.

far a.

right a.

left a.

one/22*

two/23*

three/24*

four/25

five/26

six

seven

eight

nine

ten

eleven

twelve

thirteen

fourteen

fifteen

twenty

twenty-one

twenty-two

twenty-three

twenty-four

twenty-five

twenty-six

thirty

thirty-one

thirty-two

thirty-three

forty

forty-one

fifty

sixty

sixty-one

sixty-two

seventy

seventy one

eighty

eighty-one

ninety

one hundred

one hundred one

five hundred

one thousand

one thousand one

today

yesterday

1 day before y.-day

2 days before y.-day

3 days before y.-day

4 days before y.-day

tomorrow

1 day after tomorrow

2 days after tomorrow

3 days after tomorrow

4 days after tomorrow

carpenter

singer

5A.1.2 Noun Phrases

‘green grass’

‘fresh food’

‘water spring’

‘dry grass’

‘black hair’

‘barren land’

‘cold milk’

‘mountain top’

‘hot summer’

5A.1.3 Sentences

‘Santosh cooked food’

‘Ram saw a/the small boy today’

‘The children played and got tired’

‘Ram saw a/the small girl today’

‘Ram saw (the) small children today’

‘Ram saw a/the small house today’

5A.2 Vocabulary Comparison Tables

The vocabulary comparison tables are provided in full on the following pages.

In the tables we use the following notational conventions. Abbreviations (italicized in the tables) are used for the village names: Sangla (Sa), Nichar (Ni), Kalpa (Ka), Ropa (Ro), Chitkul (Ch), Labrang (La), Poo (Po), Kuno (Ku), Nako (Na). Vocabulary items refer to concepts and are identified by English words (or phrases on a few occasions) in small caps. Swadesh list items are further identified by their Swadesh list number added to the end of the English word and separated from the word by a slash: laugh/100. Items without this number do not appear in the Swadesh list. There are 88 Swadesh list concepts in the questionnaire (see above). If a Swadesh list item is marked with an asterisk, this means that the item is in the subset of 40 Swadesh list items found to be the most stable globally by Holman et al. (2008). There are altogether 25 out of these 40 items in the questionnaire (see Section 5.7).

The longer noun and adjective tables are arranged with the English concept glosses in alphabetical order. The other tables are arranged according to other principles (semantically or by Swadesh number). In the correspondence tables, numerical indices in square brackets appear in each cell to identify the language varieties which share a form for this concept, i.e. items considered the same according to the formal principles presented above in Section 4.2. Multiple indices in the same cell are separated by slashes.

Note that since the investigations described in this chapter were conducted before undertaking the more detailed phonological analysis underlying the phonemic orthography used in Chapter 2, the transcription system used for (Sangla) Kinnauri in Tables 52–59 below differs somewhat from that used in Chapter 2. However, in the interest of verifiability and reproducibility of results, I have elected to retain the earlier, less phonemic transcription here.

Table 52

Automatic comparison of kinship terms

Sa [1]

Ni [2]

Ka [3]

Ro [4]

Ch [5]

La [6]

Po [7]

Ku [8]

Na [9]

m.grand-father

[1/2/3/4/5] tete

[1/2/3/4/5] (maperɔŋ) tete

[1/2/3/4/5] (mapɔ) tete

[1/2/3/4/5] (maːpo) tete

[1/2/3/4/5] (matʃa) tete

[6/7/8/9] meme

[6/7/8/9] meme

[6/7/8/9] meme

[6/7/8/9] mème

m.grand-mother

[1/3/4/5/6] api

[2]

(maperɔŋ) ai

[1/3/4/5/6] (mapɔ) api

[1/3/4/5/6] (maːpo) api

[1/3/4/5/6] (matʃa) api

[1/3/4/5/6] api

[7/8/9]

aʋi

[7/8/9]

aʋi

[7/8/9]

áʋi

p.grand-father

[1/2/3/4/5] tete

[1/2/3/4/5] tete

[1/2/3/4/5] tete

[1/2/3/4/5] tete

[1/2/3/4/5] tete

[6/7/8/9] meme

[6/7/8/9] meme

[6/7/8/9] meme

[6/7/8/9] mème

p.grand-mother

[1/3/4/5/6] api

[2]

ai

[1/3/4/5/6] api

[1/3/4/5/6] api

[1/3/4/5/6] api

[1/3/4/5/6] api

[7/8/9]

aʋi

[7/8/9]

aʋi

[7/8/9]

áʋi

wife/40

[1/2/4/6] gone;

tsʰɛsmi

[1/2/4/6] gone

[3]

goʋene

[1/2/4/6] gone

[5]

bore

[1/2/4/6] gone

[7/9]

nama

[8]

tʃɛnmo

[7/9]

náma

mother/42

[1/3/4/5/6/ 7/8/9]

ama, ǝma

[2]

[1/3/4/5/6/ 7/8/9]

ama

[1/3/4/5/6/ 7/8/9]

ama

[1/3/4/5/6/ 7/8/9]

ama

[1/3/4/5/6/ 7/8/9]

ama

[1/3/4/5/6/ 7/8/9]

ama

[1/3/4/5/6/ 7/8/9]

ama

[1/3/4/5/6/ 7/8/9]

áma

daughter

[1/2/4]

tʃimɛd

[1/2/3/4]

tʃimɛ(d)

[2/3/4]

tʃimɛt

[1/2/3/4] tʃimetEQ031A

[5]

ɖju atʃi

[6]

tsamɛd

[7/8/9]

pomo

[7/8/9]

pomo

[7/8/9]

pòmo

older

sister

[1] dɔuts

[2] dai

[3] dao

[4] atʃʰe

[5] atʃa

[6] apu

[7/8/9] aʒi

[7/8/9] aʒi

[7/8/9] áʒi

younger sister

[1/2/3]

bǝɪts; ɖekʰraːts

[1/2/3]

baits

[1/2/3]

baɪts

[4/5]

baja

[4/5]

baja

[6]

bete

[7/8/9]

nomo

[7/8/9]

nomo

[7/8/9]

nòmo (tʃʉn); nòmo (tʃun)

husband/41

[1/2/3/4/5] tʃʰɔŋmi; dats

[1/2/3/4/5] dats

[1/2/3/4/5] dats

[1/2/3/4/5] datEQ031A

[1/2/3/4/5] dats

[6]

pruŋ

[7]

ɖuŋmi

[8]

dakpo

[9]

mákpa

father/42

[1/3]

boʋa; bapu

[2]

baba

[1/3]

bɔba

[4/6/7/8/9] apa

[5]

au

[4/6/7/8/9] apa

[4/6/7/8/9] apa

[4/6/7/8/9] apa

[4/6/7/8/9] áʋa

son

[1/2/3/4/6] tʃʰaŋ

[1/2/3/4/6] tʃʰaŋ

[1/2/3/4/6] tʃʰaŋ

[1/2/3/4/6] tʃʰaŋ

[5] ɖe atʃi

[1/2/3/4/6] tʃʰaŋ

[7/8/9]

ʈuː

[7/8/9]

ʈuː

[7/8/9]

ʈúː

older

brother

[1/2/3/5]

ate

[1/2/3/5]

ate

[1/2/3/5]

ate

[4/6/7/8]

atʃo

[1/2/3/5]

(teɪ) ate

[4/6/7/8]

atʃo

[4/6/7/8]

atʃo

[4/6/7/8]

atʃo

[9]

áʒo

younger brother

[1/2/5]

(gaʈo) ate; bǝɪts

[1/2]

baits

[3/4]

baja; baɪa

[3/4]

baja

[1/5]

(atsu) ate

[6]

bete

[7]

nono

[8/9]

no

[8/9]

maternal uncle

[1/2/3/5/6] mɔma

[1/2]

mama

[1/3/5/6] moma

[4/7/8/9] adʒaŋ

[1/3/5/6] moma

[1/3/5/6] moma

[4/7/8/9]

aʒaŋ

[4/7/8/9]

aʒaŋ

[4/7/8/9]

áʒaŋ

maternal

aunt

[1]

ama; nane

[2]

autse

[3]

amni

[4/7]

ane

[5]

(matʃa) ene

[6]

tsema

(‘older m.a.’)

[4/7]

ane

[8]

matʃuŋ

[9]

mèʒoŋ

paternal uncle

[1/2/5]

(teg) bɔʋa; bapu

[1/2]

baba; babats

[3]

babu

[4]

tsipa

[1/5]

bapu

[6]

aku

[7]

apatʃu(n)

[8]

aʊtʃuŋ

[9]

éu

paternal

aunt

[1/2/3]

nane

[1/2/3]

nane; nai

[1/2/3]

nane

[4]

tsima

[5]

ene

[6/7/8/9]

ane

[6/7/8/9]

ane

[6/7/8/9]

ane

[6/7/8/9]

áne

Table 53

Automatic comparison of terms for body parts

Sa [1]

Ni [2]

Ka [3]

Ro [4]

Ch [5]

La [6]

Po [7]

Ku [8]

Na [9]

head/72

[1/2/3/4]

bal

[1/2/3/4]

bal

[1/2/3/4]

baːl

[1/2/3/4]

bal

[5]

pitʃaː

[6]

piʃa

[7/8/9]

ⁿgɔ

[7/8/9]

go

[7/8/9]

ⁿgɔ̀

face

[1/3/4]

to

[2]

sto

[1/3/4]

to

[1/3/4]

to

[5]

mukʰaŋ

[6]

mumi

[7]

ŋɔnɔŋ

[8]

donok

[9]

ŋòdo(ŋ)

hair (head)/71*

[1/2/3/4/5/6] kra

[1/2/3/4/5/6] kra

[1/2/3/4/5/6] kra

[1/2/3/4/5/6] kra

[1/2/3/4/5/6] kra

[1/2/3/4/5/6] kra

[7/8/9]

ʈa

[7/8/9]

ʈa

[7/8/9]

ʈá

tail/69

[1]

pǝtsnɪŋ

[2]

pantsiŋ

[3]

pǝtsǝnɪŋ

[4]

pikon

[5]

mɛts

[6]

mɛkɔn

[7/8/9]

ŋama

[7/8/9]

ŋama

[7/8/9]

ŋáma

ear/73*

[1/2/3/4] kǝnaŋ

[1/2/3/4] kanaŋ

[1/2/3/4] kanaŋ

[1/2/3/4] kanaŋ

[5]

rɔts

[6]

repaŋ

[7/9] namdʒɔkEQ031A

[8]

namtʃɔk

[7/9] námdʒɔkEQ031A

eye/74*

[1/2/4/5/6/ 7/8/9] mig

[1/2/4/5/6/ 7/8/9] mig

[3/4/5/6/ 7/8/9] mik

[1/2/3/4/5/6/ 7/8/9] mikEQ031A

[1/2/3/4/5/6/ 7/8/9] mi

[1/2/3/4/5/6/ 7/8/9] mi

[1/2/3/4/5/6/ 7/8/9] mikEQ031A

[1/2/3/4/5/6/ 7/8/9] mɪkEQ031A

[1/2/3/4/5/6/ 7/8/9] míkEQ031A

nose/75*

[1/3/4]

takuts

[2]

stakʊts

[1/3/4]

takuts

[1/3/4]

takuts

[5]

rim

[6]

mur

[7/9]

na

[8]

naʊ

[7/9]

mouth/76

[1/2/3/4] kʰǝkaŋ

[1/2/3/4] kʰakɔŋ

[1/2/3/4] kʰakaŋ

[1/2/3/4] kʰakaŋ

[5]

[6]

agor

[7/8/9]

kʰa

[7/8/9]

kʰa

[7/8/9]

kʰá

tooth/77*

[1/2/3/4]

gar

[1/2/3/4]

gar

[1/2/3/4]

gar

[1/2/3/4]

gar

[5]

sua

[6]

sʋa

[7/8/9]

so

[7/8/9]

so

[7/8/9]

hand/83*

[1/2/4/6]

gʊd

[1/2/4/6]

gʊd

[3/4/6]

gʊt

[1/2/3/4/6] gʊtEQ031A

[5]

lau

[1/2/3/4/6] gʊtEQ031A

[7/8/9]

lakpa

[7/8/9]

lakpa

[7/8/9]

làkpa

foot/80

[1/2/3/4]

baŋ

[1/2/3/4]

baŋ

[1/2/3/4]

baŋ

[1/2/3/4]

baŋ

[5]

boŋ

[6]

baŋkʰan

[7/8/9]

kaŋba

[7/8/9]

kaŋpa

[7/8/9]

káŋba

Table 54

Automatic comparison of other basic nouns

Sa [1]

Ni [2]

Ka [3]

Ro [4]

Ch [5]

La [6]

Po [7]

Ku [8]

Na [9]

animal/44

[1/2/3/4/8/9] sakɔ; semtʃen; sem

[1/2/3/4/8/9] sɛmtʃɛn

[1/2/3/4/8/9] sɛmtʃɛn

[1/2/3/4/6/8/9] tʃʰuma; sɛmtʃɛn

[5]

rat

[4/6]

tʃʰuma

[7]

sɪmtʃɪn

[1/2/3/4/8/9] sɛmtʃɛn

[1/2/3/4/8/9] sémtʃen

autumn

[1/2/3/4/5] ʈʃarmi

[1/2/3/4/5] ʈʃarmi

[1/2/3/4/5] tʃarmi

[1/2/3/4/5] tʃarmi

[1/2/3/4/5] tʃarmi

[6]

nuŋ

[7]

namle

[8]

nam

[9]

tóŋga

bird/46

[1/2/3/5]

pjats

[1/2/3/5]

pjats

[1/2/3/5]

pjats

[4]

pjad

[1/2/3/5]

pjats

[6]

pɪat

[7]

tʃiu

[8]

dʒa

[9]

tʃà

blood/64*

[1/2/3/5] polats

[1/2/3/5] polats

[1/2/3/5] pɔlats

[4/5/6]

polad; ʃʋi

[1/2/3/4/5] pola

[4/6]

ʃʋi

[7/8/9]

ʈʰak

[7/8/9]

ʈʰak

[7/8/9]

ʈʰákEQ031A

bone/65*

[1/2/3/5] hǝraŋ

[1/2/3/5] harɔŋ

[1/2/3/5] haraŋ

[4/6]

harko

[1/2/3/5] haraŋ

[4/6]

harko

[7]

rukok

[8]

rufa

[9]

rùːgokEQ031A

butter

[1/2/3/4/5/6/ 7/8/9] mar

[1/2/3/4/5/6/ 7/8/9] mar

[1/2/3/4/5/6/ 7/8/9] mar

[1/2/3/4/5/6/ 7/8/9] mar

[1/2/3/4/5/6/ 7/8/9] mar

[1/2/3/4/5/6/ 7/8/9] mar

[1/2/3/4/5/6/ 7/8/9] mar

[1/2/3/4/5/6/ 7/8/9] mar

[1/2/3/4/5/6/ 7/8/9] màr

cat (f; m)

[1/3/4/5/6/8] piʃi

[2] pǝʃ (mǝnʈr, skjo)

[1/3/4/5/6/8] piʃi

[1/3/4/5/6/8] piʃi

[1/3/4/5/6/8] piʃi

[1/3/4/5/6/8] piʃi

[7/9]

puʃi

[1/3/4/5/6/8] piʃi

[7/9]

púʃi

child/39

[1/2/3/4]

tʃʰaŋ

[1/2/3/4]

tʃʰaŋ

[1/2/3/4]

tʃʰaŋ

[1/2/3/4]

tʃʰaŋ

[5]

atʃi

[6]

tʃigdʒja

[7/8/9]

tʃiʋa

[7/8/9]

tʃiʋa

[7/8/9]

tʃìʋa

cloud/160

[1/2/3/4/5]

dʒu

[1/2/3/4/5]

dʒu

[1/2/3/4/5]

dʒu

[1/2/3/4/5] dʒuː

[1/2/3/4/5]

zu

[6]

mukpa

[7]

makpa

[8]

tin

[9]

ʈín

copper

[1/2/3/4] ʈromaŋ

[1/2/3/4] ʈromaŋ

[1/2/3/4] ʈromaŋ

[1/2/3/4] ʈromaŋ

[5]

ʈamaŋ

[6]

tromaŋ

[7/9]

saɖ

[8]

saŋma

[7/9]

sã́ː

day/178

[1/2/3/4]

dear; lae

[1/2/3/4]

lae

[1/2/3/4]

laje

[1/2/3/4]

laje

[5]

niri

[6]

nir

[7/8/9]

ɲɪnmo

[7/8/9]

ɲɪnmo; tiriŋ

[7/8/9]

ɲìnmo

dog/47*

[1/2/3/4]

kui

[1/2/3/4]

kui

[1/2/3/4]

kui

[1/2/3/4]

kui

[5]

kʰui

[6/7/8/9]

kʰi

[6/7/8/9]

kʰi

[6/7/8/9]

kʰi

[6/7/8/9]

kʰí

egg/67

[1]

litr

[2]

lito

[3/4/5]

liʈ

[3/4/5]

liʈ

[3/4/5]

liː

[6]

lili

[7]

guʋa

[8]

goŋa

[9]

gòã

fire/167*

[1/2/3/4/5/6/ 7/8/9] me

[1/2/3/4/5/6/ 7/8/9] me

[1/2/3/4/5/6/ 7/8/9] me

[1/2/3/4/5/6/ 7/8/9] me

[1/2/3/4/5/6/ 7/8/9] me

[1/2/3/4/5/6/ 7/8/9] me

[1/2/3/4/5/6/ 7/8/9] mɛ

[1/2/3/4/5/6/ 7/8/9] me

[1/2/3/4/5/6/ 7/8/9] mè

food

[1/2/3]

kʰɔʊ

[1/2/3]

kʰau

[1/2/3]

kʰau

[4]

tʰaktʰukEQ031A; tsas

[5]

kɔn

[6]

ʈʰaktur

[7]

takʰtukEQ031A

[8]

saptuŋ

[9]

sèptuŋ

forest/52

[1/2/3/4/5/6] zaŋgal

[1/2/3/4/5/6] zaŋgal

[1/2/3/4/5/6] dʒaŋgal

[1/2/3/4/5/6] dʒaŋgal

[1/2/3/4/5/6] zaŋgal

[1/2/3/4/5/6] dʒaŋgal

[7/9]

rija

[8]

riga

[7/9]

rìa

fruit/54

[1/2/3/5/6] fɔlaŋ

[1/2/3/5/6] fɔlaŋ

[1/2/3/5/6] fɔlaŋ

[4]

uʃo

[1/2/3/5/6] pʰolaŋ

[1/2/3/5/6] fɔlaŋ

[7]

ʃɪntʃukEQ031A

[-] -

[-] -

glacier

[1/2/3/5]

risur

[1/2/3/5]

risur

[1/2/3/5]

risur

[4/6]

lisur

[1/2/3/5]

risur

[4/6]

lisʊr

[7/8/9]

rut

[7/8/9]

rutEQ031A

[7/8/9]

rùːtEQ031A

goat

[1/2/3/4] bakʰɔr; adʒ (male)

[1/2/4]

bakʰɔr; bɔlu (male)

[1/3]

bakʰaraŋ (f); adʒ (male)

[1/2/4]

bakʰɔr

[5/6]

tet

[5/6]

tɛtEQ031A

[7/8/9]

rama

[7/8/9]

rama; raʋo

[7/8/9]

ràma

gold

[1/2/3/4/5/6] zaŋ

[1/2/3/4/5/6] zaŋ

[1/2/3/4/5/6] zaŋ

[1/2/3/4/5/6] zaŋ

[1/2/3/4/5/6] zaŋ

[1/2/3/4/5/6] zaŋ

[7]

sir

[8/9]

ser

[8/9]

sér

grass/60

[1/2/3/4/5]

tʃi

[1/2/3/4/5]

tʃi

[1/2/3/4/5]

tʃi

[1/2/3/4/5]

tʃiː

[1/2/3/4/5]

tʃiː

[6]

tsi

[7/8/9]

sa

[7/8/9]

sa

[7/8/9]

hen

[1/2/3/4/5/6] (maɳʈr) kukari

[1/2/3/4/5/6] (manʈr) kukari

[1/2/3/4/5/6] (manʈu) kukari

[1/2/3/4/5/6] kukari

[1/2/3/4/5/6] kukari

[1/2/3/4/5/6] kukari

[7/9]

tʃamo

[8]

kukuri

[7/9]

tʃàmo

iron

[1/2/3/4/5]

rɔn

[1/2/3/4/5]

rɔn

[1/2/3/4/5]

rɔn

[1/2/3/4/5]

rɔn

[1/2/3/4/5] r

ɔn

[6/9]

tʃakʰ

[7/8/9]

tʃak

[7/8/9]

tʃak

[6/7/8/9]

tʃáː; tʃák

lamb

[1/2/3/4]

kʰats

[1/2/3/4]

kʰats

[1/2/3/4]

kʰats

[1/2/3/4]

kʰats

[5]

krats

[6]

krat

[7/9]

lu

[8]

lugu

[7/9]

lùː

leaf/56*

[1/3]

patʰraŋ

[2]

patraŋ

[1/3]

patʰraŋ

[4/6]

patalaŋ

[5]

patʰǝraŋ

[4/6]

patəlaŋ

[7]

hɔk

[8]

hɔg

[9]

lìpEQ031A

man (human)/38

[1/2/3/4/5/6/ 7/8/9] mi

[1/2/3/4/5/6/ 7/8/9] mi

[1/2/3/4/5/6/ 7/8/9] mi

[1/2/3/4/5/6/ 7/8/9] mi

[1/2/3/4/5/6/ 7/8/9] mi

[1/2/3/4/5/6/ 7/8/9] mi

[1/2/3/4/5/6/ 7/8/9] mi

[1/2/3/4/5/6/ 7/8/9] mi

[1/2/3/4/5/6/ 7/8/9] mì

man (male)/37

[1]

ɖekʰres

[2]

ɖɛkʰros

[3]

dʒʋan

[4]

ɖɛkʰrad

[5]

boiŋ

[6/8]

mi

[7/8]

kʰjɔktɔŋ

[6/7/8] kʰjɔktɔŋ; mi

[9]

pʰúʒa

meat/63

[1/2/3/4/5/6/ 8/9] ʃa

[1/2/3/4/5/6/ 8/9] ʃa

[1/2/3/4/5/6/ 8/9] ʃa

[1/2/3/4/5/6/ 8/9] ʃa

[1/2/3/4/5/6/ 8/9] ʃa

[1/2/3/4/5/6/ 8/9] ʃa

[7]

ʃia

[1/2/3/4/5/6/ 8/9] ʃa

[1/2/3/4/5/6/ 8/9] ʃá

milk

[1]

kʰiraŋ

[2]

kʰiroŋ

[3/4/5]

kʰɛraŋ

[3/4/5]

kʰeraŋ

[3/4/5]

kʰeraŋ

[6]

kʰatipɛl

[7/8/9]

(h)oma

[7/8/9]

oma

[7/8/9]

òma

moon/148

[1/2/3/4/5/6] golsaŋ

[1/2/3/4/5/6] golsaŋ

[1/2/3/4/5/6] golsaŋ

[1/2/3/4/5/6] golsaŋ

[1/2/3/4/5/6] golsaŋ

[1/2/3/4/5/6] golsaŋ

[7/8]

daʋa

[7/8]

daʋa

[9]

ⁿdàːr

mountain/ 171*

[1/2/4]

ɖɔkʰaŋ; raŋ

[1/2]

ɖɔkʰaŋ

[3]

ɖɔkaŋ

[1/4]

raŋ

[5]

ʈʰol

[6]

ʋe

[7/8/9]

la

[7/8/9]

la

[7/8/9]

night/177*

[1/2/3/4]

tʊr; ratɪŋ

[1/2/3/4]

ratɪŋ

[1/2/3/4]

ratɪŋ

[1/2/3/4]

ratɪŋ

[5]

muni

[6]

gɔ̃

[7]

goŋmo

[8]

tsaŋmo

[9] gø̀emo; gòemo

pond; lake

[1/2/3/5]

sɔraŋ

[1/2/3/5]

sɔraŋ

[1/2/3/5]

sɔraŋ

[4]

soː

[1/2/3/5] soraŋ

[6/7/8/9]

tʃɔ

[6/7/8/9]

tʃo

[6/7/8/9]

tʃo

[6/7/8/9]

tʃó

rain/151

[1/2/4/6/7/8/9] gʋɛnɪŋ; tʃʰarʋa

[1/2]

gʋɛnɪŋ

[3]

lagɛts

[1/4/6/7/8/9] tʃʰarʋa

[5]

gojnɪŋ

[1/4/6/7/8/9] tʃʰarba

[1/4/6/7/8/9] tʃʰã́rʋa

[1/4/6/7/8/9] tʃʰarʋa

[1/4/6/7/8/9] tʃʰárʋa

river/152

[1/2/5]

garǝŋ

[1/2/5]

garaŋ

[3]

sɔmɔndraŋ

[4]

nalaŋ

[1/2/5]

garaŋ

[6]

luŋpʰa

[7]

tsaŋbo

[8/9]

tsaːnpʰo; tsaːnpʰoŋ

[8/9]

tsáːnfo

root/57

[1/3/4/5/6] dʒilaŋ

[2]

dziloŋ

[1/3/4/5/6] dʒilaŋ

[1/3/4/5/6] dʒilaŋ

[1/3/4/5/6] dʒilaŋ

[1/3/4/5/6] zilaŋ

[7]

batak

[8]

patak

[9]

pàdakEQ031A

rope/61

[1/2/3]

baʂ

[1/2/3]

bǝʃ

[1/2/3]

bǝʃ

[4/6/7/8/9] tʰakpa

[5]

lat

[4/6/7/8/9] tʰakpa

[4/6/7/8/9] tʰakpa

[4/6/7/8/9] tʰakpa

[4/6/7/8/9] tʰákpa

salt/155

[1/2/3/4/5/6/ 7/8/9] tʃa

[1/2/3/4/5/6/ 7/8/9] tʃa

[1/2/3/4/5/6/ 7/8/9] tʃa

[1/2/3/4/5/6/ 7/8/9] tʃa

[1/2/3/4/5/6/ 7/8/9] tʃa

[1/2/3/4/5/6/ 7/8/9] tʃa

[1/2/3/4/5/6/ 7/8/9] tʃa

[1/2/3/4/5/6/ 7/8/9] tʃa

[1/2/3/4/5/6/ 7/8/9] tʃá

seed/55

[1/3/4/5]

bijaŋ

[2]

bijoŋ

[1/3/4/5]

bijaŋ

[1/3/4/5]

bijaŋ

[1/3/4/5]

bijaŋ

[6]

pʊdzad

[7/8/9]

saŋon

[7/8/9]

saŋon

[7/8/9]

sáŋɔn; sáŋøn

sheep

[1/2/3]

zɛd

[1/2/3] (mal) zɛd

[1/2/3]

(mɔl) zɛd

[4]

kʰas

[5]

modzat

[6]

braŋ

[7/8/9]

mamo

[7/8/9]

mamo

[7/8/9]

màmo (‘ewe’)

silver

[1/2/3/4/5/6/ 7/8/9] mul

[1/2/3/4/5/6/ 7/8/9] mul

[1/2/3/4/5/6/ 7/8/9] mul

[1/2/3/4/5/6/ 7/8/9] mul

[1/2/3/4/5/6/ 7/8/9] mul

[1/2/3/4/5/6/ 7/8/9] mul

[1/2/3/4/5/6/ 7/8/9] mul

[1/2/3/4/5/6/ 7/8/9] mʊl; mʉl

[1/2/3/4/5/6/ 7/8/9] múl

sky/162

[1/2/3/4/5] sorgaŋ

[1/2/3/4/5] sorgaŋ

[1/2/3/4/5] sorgaŋ

[1/2/3/4/5] sorgaŋ

[1/2/3/4/5] sorgaŋ

[6/7/8/9]

nam

[6/7/8/9]

nam

[6/7/8/9]

nam

[6/7/8/9]

nám

snake/49

[1]

sapes

[2/3/6]

sapas

[2/3/6]

sapas

[4]

saʋəs

[5]

sapa

[2/3/6]

saʋas

[7/8/9]

ɖul

[7/8/9]

ɖʊl

[7/8/9]

ɖùl; ɖỳl

snow/164

[1/2/3/4]

pom

[1/2/3/4]

pom

[1/2/3/4]

pom

[1/2/3/4]

pom

[5]

haŋ

[6]

ras

[7/8/9]

kʰa

[7/8/9]

kʰa

[7/8/9]

kʰáː

spring (season)

[1/2/3/4/5] rɛnam

[1/2/3/4/5] rɛnam

[1/2/3/4/5] rɛnam

[1/2/3/4/5] rena(m)

[1/2/3/4/5] renam

[6]

gjanəm

[7]

tʃʰarko

[8]

tonka

[9]

píka

star/149*

[1/5]

tar; kar

[2/3]

skar

[2/3]

skar

[4/6/7/8/9] karma

[1/5]

kaːr

[4/6/7/8/9] karma

[4/6/7/8/9] karma

[4/6/7/8/9] karma

[4/6/7/8/9] kárma

stone/156*

[1/2/4/5/6]

rag

[1/2/4/5/6]

rag

[3]

runiŋ

[1/2/4/5/6] ra(g)

[1/2/4/5/6]

ra

[1/2/4/5/6] kɔlɔn; ra

[7/8/9]

dua

[7/8/9]

dua

[7/8/9]

dùa

sugar

[1/2/3/4/5/ 7/8/9] tʃini

[1/2/3/4/5/ 7/8/9] tʃini

[1/2/3/4/5/ 7/8/9] tʃini

[1/2/3/4/5/ 7/8/9] tʃini

[1/2/3/4/5/ 7/8/9] tʃiniː

[6]

sini

[1/2/3/4/5/ 7/8/9] tʃini

[1/2/3/4/5/ 7/8/9] tʃiniː; ŋarmo

[1/2/3/4/5/ 7/8/9] tʃíːni

summer

[1/2/3]

ʂɔl

[1/2/3]

ʃɔl

[1/2/3]

ʃɔl

[4]

ʃolo

[5]

sol

[6]

hɔlaŋ

[7/9]

jarka

[8]

ɛrka

[7/9]

járka

sun/147*

[1/3/4]

june

[2]

jun

[1/3/4]

june

[1/3/4]

junekEQ031A

[5/6]

ni

[5/6]

ni

[7/8/9]

ɲima

[7/8/9]

ɲima

[7/8/9]

ɲìma

tree/51*

[1/2/3/6] boʈʰaŋ

[1/2/3/6] boʈʰaŋ

[1/2/3/6] boʈʰaŋ

[4]

botaŋ

[5/7/8/9]

paːŋ

[1/2/3/6] boʈʰaŋ

[5/7/8/9]

paŋ

[5/7/8/9]

paŋ

[5/7/8/9]

páŋ

village

[1/2/3/4/5/6] deʃaŋ

[1/2/3/4/5/6] deʃaŋ

[1/2/3/4/5/6] deʃaŋ

[1/2/3/4/5/6] deʃaŋ

[1/2/3/4/5/6] dɛʃaŋ

[1/2/3/4/5/6] deʃaŋ

[7/8/9]

jul

[7/8/9]

jʊl; jʉl

[7/8/9]

jùl

water/150*

[1/2/3/4/5/ 6/9] ti

[1/2/3/4/5/ 6/9] ti

[1/2/3/4/5/ 6/9] ti

[1/2/3/4/5/ 6/9] ti

[1/2/3/4/5/ 6/9] ti

[1/2/3/4/5/ 6/9] ti

[7/8/9]

tʃʰu

[7/8/9]

tʃʰu

[1/2/3/4/5/6/ 7/8/9] tʃʰú; tí

wind/163

[1/2/3/4/5/6] laːn

[1/2/3/4/5/6] laːn

[1/2/3/4/5/6] lan

[1/2/3/4/5/6] laːn

[1/2/3/4/5/6] laːn

[1/2/3/4/5/6] laːn

[7]

lagda

[8]

lagpa

[9]

lágdɛ

winter

[1/2/3]

gun

[1/2/3]

gun

[1/2/3]

gun

[4]

guno

[5]

guni

[6]

gʊnaŋ

[7]

gunkʰa

[8]

gunka

[9]

gùnga

woman/36

[1/2]

tsʰɛsmi

[1/2]

tsʰɛsmi

[3]

tsʰɛtsɛs; tsʰetses

[4]

tsʰesemi

[5]

mɔrɪŋ

[6]

mʊnʃɪŋ

[7/9]

kʰimamo

[8]

pomo

[7/9]

kʰímamo

yak

[1/4/5/6/8/9] jak

[2/3/4/5/6/7] jag

[2/3/4/5/6/7] jag

[1/2/3/4/5/6/ 7/8/9] jakEQ031A

[1/2/3/4/5/6/ 7/8/9] ja

[1/2/3/4/5/6/ 7/8/9] jaː

[2/3/4/5/6/7] jag

[1/4/5/6/8/9] jak

[1/4/5/6/8/9] jàk

yak (female)

[1/3/5]

brime

[-] -

[1/3/5]

brime

[4/6]

brimo

[1/3/5]

brime

[4/6]

brimo

[7/8]

ɖimo

[7/8]

ɖimo

[9]

jakmo

year/179

[1/3/4/5/6] boʂaŋ

[2]

borʃaŋ

[1/3/4/5/6] boʃaŋ

[1/3/4/5/6] boʃaŋ

[1/3/4/5/6] boʃaŋ

[1/3/4/5/6] boʃaŋ

[7/8/9]

[7/8/9]

lo

[7/8/9]

lò; lɔ̀

Table 55

Automatic comparison of adjectives

Sa [1]

Ni [2]

Ka [3]

Ro [4]

Ch [5]

La [6]

Po [7]

Ku [8]

Na [9]

bad/186

[1/3]

mari

[2]

maːr

[1/3]

mari

[4/6]

halam

[5]

maʃəro

[4/6]

halam

[7]

akʰe

[8]

tʰʊa

[9]

ŋànba

beautiful

[1/2/3/4/5/6] ʃare

[1/2/3/4/5/6] ʃare

[1/2/3/4/5/6] ʃare

[1/2/3/4/5/6] ʃare

[1/2/3/4/5/6] ʃare

[1/2/3/4/5/6] ʃare

[7]

laho

[8]

lakpo

[9]

làːfo

big/27

[1/2/3/4]

teg; teːg

[1/2/3/4]

teg

[1/2/3/4]

teg

[1/2/3/4]

teg

[5]

tɛi

[6]

tʃei

[7/8]

tʃʰepo

[7/8]

tʃʰepo

[9]

tʃʰétpo

black/176

[1/2/3/4]

rok

[1/2/3/4]

rok

[1/2/3/4]

rɔk

[1/2/3/4]

rɔkEQ031A

[5/6]

kʰai

[5/6]

kʰai

[7/8/9]

nakpo

[7/8/9]

nakpo

[7/8/9]

nàkpo

cold/181

[1/2/3/4]

lis; tʃɪk; sɔk

[1/2/3]

tʃɪk

[1/2/3/4]

sɔk; tʃɪk

[1/3/4]

sɔtEQ031A

[5/6]

kʰati

[5/6]

kʰati

[7/8/9]

ʈaŋmo

[7/8/9]

ʈaŋmo

[7/8/9]

ʈàŋmo

dry/195

[1]

ʈʃarmu

[2/3]

ʈʃarts

[2/3]

ʈʃarts

[4]

tʃar

[5/6]

fɔsi

[5/6]

fɔsi

[7/8/9]

kampo

[7/8/9]

kambo

[7/8/9]

kámpo

good/185

[1/2/3/4]

dam

[1/2/3/4]

dam

[1/2/3/4]

dam

[1/2/3/4]

dam

[5]

dzoi

[6/8]

epo

[7/9]

gaŋʃɪn (people); ʃɪmbo (inan.); demo

[6/8]

epo

[7/9]

dèmo; zàŋbo; ʃímpo; ʃímbo; ètpo

green/173

[1/2]

raːg

[1/2]

raːg

[3/4]

rak

[3/4]

raːk

[5]

pʰi

[6]

tiŋ

[7/8/9]

ŋonpo;

ŋønpo

[7/8/9]

ŋonpo; ɖompo

[7/8/9]

ŋǿnpo

(blue-green)

long/28

[1]

lames

[2/3/4]

lamɔs

[2/3/4]

lamas

[2/3/4]

lamas

[5]

rui

[6]

ʃui; sarpa

[7/8/9]

riŋpo

[7/8/9]

riŋpo

[7/8/9]

rìŋpo

new/183*

[1/2/3/4]

ɲug; ɲuːg

[1/2/3/4]

ɲuːg

[1/2/3/4]

ɲukEQ031A

[1/2/3/4]

ɲukEQ031A

[5/6]

nui

[5/6]

nui

[7/8/9]

soma

[7/8/9]

soma

[7/8/9]

sóma

old/184

[1/2] ʊʂk

[1/2] ʊʃk

[3] ɔʃkEQ031A

[4] ʊʃ

[5] hui

[6] uʃi

[7/8/9] ɲiŋpa

[7/8/9] ɲɪŋpa

[7/8/9] ɲìŋba

red/172

[1/2/3]

ʃʋig

[1/2/3]

ʃʋig

[1/2/3]

ʃʋig

[4]

ʃʋik

[5/6]

mãĩ

[5/6]

mãĩ

[7/8/9]

marbo

[7/8/9]

marbo

[7/8/9]

márʋo

round/190

[1/4/7/9] baʈlɛs; girgir

[2]

baʈlos

[3]

baʈlas

[1/4/7/9]

girgir

[5/6/9]

kirkir

[5/6/9]

kirkir

[1/4/7/9]

girgir

[8]

tɔktɔk

[1/4/5/6/7/9] kírkir; gìrgir

small/32

[1/2/3/4] zɪgɪʦ; gaʈo

[1/2]

gaʈo

[1/3/4]

dzigits

[1/3/4]

dzigitEQ031A

[5]

ətsə

[6]

tsɪgdza

[7/8/9]

tʃun

[7/8/9]

tʃun

[7/8/9]

kúrkur; tʃýn; tʃún

straight/189

[1/2]

sɔlɖɛs

[1/2]

sɔlɖɛs

[3]

sɔlɖas

[4]

silʈa

[5]

pɔdəra

[6]

kʰosra

[7]

ʈaŋbo

[8]

ombo

[9]

ʈʰáŋbo

warm/180

[1/2/3/4]

bɔk

[1/2/3/4]

bɔk

[1/2/3/4]

bɔk

[1/2/3/4]

bɔkEQ031A

[5]

tatʰəra

[6]

kotʃʰra

[7/9]

ʈønmo;

ʈonmo

[8]

toŋpa

[7/9]

ʈø̀nmo

wet/194

[1/4]

pintʃ; tʰis

[2]

spenǝk

[3]

pɪnk

[1/4]

tʰis

[5]

rakʃiː

[6]

tʰɪsi

[7]

lʉnpa

[8]

lemba

[9]

lánte

white/175

[1/2/3/4]

ʈʰog

[1/2/3/4]

ʈʰog

[1/2/3/4]

ʈʰog

[1/2/3/4]

ʈʰog

[5/6]

tʃãĩ

[5/6]

tʃai

[7/8/9]

karʋo

[7/8/9]

karbo

[7/8/9]

kárʋo

yellow/174

[1/2]

pig

[1/2]

pig

[3/4]

pik

[3/4]

piːk

[5/6]

lei

[5/6]

lei

[7/8/9]

serʋo

[7/8/9]

sɛrbo

[7/8/9]

sérʋo

Table 56

Automatic comparison of some adverbs of time

Sa [1]

Ni [2]

Ka [3]

Ro [4]

Ch [5]

La [6]

Po [7]

Ku [8]

Na [9]

today

[1/2/3/4/6]

tɔrɔ

[1/2/3/4/6]

tɔrɔ

[1/2/3/4/6] toro

[1/2/3/4/6]

tɔrɔ

[5]

tʰan

[1/2/3/4/6]

tɔrɔ

[7/9]

tirɪŋ

[8]

derɪŋ

[7/9]

tìrɪŋ

yesterday

[1/2/3]

me

[1/2/3]

[1/2/3]

me

[4]

mɛʃpa

[5]

nei

[6]

ʃɪraŋ

[7/8/9]

daŋ

[7/8/9]

daŋ

[7/8/9]

ⁿdã̀ŋ

1 day bef. y.

[1/2/3/4]

ri

[1/2/3/4]

ri

[1/2/3/4]

ri

[1/2/3/4]

ri

[5]

tubrja

[6]

tʊʃɪraŋ

[7/9]

kʰeniʃak

[8]

kʰarɲɪŋ

[7/9]

kʰɛ́niʃakEQ031A

2 days bef. y.

[1]

rɪgtsɔmja

[2/3/4] rɪktsɔmja

[2/3/4] rɪktsɔmja

[2/3/4] riktsɔmja

[-] -

[6]

pitu ʃɪraŋ

[7]

dʒinɪŋ

[8]

dʒerɲɪŋ

[9]

dʒìniʃak

3 days bef. y.

[1]

rɪktsʊ ɔmja

[2] r

ɪktsɔmjʊ ɔmja

[3]

rɪktʃamjaʊ ɔmja

[4] rɪktsɔmjaktsʊ ɔmɪa; rɪktsumɪa omɪa

[-] -

[6]

itu ʃɪraŋ

[-] -

[-] -

[9]

gùniʃak

4 days bef. y.

[-] -

[-] -

[-] -

[-] -

[-] -

[-] -

[-] -

[-] -

[9]

tʃʰúniʃak

tomorrow

[1/2/3/4]

nab; nasom

[1/2/3]

nab

[1/2/3]

nab

[1/4]

nasom

[5]

obi

[6]

ŋaɪro

[7/8/9]

naŋmo

[7/8/9]

naŋmo

[7/8/9]

nàŋmo

1 day aft. t.

[1/3/4]

rɔmi

[2/6]

rome

[1/3/4]

rɔmi

[1/3/4]

rɔmi

[5]

nirja

[2/6]

rɔmɛʈ

[7/8/9]

naŋ

[7/8/9]

naŋ

[7/8/9]

náŋ; náː

2 days aft. t.

[1/3]

paŋe

[2/4/6]

pãɛ̃

[1/3]

paŋe

[2/4/6]

pãẽ

[5]

barja

[2/4/6]

pajɛʈ

[7/8/9]

dʒe

[7/8/9]

dʒe

[7/8/9]

dʒèj

3 days aft. t.

[1/3]

tʃɛŋe

[2]

ɛ̃ẽ

[1/3/6]

tʃɛŋe; ɛŋe

[4]

emi

[5]

tʰerja

[3/6]

ɛŋɛʈ

[7]

guiʃak

[8]

naŋmo naŋdʒe

[9]

gùi

4 days aft. t.

[1/3]

ʈɛŋe

[2]

tʃɛ̃ẽ

[1/3]

ʈɛŋe

[4]

tʃemi

[5]

koɲa

[-] -

[7]

ʃuiʃak

[-] -

[9]

tʃʰúi

Table 57

Automatic comparison of numerals

Sa [1]

Ni [2]

Ka [3]

Ro [4]

Ch [5]

La [6]

Po [7]

Ku [8]

Na [9]

one/22*

[1/2/3/4/5/6] ɪd

[1/2/3/4/5/6] ɪd

[1/2/3/4/5/6] ɪd

[1/2/3/4/5/6] ɪd; i

[1/2/3/4/5/6]

i

[1/2/3/4/5/6]

i

[7/8/9]

tʃik

[7/8/9]

tʃɪk

[7/8/9]

tʃíkEQ031A

two/23*

[1/2/3/4/6]

nɪʃ

[1/2/3/4/6]

nɪʃ

[1/2/3/4/6]

nɪʃ

[1/2/3/4/6]

nɪʃ

[5]

niʃi

[1/2/3/4/6]

niʃ

[7/8/9]

ɲiː

[7/8/9]

ɲiː

[7/8/9]

ɲíː

three/24*

[1/2/3/4]

ʂum

[1/2/3/4]

ʃum

[1/2/3/4]

ʃum

[1/2/3/4]

ʃum

[5]

homo

[6]

hʊm

[7/8/9] sum

[7/8/9]

sum

[7/8/9]

súm

four/25

[1/2/3/4/5/6] pa

[1/2/3/4/5/6] pə

[1/2/3/4/5/6] pə

[1/2/3/4/5/6] pə

[1/2/3/4/5/6] pə

[1/2/3/4/5/6] pə

[7/8/9]

ʒi

[7/8/9]

ʒi

[7/8/9]

ʒì

five/26

[1/2/3/4/5/6/ 7/8/9] ŋa

[1/2/3/4/5/6/ 7/8/9] ŋa

[1/2/3/4/5/6/ 7/8/9] ŋa

[1/2/3/4/5/6/ 7/8/9] ŋa

[1/2/3/4/5/6/ 7/8/9] ŋa

[1/2/3/4/5/6/ 7/8/9] ŋa

[1/2/3/4/5/6/ 7/8/9] ŋa

[1/2/3/4/5/6/ 7/8/9] ŋa

[1/2/3/4/5/6/ 7/8/9] ŋá

six

[1/2/4/5/6/9] ʈug

[1/2/4/5/6/9] ʈug

[3/4/5/6/8/9] ʈʊk

[1/2/3/4/5/6/ 8/9] ʈukEQ031A

[1/2/3/4/5/6/ 8/9] ʈu

[1/2/3/4/5/6/ 8/9] ʈu

[7]

ʈʰok

[3/4/5/6/8/9] ʈuk

[1/2/3/4/5/6/ 8/9] ʈùkEQ031A

seven

[1/2/3/4/5] (s)tɪʃ

[1/2]

stɪʃ

[1/3/4/5]

tɪʃ

[1/3/4/5]

tiʃ

[1/3/4/5]

tiʃ

[6]

ʃiniʃ

[7/8/9]

dun

[7/8/9]

dun; dʉn

[7/8/9]

dùn; dỳn

eight

[1]

[2/3/4]

rajɛ

[2/3/4]

rajɛ

[2/3/4]

raje

[5]

rea

[6]

gɛtEQ031A

[7/8/9]

gjet

[7/8/9]

gjet; gjɛt

[7/8/9]

gjètEQ031A

nine

[1/3/4/5]

gui

[2]

sgui

[1/3/4/5]

gui

[1/3/4/5]

gui

[1/3/4/5]

gui

[6/7/8/9]

gu

[6/7/8/9]

gu

[6/7/8/9]

gu

[6/7/8/9]

ten

[1]

[2/3/4]

sajɛ

[2/3/4]

sajɛ; saje

[2/3/4]

saje

[5]

sja

[6]

sa

[7/8/9]

tʃu

[7/8/9]

tʃu

[7/8/9]

tʃú

twenty

[1/2/3/4/5] niza

[1/2/3/4/5] niza

[1/2/3/4/5] niza

[1/2/3/4/5] niza

[1/2/3/4/5] niza

[6]

nisa

[7/8]

ɲiʃu

[7/8]

ɲiʃu

[9]

nìʃu

thirty

[1]

nizo sɛ

[2/3]

nizo sajɛ

[2/3]

nizɔ saje

[4]

nizau saje

[5]

nizaɔ sja

[6/8]

sumtʃu

[7/9]

sumdʒu

[6/8]

ɲiʃu naŋ tʃu; sumtʃu

[7/9]

súmdʒu

thirty-one

[1/2]

nizo sigit

[1/2]

nizo sigit

[3/5]

nizaɔ sigit

[4]

nizau sigitEQ031A

[3/5]

nizaɔ sigit

[6]

nisau sait

[7]

sumdʒu ʃɔkʃɪk

[8]

ɲiʃu naŋ tʃugʃik;

sumtʃu tʃik

[9]

súmdʒu sɔkʃɪkEQ031A

forty

[1/2/3/4/5]

niʃ niza

[1/2/3/4/5]

niʃ niza

[1/2/3/4/5]

niʃ niza

[1/2/3/4/5]

niʃ niza

[1/2/3/4/5]

niʃ niza

[6]

niʃ nisa

[7/8/9]

ʒɪptʃu

[7/8/9]

ɲiʃuʋa ɲiː; ʒiptʃu

[7/8/9]

ʒìptʃu

forty-one

[1/2]

niʃ nizo ɪd

[1/2]

niʃ nizo id

[3]

niʃ nizaɔ ɪd

[4/5]

niʃ nizau i(d)

[4/5]

niʃ nizau i

[6]

niʃ nisau id

[7]

ʒɪptʃu ʃɔkʃɪk

[8]

ɲiʃuʋa ɲinaŋ tʃɪk; ʒiptʃu tʃik

[9]

ʒìptʃu ʒakʃɪkEQ031A

fifty

[1]

niʃ nizo sɛ

[2]

niʃ nizo sajɛ

[3]

niʃ nizaɔ saje

[4]

niʃ nizau adʰaŋ

[5]

pãẽ

[6]

ʈai nisa

[7]

ɲabtʃu

[8]

ɲiʃuʋa ɲinaŋ tʃu; ɲaptʃu

[9]

ɲèptʃu

sixty

[1/2/3/4/5]

ʃum niza

[1/2/3/4/5]

ʃum niza

[1/2/3/4/5]

ʃum niza

[1/2/3/4/5]

ʃum niza

[1/2/3/4/5]

ʃum niza

[6]

hum nisa

[7]

ʈʰuktʃu

[8/9]

ɲiʃuʋa sum; ʈugtʃu; ʈuktʃu

[8/9]

ʈùktʃu

seventy

[1]

ʃum nizo sɛ

[2]

ʃum nizo sajɛ

[3]

ʃum nizaɔ saje

[4]

ʃum nizau saje

[5]

ʃum nizaɔ sja

[6]

hum nisaɔ sa

[7/8/9]

duntʃu

[7/8/9]

ɲiʃuʋa sumnaŋ tʃu; duntʃu; dontʃu

[7/8/9]

dùntʃu

seventy-one

[1/2]

ʃum nizo sigit

[1/2]

ʃum nizo sigit

[3/5]

ʃum nizaɔ sigit

[4]

ʃum nizau sihi(d)

[3/5]

ʃum nizaɔ sigit

[6]

hu(m) nisau sait

[7]

duntʃu donʃɪk

[8]

ɲiʃuʋa sumnaŋ tʃugʃɪk;

duntʃu tʃik

[9]

dòntʃu dɔkʃɪkEQ031A

eighty

[1/2/3/4/5]

pə niza

[1/2/3/4/5]

pə niza

[1/2/3/4/5]

pə niza

[1/2/3/4/5]

pǝ niza

[1/2/3/4/5]

pǝ niza

[6]

pǝ nisa

[7/8]

gjaʒu

[7/8]

ɲiʃuʋa dʒi; gjaʒu

[9]

gʰèdʒu

ninety

[1]

pə nizo sɛ

[2]

pə nizo sajɛ

[3]

pə nizaɔ saje

[4]

pǝ nizau saje

[5]

pǝ nizao sja

[6]

pɔ nisaɔ sa

[7/8/9]

guptʃu

[7/8/9]

ɲiʃuʋa dʒinaŋ tʃu; guptʃu

[7/8/9]

gùptʃu

one hundred

[1/2/3/4/5]

ra

[1/2/3/4/5]

ra

[1/2/3/4/5]

ra

[1/2/3/4/5]

ra

[1/2/3/4/5]

ra

[6/7/8/9]

gja

[6/7/8/9]

gja

[6/7/8/9]

gja

[6/7/8/9]

gjà

five hundred

[1/2/3/4/5] ŋara

[1/2/3/4/5] ŋara

[1/2/3/4/5] ŋara

[1/2/3/4/5] ŋara

[1/2/3/4/5] ŋara

[6]

ŋagja

[7/8/9]

ŋabgja

[7/8/9]

ŋabgja

[7/8/9]

ŋábgja

one

thousand

[1/2/3/4/5/6] hazaːr

[1/2/3/4/5/6] hazaːr

[1/2/3/4/5/6] hadʒaːr

[1/2/3/4/5/6] hazaːr

[1/2/3/4/5/6] hǝzar

[1/2/3/4/5/6] hadʒaːr

[7/8/9]

tɔŋ

[7/8/9]

tɔŋ

[7/8/9]

tɔ́ŋ; tóŋ

one

thousand

one

[1/3/4/6] hazaːru id; hazaːr id

[2]

id hazaːr id

[1/3/4/6] hadʒaːru ɪd

[1/3/4/6] hazaːru i(d)

[5]

i hǝzar i

[1/3/4/6] hadʒaːru id

[7/8]

tɔŋtʃik

[7/8]

tɔŋtʃɪk naŋtʃɪk; tɔŋtʃɪk

[9]

tɔ́ŋraŋ tʃɪkEQ031A

Table 58

Automatic comparison of question words

Sa [1]

Ni [2]

Ka [3]

Ro [4]

Ch [5]

La [6]

Po [7]

Ku [8]

Na [9]

who/11

[1/4]

had

[2/3/4]

hat

[2/3/4]

hat

[1/2/3/4]

hatEQ031A

[5/7/8/9]

su

[6]

ʊŋ

[5/7/8/9]

su

[5/7/8/9]

su

[5/7/8/9]

what/12

[1]

tʃʰəd

[2]

ʈʰɔ

[3]

tʰǝ(d)

[4]

ʈʰətEQ031A

[5]

kʰe

[6]

tʃʰe

[7/8/9]

tʃi

[7/8/9]

tʃi

[7/8/9]

tʃí

where/13

[1/2/3/4]

ham

[1/2/3/4]

ham

[1/2/3/4]

ham

[1/2/3/4] ha(m)

[5]

go

[-] -

[7]

kana

[-] -

[9]

kànɖu

when/14

[1/2/3/4] teraŋ; tɛraŋ

[1/2/3/4]

terɔŋ

[1/2/3/4]

tɛraŋ

[1/2/3/4]

teraŋ

[5]

home

[6]

taʃpa

[7/8/9]

nam

[7/8/9]

nam

[7/8/9]

nàm

how/15

[1]

hala

[2]

halɛs

[3/4/5]

hale

[3/4/5]

hale

[3/4/5]

hale

[6]

ale

[7/8/9]

tʃukEQ031A

[7/8/9]

tʃuk

[7/8/9]

tʃúkEQ031A

Table 59

Automatic comparison of personal pronouns

Sa [1]

Ni [2]

Ka [3]

Ro [4]

Ch [5]

La [6]

Po [7]

Ku [8]

Na [9]

1sg/1*

[1/2/3/4/5/6] gǝ

[1/2/3/4/5/6] gǝ

[1/2/3/4/5/6] gǝ

[1/2/3/4/5/6] gə

[1/2/3/4/5/6] gə

[1/2/3/4/5/6] gǝ; gʉ

[7/8/9]

ŋa; maŋ

[7/8/9]

ŋa

[7/8/9]

ŋà; mà

2sg.h/2

[1/2/3/4/5]

ki

[1/2/3/4/5]

ki

[1/2/3/4/5]

ki

[1/2/3/4/5]

ki

[1/2/3/4/5]

ki

[6]

giraŋ

[7]

ɲetEQ031A

[8]

rue

[9]

kʰóŋ

2sg.nh/2*

[1/2/3/4/5/6] kǝ; ka

[1/2/3/4/5/6] ka

[1/2/3/4/5/6] ka

[1/2/3/4/5/6] ka

[1/2/3/4/5/6] ka

[1/2/3/4/5/6] ka

[7/8/9]

kʰøtEQ031A

[7/8/9]

kʰøtEQ031A

[7/8/9]

kʰǿtEQ031A

3sg/3

[1/2/3/4/5/6] dɔ; hɔdɔ; hɔnɔ; nɔ; hɔjɔ;

[1/2/3/4/6]

nɔ; dɔ

[1/2/3/4]

do

[1/2/3/4]

ʊno; dɔ

[1/5]

hojo

[1/2/6]

[7/8/9]

kʰɔ

[7/8/9]

kʰɔ

[7/8/9]

kʰɔ

1pl.incl/4*

[1/3]

niŋa

[2/4/5]

niŋ

[1/3]

niŋa

[2/4/5]

nɪŋ

[2/4/5]

niŋsa; niŋ

[6]

nɪŋpaŋ

[7]

maŋʃak

[8]

hotset

[9]

òn

1pl.excl/4

[1]

kiʃaŋ

[-] -

[3]

niʃi

[4]

kaʃaŋ

[-] -

[6]

kirapaŋ

[-] -

[-] -

[9]

màʃakEQ031A; ɲètEQ031A

2pl.h/5

[1/2]

kinɔ

[1/2]

kino

[3]

kiʃi

[4]

kin

[5]

katʃaŋ

[6]

kɪnpaŋ

[7]

ɲiʃak

[8]

kʰeraŋ

[9]

kʰóŋʃakEQ031A; kʰóŋdʒakEQ031A

2pl.nh/5

[1]

kano

[2]

kanego

[-] -

[4]

kan

[-] -

[-] -

[7]

kʰjoʃak

[-] -

[9]

kʰóʋatEQ031A; kʰóʃakEQ031A

3pl/6

[1/2/3]

dɔgɔ; hɔdɔgɔ; honogɔ; nɔgɔ

[1/2/3]

nʊgɔ; dɔgɔ

[1/2/3]

dogo

[-] -

[5]

homo tetpaŋ

[6]

dɔmi

[7]

pia

[8/9]

kʰoʋa

[8/9]

kʰóʋatEQ031A; kʰóʃakEQ031A

1

The first four villages are situated in Lower Kinnaur, Kuno and Labrang are located in Middle Kinnaur, while the last three are villages of Upper Kinnaur (see Chapter 1).

2

Because the investigations described in this chapter were conducted before undertaking the more detailed phonological analysis underlying the phonemic orthography used in Chapter 2, the transcription system used for (Sangla) Kinnauri in this chapter for all lexical comparisons differs in some details from that used in Chapter 2. However, in the interest of verifiability and reproducibility of results, we have elected to retain the earlier, less phonemic transcription here.

3

No verbs are included in the comparisons. Verbs were included in the basic vocabulary questionnaire (see Appendix 5A to this chapter), but were provided in such a variety of different (basic) forms by language consultants, that it was not feasible to attempt to harmonize them at this stage, without much more knowledge of each of the varieties.

4

Kuno is not unique in this respect among ST languages. Mazaudon (2010: 124–131) describes parallel decimal and vigesimal numeral systems in Dzongkha (dzo), and a similar situation is found in Bunan (bfu) (Widmer 2017) and Kanashi (xns) (Saxena and Borin 2022a).

5

Written Tibetan: stoṅ ‘thousand’ (Bielmeier et al. MS 2008).

6

The term “base” is used here to refer to the number system base, 10 or 20 in the KST varieties under discussion, and its multiples.

7

The elements added between tens and ones in Poo and Nako resemble those of the (Lhasa) Tibetan system, where a different element is used for each decade (Bell 1939: 68 f.). See also Chapter 3, Section 3.5.

8

We have more detailed data of Kinnauri, which exhibits a range of third person pronominal forms (see Chapter 2). The forms found in the various KST varieties of the Sangla group show some similarity with one or the other form found in Kinnauri. The only exception is Ropa, which has ʊno as the third person singular pronoun. This form is not found in Kinnauri.

9

Out of all the KST varieties investigated, the data on the Sangla variety is the most extensive (see Chapter 2).

10

Third person non-anaphoric pronouns (in object form) in Sangla are, for example, hudu, do-u.

11

Parallel entries for all three dialects are found for many, but not in all cases.

12

An exception in this regard is Webster (1991).

13

While the “Rung” label has been used at least since Thurgood (1984), its actual content has varied, it is not generally accepted among Sino-Tibetanists, and Thurgood (2017: 24 f.) himself seems to have abandoned it (although this is not completely clear from the presentation in Thurgood 2017). However, the fact that it is presented in a handbook-style publication such as Thurgood and LaPolla (2017) motivates its inclusion here.

14

Hindi, which is the official state language of Himachal Pradesh, is generally understood by the people of Kinnaur. During data collection, when needed, Hindi was used as the contact language, as it is more widely understood than, e.g., English.

  • Collapse
  • Expand