given language by the speech community. It is our observation that in Tibeto-Burman languages, Indigenous concepts, categories and classifications of time are usually grammatically encoded in adverbial forms, while the influential Indo-Aryan languages of the region mostly make use of nominal morphology in order to express temporal concepts. In addition, reflexes of Proto-Tibeto-Burman (hereafter PTB) nouns are still evident across the language family. To conclude, we position this survey as a comparative and analytical contribution which focuses attention on the region’s rich linguistic variation and the importance of rigorous documentation, conservation and revitalisation programs for Indigenous languages of the Tibeto-Burman family, as the communities who speak these languages continue to grapple with severe socio-political challenges and face the hegemonic pressures of linguistic assimilation.

Keywords: Comparative linguistics, Endangered languages, Himalaya, Language documentation, Tibeto-Burman, Time

Introduction^[1]

How people experience, categorise, and realise time is anything but uniform across human languages. In our globalised and interconnected world, standardised international systems of time and temporal management often run counter to or challenge culturally-specific temporal metrics, some of which we highlight in this contribution. Much like colour (see Turin & Chung 2018), time represents a cognitive abstraction expressed through a rationalisation and observation of the physical world; principally the sun’s movement throughout the day, manifested in dawns and dusks, sunrises and sunsets, the changing phases of the moon, and the seasons of the year. However, while research into the physiology of colour and its interpretation has significantly shaped our understanding of how—both scientifically and culturally—our retinal cones and cognition determine what, for example, ‘blue’ is and how it is perceived, time remains more opaque and under-theorised in ways that invite greater scrutiny and attention.

In this article, we present and compare the time systems of thirteen Tibeto-Burman languages, which constitute a subgroup of the greater Sino-Tibetan family through a careful analysis of lexicon and grammar (van Driem 2011). Our primary focus is on the Tibeto-Burman languages spoken in Nepal, a recognized ‘language hotspot,’ and within that, the ethnolinguistic Rāī-Kiranti language group spoken in the eastern part of Nepal (Turin 2008; Harrison 2008, 2). We supplement this analysis with data from neighbouring Tibeto-Burman languages. Drawing on Turin’s long-term fieldwork with the Thangmi community, also known as ‘Thami,’ the article focuses in particular on the varieties of Thangmi spoken in Dolakhā and Sindhupālcok districts of central eastern Nepal.

Tibeto-Burman languages are a subgroup within the contested Sino-Tibetan family, the latter including the diverse range of Chinese languages. Taxonomic and genetic descriptions of the Tibeto-Burman family are rife with complexity and disagreement. For example, van Driem (2011) proposes the term “Trans-Himalayan” to better describe and acknowledge the linguistic geography of the region (32). Tibeto-Burman languages are spoken from Kashmir to Vietnam, and thanks to massive aerial variation and internal diversity, genetic affiliations remain disputed (Bradley 1997; DeLancey 1987; Kansakar 1993; Thurgood & LaPolla 2003). The term Tibeto-Burman, however, continues to be widely used to describe this language grouping. Given the level of diversity within the language family, then, it is perhaps unsurprising that conceptualisations of time also vary widely.

Loan Words in Tibeto-Burman Language

When discussing loans, we would do well to bear in mind Wierzbicka’s (2008) good counsel: We must challenge ourselves to distinguish between instances when there is an actual lexical gap versus situations that are better explained as an alternative viewpoint or worldview that cannot be easily lexicalised or rationalised in another language. To that end, we take no position on factors influencing specific borrowing, and we restrict ourselves to locating and describing cognates and loans of particular note and interest when present in the data. Well-attested across the Tibeto-Burman family, lexical borrowings can and do arise for a multitude of reasons.

However, in acknowledgement of the diversity and vastness of the language family, we find ourselves under-qualified to retrace and reconstruct the more complex histories of borrowed terms that we identify in the data. Many Tibeto-Burman languages—and certainly those spoken in Nepal—borrow lexical terms from socially dominant and politically standardised languages such as Nepali, an Indo-Aryan language which is constitutionally enshrined as the official language of Nepal. In some cases, two-time systems that differentiate between segmented time (e.g. hours and minutes) and observable episodes (e.g. midday, sunset/sunrise) may exist in either free variation or complementary distribution in the same lexicon. Simply put, foreign time systems and temporal forms can either replace existent Indigenous forms in a lexicon or work in concert with them to appear in specific environments. Sunwar offers one such example, in which Nepali is used alongside an Indigenous system to express concepts of time (Borchers 2008). Wambule and Jero examplify languages that have integrated and hybridised time-related terms from Nepali into their Indigenous lexicons (Opgenort 2004).

In his work on lexical borrowing, Grzega (2003) identifies a number of factors that can help us describe the abundance of borrowed terms relating to time in Tibeto-Burman languages. Explanations for borrowing include, among others: The “feeling of insufficiently differentiated conceptual fields [between speaking groups]” (Weinreich 1954, 59 as cited by Grzega 2003, 23); the “rise of a specific conceptual field [like new technologies]” (Grzega 2002, 1030 as cited by Grzega 2003, 23); the “political or cultural dominion of one people by another” (Fritz 1988, 1622 as cited by Grzega 2003, 23); the “mere oversight or temporary lack of remembering the indigenous name [or word]” (Weinreich 1954, 60; Baranow 1973, 138; Tersh 1978, 209, 214 as cited in Grzega 2003, 23); and the “low frequency of indigenous words and instability of words within a region” (Weinreich 1954, 57; Scheler 1977, 88 as cited in Grzega 2003, 24).

In the context of the Tibeto-Burman languages spoken in Nepal, the most salient of the factors outlined above to explain borrowing is the “political or cultural dominion of one people by another” (Fritz 1988, 1622 as cited by Grzega 2003, 23). The rapid transformation of traditional lives and livelihoods—through rampant urbanisation, the impact of state media, economic pressures, social upheaval and compulsory education—may further accentuate the “feeling of insufficiently differentiated conceptual fields” (Weinreich 1954, 59 as cited in Grzega 2003, 24), the subsequent “rise of a specific conceptual field” that may have been previously absent in these languages or, in the case of time, the emergence of an unmeasured category (Grzega 2002, 1030 as cited in Grzega 2003, 23). In addition, Wallace (2005) suggests that time is inherently political and that it serves as a mechanism to maintain ideologies of power. As he explains,

all societies produce and maintain maps of time, historical and futuristic frames of reference, large constructions, generally beginning in the past with a creation myth, providing chronicles of relatively recent histories of significant events (the ‘now’), and in some cases extending far into the future, possibly to an end of the world or to a cyclical renewal. (Wallace 2005, 5)

Time can be conceptually and systematically regulated and used to enforce the legitimacy of a nation state, a dominant culture or a political ideology. It is therefore quite reasonable to suggest that terminology relating to time might readily be substituted by, or assimilate to, the terminology of a more dominant group.

One diffusional feature of cultural significance in the High Himalayan region is the presence of regionally dominant religions, liturgical traditions, and their effects on local customs and Indigenous religions. Grounded in the heritage of the Indian subcontinent, astrological principles such as the kālachakra (the wheel of time) observe lunar cycles (Ramble 2013). These Indian influences have contributed foundational features to Tibetan Buddhism through, in particular, the introduction of lunar tables in timekeeping (Erlewine 2012, 36). Moreover, Erlewine (2012) suggests that “Tibetan astrology is inextricably bound to Tibetan Buddhism” (31), and argues that the two concepts cannot be disassociated from one other.

Referring to more culturally Chinese areas of influence, Ying-chin (1999) notes that temporal borrowings have considerable sociolinguistic and philological significance. Acknowledging the contemporary use of the Gregorian calendar throughout larger China, Ying-chin emphasises that, “by observing what minorities without independent written characters call the twelve months, we may gain insight into their speakers’ Indigenous conceptualisations of time, as reflected in their own languages” (1999, 71). Well-documented Qiangic languages like Ersu and Gyarong, both spoken in the Chinese province of Sichuan, offer compelling illustrations of the adoption of Chinese temporal elements, most evident in the annual sequence of the Chinese zodiac. However, given that the observation of heavenly bodies is a universal human tendency—across time and space—it is impossible to say how much the Indian or Chinese lunar systems have directly influenced Indigenous calendars in the Himalayas that may already have been lunar-centric prior to sustained contact with dominant regional and religious traditions.

We identify the possible religious motivation for the adoption of foreign vocabulary into Indigenous Himalayan lexicons as a fertile area for future research. Nevertheless, as noted above, borrowings are not the focus of this paper. In the comparative analysis that follows, we focus on similarities, trends and differences in the temporal systems pf Tibeto-Burman languages, making reference to loaned terminologies when relevant to the social and linguistic context.

Methodological Framework

In this contribution, we concentrate on contrasting segmental and specific time (e.g. ‘the day after tomorrow,’ ‘dates,’ ‘months,’ ‘years’) rather than focusing on a comparative analysis of relative time constructs (e.g. ‘soon,’ ‘then,’ ‘earlier’). Underscoring the complexity of understanding time cross-linguistically, Wallace reminds us that words like ‘now’ can encompass all measures of time depending on the context in which the term is used (2005, 1). Although time is in many ways relative, the research data we present invite further discussion and comparison of the various temporal systems and constructs used across Indigenous groups in the Himalayan region. Temporal polysemy is a related and interesting avenue of research, although not our primary emphasis in this contribution (Evans 2005).

Our analysis in this paper moves from the more reserved, smaller temporal systems to the more complex and idiosyncratic ones. While the focus of this paper is centred on the linguistic presentation and analysis of temporal systems and less on the related cultural contexts in which these language are spoken, this decision is

simply a practical choice in response to space constraints and in no way a judgement on the relative merits of one approach over another.

When immediately relevant to the linguistic analysis in ways that would aid better understanding of temporal frames, we provide ethnographic and cultural context. We encourage interested readers to engage with the references identified in our bibliography for more detailed information about the livelihoods and cultures of the speakers whose languages we celebrate in this article.

Although none of the languages we review in this comparative survey could be called ‘temporal outliers,’ some systems are strikingly more intricate than others. Moreover, many languages with more developed temporal schemes borrow and appear to have assimilated time systems from neighboring languages, while at the same time still maintaining a large Indigenous temporal lexicon. For this reason, such languages in particular excite our curiosity with regard to the cognitive and social aspects of time management and organisation, and the many ways that these are expressed together. Beginning in the next section, “Basic Tibeto-Burman terminology in a Comparative Perspective, we present an in-depth analysis of what we consider to be more standard time terminology in the Tibeto-Burman family. This is then followed in the next section, Complex Tibeto-Burman Time Terminology through a Comparative Perspective, by a presentation and analysis of the more idiosyncratic systems attested in other languages in the language family. Consolidated comparisons between both data sets follow in the Discussion Section with an emphasis on apparent cognates and shared paradigms.

Basic Tibeto-Burman TIme Terminology in a Comparative Perspective

The languages presented in this section include Lepcha, Sunwar, Dhimal, Wambule, Jero, Kham, and Dolakha Newar. In these languages, we identify more predictable and common terminology relating to time including loans from Nepali or other Indo-Aryan languages, such as Hindi.

Lepcha

Lepcha is a language spoken in Sikkim, Darjeeling, the Ilām District in Nepal and areas of Bhutan, and is home to a relatively classic temporal paradigm within the parameters of the Tibeto-Burman family.

Like its sibling languages, most Lepcha terminology that relates to specific points in time is constructed through the compounding of lexemes. Compounding is a common mechanism with which to create new meaning from existing lexical and grammatical forms. As a result of the many monosyllabic morphemes in Tibeto-Burman languages, compounding is unsurprising in time terminology (Matisoff 1978). For example, Lepcha lúkʔál can be glossed as ‘tomorrow’: a construction comprised of the verbal and nominal lúk ‘to get up, rise’ or ‘morning’ and ʔál, which means ‘new’ (Plaisier 2006, 99). Certain suffixes also combine to describe sequences of days in the near past in the same manner as in Table 1 (Plaisier 2006, 99). Nevertheless, some terms in Lepcha such as ‘today’ and ‘yesterday’ appear to be monomorphemic, such as saróng and tasó (shortened to só) respectively (Plaisier 2006, 99).

In Lepcha, specifying the time of day distinguishes ‘daytime’ (sanyí or sanyím or suknyím) from ‘nighttime’(sonáp). While saʔyák refers to ‘day and night,’ and thus constitutes a full cycle, an alternative way of expressing one full cycle or time period is simply to add ‘daytime’ and ‘night’ together: Sanyí sonáp (Plaisier 2006, 99). The concept of ‘midday’ is constructed by adding the suffix phet ‘half’ to sanyím to create sanyím phet (or the shortened, nyímphet). By extension, ‘midnight’ is sonápphet (Plaisier 2006, 99). From this we note that while compounding is common in Lepcha, the hybrisation of Indigenous and borrowed time systems is not attested in the language although this phenomenon is common in other closely-related languages.

Sunwar

Sunwar, spoken in central eastern Nepal, exhibits a relatively transparent array of temporal adverbs when compared to the systems of neighbouring Tibeto-Burman languages. Many time-related terms in Sunwar are also borrowed from Nepali. Borchers (2008) attests that, “references to a certain time of day are always made in Nepali,” as in dui baje, ‘two hour or two o’clock’ (Nep.) [sic, recte: ‘at two o-clock’] (89). Nevertheless, as in other Tibeto-Burman languages, compounding is prevalent in Sunwar. Common bases for Sunwar compounds include nāt ‘day’ and lādo ‘night’ (Borchers 2008, 91).

In Table 2 above, the Sunwar morpheme mul can be isolated to mean ‘new,’ conceptually similar to constructions with ʔál in Lepcha (Borchers 2008, 89). Other common Sunwar terms that constitute the present or near future include disā ‘tomorrow, next day,’ ici ṅā ‘now, nowadays,’ mulayo ‘now, today,’ nātre ‘all day,’ and nādore ‘all night.’ While disā signifies ‘tomorrow or next day,’ the morphologically and phonologically similar term adisā ‘next day’ does not mean ‘tomorrow’ (Borchers 2009, 90). Here, adisā references future

points in time that remain unspecified, while disā can refer to both general and specific points in the future. The following narrative excerpts of Sunwar dialogues present the contextual use and application of the term adisā.

In 1 and 2 above, Sunwar adisā refers to the future, but not to a specifically-defined ‘tomorrow.’ Rather, adisā signifies temporal progression as a whole and does not appear to fix time with the precision that disā would.

Dhimal

Dhimal is a language spoken in the Jhāpā and Moraṅ districts of southeastern Nepal. Like Sunwar, specific time is usually expressed through Nepali numerals using Nepali terms for hours and days (King 2009).

In Dhimal, ni or nani are synonyms for ‘day’ (King 2009, 89, 566). It may be that ni and nani, which both contain the morpheme <-ni>, are modern reflexes of the PTB *nəy meaning ‘day’ and thus cognate with many other extant Tibeto-Burman languages (Opgenort 2005, 82).

Unlike Sunwar and other Tibeto-Burman languages, however, the descriptive scope of time in Dhimal is noticeably more restrictive. Starting from the present, temporal increments only generally extend to two steps into the past or future.

It is unclear to what degree nhuʔdina ‘the following day, next day’ can extend further into the future or whether it is synonymous with jumni ‘tomorrow.’ Its usage may resemble Sunwar adisā in terms of indexing the general future rather than a temporally specific ‘tomorrow’ (Borchers 2008, 90).

In Dhimal, the scope of years resembles the structure noted above for days. And, in the same manner that all day-related terms contain the final element <-i>, so too do all year-related words contain <-bare>.

The form nanibare ‘this year’ is transparently constructed from nani ‘today’ and bare; the latter may be borrowed from Nepali bār meaning ‘day of the week, time, turn, occasion.’ The term anhebare ‘last year’ somewhat resembles to anji ‘yesterday,’ possibly related through the potential prefix <an-> (King 2009, 488).

Dhimal speakers do not appear to use Indigenous Dhimal terms for the concepts ‘week’ or ‘month.’ ‘Week’ is borrowed from the Indo-Aryan eʔ-athar and manifests in Dhimal as a’thar (King 2009, 489, 528). Dhimal speakers can specify at what time events occur through the use of rhi’ma ‘in the morning,’ dilidili ‘around the evening,’ and nhisiŋ ‘night’ (King 2009, 512, 568, 582). If Dhimal speakers seek less specificity but still reference a given time period within a frame, they can use koko ‘same time period’ (King 2009, 544). In a number of these terms, reduplication is attested, also common to Tibeto-Burman languages (Abbi 1990).

Wambule and Jero

Reduplication is a characteristic also observed in Wambule and Jero. Both spoken in eastern Nepal, these two Kiranti languages are very closely related to one another. For this reason, our current analysis is constructed to facilitate their lexical comparison. Nouns of time in both Wambule and Jero are Nepali loans, while adverbs are Indigenous (Opgenort 2005). This further underscores our observation that in Tibeto-Burman languages, the Indigenous concepts, categories and classifications of time are usually encoded in the grammar in adverbial form, whereas the influential Indo-Aryan languages of the region use nominal morphology to convey time. Other languages in which this tendency is observed include Darma and Thangmi.

The Indigenous temporal terminologies of Wambule and Jero are complex and fuse with Indo-Aryan loans to create unique local forms. For example, in Wambule, byala ~ byal ‘time’ from Nepali can co-exist in forms with the Indigenous demonstratives <-im> ‘that’ and <-yas> ‘this’ to mean ‘then, that time, at that moment’ and ‘now, this time, at this moment’ respectively (Opgenort 2004). However, the majority of time adverbs are 

indisputably Wambule in origin and convey relative and general time, not specific increments. Incremental time is generally conveyed using Nepali terms.

The Āmboṭe dialect of Jero attests the suffix <-so>, which resembles the identical Wambule suffix that signifies ‘day’ (Opgenort 2005, 82). Likewise, <-ni> of the Āmboṭe dialect of Jero is cognate with Wambule <-ɖi>, also signifying ‘day,’ and Khaling <-ne> as evidenced in aathaasne ‘day before yesterday’ (Opgenort 2005, 82). This morpheme appears to be a reflex of PTB *ney, which has been reconstructed to signify either ‘day’ or ‘sun’ (Opgenort 2005, 82).

Many Wambule and Jero adverbs of time are identical or near identical, specifically in the southern Mohanṭāre dialect of Jero. For example, thaːthaccum ‘three days ago’ is the same in both Wambule and Jero (Opgenort 2005, 122; Opgenort 2004, 242).

Wambule and Jero also share terms for ‘two days ago’ and ‘yesterday,’ thaccum and saiso respectively.

Regarding annual increments, the Āmboṭe dialect of Jero makes use of both the suffix <-thoce>and the prefix <tho-> to mean ‘year,’ while in the Mohanṭāre dialect, only the suffixal form is attested (Opgenort 2005, 122). If one compares this morpheme and its variants with the terms in Table 7 (Indigenous Annual Time Constructs in Wambule), it is evident that these constructs arise through compounding with a cognate suffix in Wambule. For example, tyaŋthoce ~ tyaŋthoɖ ~ tyathoce ~ tyathoɖ‘this year’ is likely prefixed with the attested tyaŋo ~ tyaŋ ‘now, from now on’ to signify ‘this year’ or more literally, the year from now on (Opgenort 2007, 240). 

Notwithstanding phonetic alterations across dialects, Tibeto-Burman tendencies clearly surface in both Wambule and Jero in the form of compounding and the reduplication of temporal lexicon.

Khan

Kham is spoken in the Rukum and Rolpā districts of Nepal. Kham’s base temporal adverbial inventory is relatively restricted, with only about twenty specific time terms and approximately twenty relative time terms (Watters 2002). Like Lepcha, after a certain point, Kham uses a specific set of compounded terms to express days (e.g. ‘hence’) with a prefix serving as the element of a lexical item that alters meaning.

Kham <-chyã> and <-chim>, which occur frequently in temporal constructs, both signify ‘day,’ and are likely related to one other. Moreover, in isolation, chyam signifies ‘day’ (e.g. ‘that day’) (Watters 2002, 144) with achim functioning as a combination of the proximate locative element <a-> and the word for ‘day’ to signify ‘today’ (Watters 2002, 129, 144). This locative prefix also appears in achya ‘earlier today.’ It is worth noting that the two terms for ‘day’ can combine with the same prefix to produce different meanings, making them semantically distinctive.

While Watters verifies that the prefix <nihm-> is derived from ‘two’ in Kham (<neh->) and also appears in nihmni ‘year after next,’ the internal etymology of other particles that appear on ‘three days after’ and so on are less transparent.

Although <-ni> is a commonly occurring morpheme in Kham, it does not appear to be related to reflexes in Wambule, Jero, or Khaling that signify ‘day.’ Instead, in Kham, <-ni> is likely derived from the PTB *s-niŋ ‘year’ (Watters 2002, 145). In fact, aĩhsi is a clear derivative of the PTB *asniŋ ‘this year,’ which manifests as asni in the Maikot dialect and as ahiŋ in the Gamale dialect of Kham (Watters 2002, 145).

In common with other Tibeto-Burman languages, Kham adverbs can be compounded to create new forms. For example, achya ‘earlier today’ can combine with chəkalnya to mean ‘this morning’ (achya chəkalnya), ahjya ‘earlier this year’ combines with uhbyali means ‘last summer’ (ahjya uhbyali), pəte ‘later today’ combines with re:-lə to create ‘tonight’ (pəte re:-lə), and pəte combines with rihm-kə ‘at dusk’ to form ‘this evening’ (pəte rihm-kə) (Watters 2002, 145).

The Kham numeral classifier <tə-> can also combine with specific time lexemes to introduce a sense of vagueness, as in tə-rim-kə ‘a few years ago’ (Watters 2002, 145). This construction is particularly useful in narratives, as shown in example 7 above, where tə-cha ‘one day’ conveys an unspecific temporal period.

It is noteworthy that the storyteller refers to the Nepali month phalgun, part of the Hindu calendar, in the same sentence as using the Kham construction ahjya uhbyali-kə ‘in earlier spring.’ This example further illustrates the mixing of time terminologies from genetically distinct language families among speakers of Himalayan languages.

In addition, in example 9, we note that in Kham, sal appears as another word for ‘year’, loaned from Hindi and Nepali (saal or sāl, respectively) and therefore likely not a reflex of PTB *s-niŋ or a form of rim as indicated by Watters (Watters 2002, 449). In Kham, as in all languages, there is variation in both frequency and preference of using loaned time terms over Indigenous ones.

Dolakhar Newar

Dolakha Newar, a conservative dialect of Newar, has numeral classifiers for a range of noun types that include time constructs. The variation between dialects of Newar can be observed in the term for ‘this year,’ which is thāpre in the Dolakha dialect and thapāle in the Kathmandu variant, showcasing epenthesis and a different liquid consonant in the latter (Genetti 2009, 39).

Within the wide range of Dolakha Newar numeral classifiers, nu is used for counting days, lā correlates with months, and da to years (Genetti 2009, 69). It is plausible that nu is a reflex of PTB *ney as attested in related forms attested in Jero, Wambule, and Khaling (Opgenort 2005, 82). Additionally, common time constructs in Dolakha Newar do not only exist as classifiers, but can also function adverbially, as in kesi ‘tomorrow’ (Genetti 2009, 202).

While the terms outlined above function as numeral classifiers, specific lexemes also exist in Dolakha Newar to express time constructs beyond simple enumeration and counting.

Examples 11 and 12 illustrate differences in the use of various forms of the term ‘day.’ While nu aligns with nis- and pe– to generate the constructions ‘two days’ and ‘four days’ respectively, khunu is modified by -ŋ, an emphasiser, together with the adjective, lita ‘next.’ Regarding relative time constructs, it is also noteworthy that Dolakha Newar libi ‘later’ carries the same gloss in Thangmi.

Example 13 is a sentence taken from A Story of Three Children originally told by Mrs. Kalam Maskey in 1989 reveals some additional terms that Dolakha Newari speakers use in relation to time: dina prati din ‘day by day’ and mās ‘month’ (Genetti 2009, 227). It is possible that mās could be an alternate for lā, which is also attested as ‘month,’ while dina prati din in example 13 is a direct loan from Nepali. The presence of this borrowed phrase indicates the influence Nepali has had on Dolakha Newar and helps highlight how such terms become integrated into contemporary discourse alongside Indigenous terminology.

Example 14 below offers a compelling illustration of the crosslinguistic tendency to utilise Nepali terms to denote specific time, a tendency also noted in Sunwar. In the section “Complex Tibeto-Burman Time Terminology through a Comparative Perspective” below, we offer further evidence of the incorporation of foreign time frameworks into Indigenous systems in terms of not only specific time measurements, but even the counting of months and years.

Complex Tibeto-Burman Time Terminology through a Comparative Perspective

Languages in this section of our analysis include Ersu, Darma, Gyarong, Thangmi, Kulung, and Chepang. Ersu and Gyarong belong to the Qiangic subdivision of the larger family and feature considerable borrowings from Sinitic languages spoken nearby. All languages in this section display divergent and interesting characteristics including distinct Indigenous calendars, extensive sequential systems, and instances of hybridisation with foreign time measures and terminology, to create unique linguistic forms for temporal reference.

When modified by -ma, the term ȵo-ma ‘day’ is used in a more abstract sense, as in ‘the day is good/bad’ rather than in a segmental sense to measure amounts of time (Zhang 2013, 221). Regarding this notion of ‘day’ itself,  Ersu is canonically and characteristically Tibeto-Burman rather than Sinitic. The English term ‘day’ glosses as ȵo-ma (which also means ‘sun’) or ȵo in Ersu. This morpheme closely resembles the PTB lexeme for‘day,’ *ney, and its various cognates (Opgenort 2005, 82; Borchers 2008, 91). In addition to highlighting the presence of an Indigenous etymology, this example showcases Ersu internal syntax and constituency order, which includes head-initial noun phrases compared to head-final ordering, as would be expected in Mandarin (Zhang 2013). Ersu speakers still generally adhere to a SOV sentence structure, despite heavy influence from Mandarin. Nevertheless, in some of following examples, this head-initial directionality is not always followed.

Similar to tə ȵo ‘one day’ in Ersu, we note that tə is also a numeral classifier in Kham with tə-cha carrying the same meaning of ‘one day’ (Zhang 2013, 222; Watters 2002, 145). Regarding temporal concepts for months, Ersu draws heavily from classical Chinese (Zhang 2013, 218). However, while the Chinese system modifies its numerals with yuè ‘moon’ to render a twelve-month system, Ersu uses ɬa while still appearing to mirror the cardinal Chinese structure.

Additionally, while there are no discrete seasons in the Ersu language attested by its speakers, a pattern of three-month installations that Zhang observes roughly correlate with the four canonical seasons:

In Ersu, months are generally divided into two categories: əI ‘white’ (an optional particle) or nua ‘black’ (which is obligatory) (Zhang 2013, 221). This distinction relates to whether a speaker is referring to the first half of the month when the moon becomes its brightest (hence ‘white’) or the latter half when it appears darkest (hence ‘black’). Zhang notes that this cycle shows that, “in Ersu, there is no number larger than ‘16’ referring to the days of a month [and] … the notion of ‘month’ is borrowed from Mandarin Chinese judging by this half-month circulation” (Zhang 2013, 220). This bi-monthly lunar sensitivity is likely a Buddhist influence (Erlewine 2012).

The Ersu term for ‘year’ is expressed by three different morphemes indexing various interpretations and meanings. First, bùtshə̀, the most common morpheme, can be preceded by a variety of markers (e.g. numeral, demonstrative, interrogative).

Second, Ersu ə^I also signifies year, but the term only functions in reference to the Chinese zodiac. Ersu ə ^I, a diminutive in Mandarin Chinese, glosses as ‘year’ and as a free morpheme in Ersu (Zhang 2013, 214). Finally, Ersu xi is also attested for ‘year’ and has a more general scope. Ersu xi is described by Zhang as a “temporal shifter,” functioning in a deictic manner to reference time similar to ȵo, but also including a distant scope as observed in example 21 below (Zhang 2013, 215-216).

Darma

Darma is a language spoken in the state of Himachal Pradesh in India as well as the Dārculā district of Nepal. Darma finds itself in a complex place both genetically and geo-politically, under influence from and in interactionwith India, Nepal, and China (Willis 2007). Historically, linguists have disagreed on Darma’s location within the Tibeto-Burman family, but most now agree that Darma forms part of the western branch of Himalayan languages (Willis 2007). Like its relatives, Darma generally expresses time through the use of adverbs that are distinguished from other locative adverbs and adverbs of manner (Willis 2007).

In Darma, while these adverbs are clause-initial, not all are common in naturally occurring speech with “few [attested] examples from direct elicitation sessions” (Willis 2007, 455). Adverbs of time that deal with the past are infrequently attested in conversational discourse (Willis 2007), further evidencing the process by which Tibeto-Burman languages grammaticalise temporal constructs as adverbial rather than nominal constructions.

We note that than, which glosses as ‘now’ in Darma, serves as the first element in than ying, allowing ‘this year’ to be analysed as a compound of than ‘now’ and ying ‘year’ (Willis 2007, 588). As compounding is common in Darma in particular and within the Tibeto-Burman family as a whole, such constructions are unsurprising and lead us to understand ‘now’ as a broad dietic reference rather than indicating a narrowly-defined given moment.

There are no terms for days of the week or months in Dharma, and the annual cycles of the speech community reflect their seasonal migrations and subsistence patterns. In Darma, ‘day’ is attested as ɟja, ‘week’ is həbta̪, and ‘month’ is la, which also means ‘moon’ in other Tibeto-Burman languages (Willis 2007, 580).

The word həbta̪ ‘week’ is actually a loan from Nepali. It is therefore of no surprise that when terms like ‘Tuesday’ do appear, they are also borrowed from regionally dominant Indo-Aryan languages (Willis 2007, 422). Indeed, the appearance of Hindi or Nepali is also evident in Darma with mahinā ‘the first month of the calendar’ and sal ‘year’ (Hindi sal or also possibly Nepali sāl) surfacing at times as ‘month’ and ‘year,’ respectively (Willis 2007, 154, 161).

Gyarong

Gyarong, also spelled Gyalrong or Jiarong, is a Qiangic language spoken in Sichuan, China. Gyarong exhibits both Tibetan and Chinese loans in its lexicon, which— in the case of colour terminology—compound together between and across the substrate language. The data presented in this paper draw on the seminal published work of Marielle Prins and relate to the Jiǎomùzú dialects of central Mǎěrkāng County.

Many Chinese loans in Gyarong relate to specific and smaller temporal intervals. For example, ʃiŋtʃi ‘week’ is a direct loan from Chinese and the days of the week further reflect the Chinese system: ʃiŋtʃiji ‘Monday,’ ʃiŋtʃitʃan 

‘Sunday,’ and ʃiŋtʃiwu ‘Friday’ (Prins 2016, 732). Prins (2016) attests that code-switching is common between both Gyarong and Chinese, and that concepts such as days, months, and years are now more typically explained through the use of Chinese numerals. Older speakers, however, may still use Gyarong numerals.

Gyarong numerals are even used for foreign understandings of timekeeping that have been adopted into the language. Like Ersu, Gyarong speakers divide a month into two parts corresponding to the phases of the moon: kətho or thowa are used for the first half of the month when the moon is waxing, equivalent to əI in Ersu. And, kəmbot or ɟowa in Gyarong, equivalent to nua in Ersu, are used for the latter when the moon is waning (Prins 2016, 205). Given that this monthly division is of Chinese origin, one might assume that Chinese lexemes would be used to describe this system. However, and perhaps somewhat surprisingly, native Gyarong numerals are attested in these contexts (Prins 2016).

The other superstrate lexicon from which Gyarong speakers borrow is Tibetan. Predictably, Tibetan influence on Gyarong is most evident in aspects of the lexicon that correspond to cultural aspects of Tibetan philosophy, and some divinations and horoscope readings use Tibetan numerals (Prins 2016). However, while the basic Gyarong counting system uses largely unmodified Tibetan terms, the Tibetan-based zodiac system instead makes use of Gyarong lexemes (see table below):

The only exception to this Indigenous influence is the third month that uses stag ‘tiger’ in Tibetan, rather than the Gyarong equivalent khoŋ (Prins 2016, p. 203).

While lo, a term borrowed from Tibetan, glosses as ‘year,’ təlo and təpa share the same meaning, also ‘year’ (Prins 2016, 215). More specifically, təlo is used for the zodiac, which may harken back to its Tibetan roots. Also functioning as a classifier, tsəla ‘month’ additionally glosses as ‘moon’ (Prins 2016, 727, 754) and the second element <-la> is widely attested in other Tibeto-Burman languages meaning ‘moon, month.’ The concept of ‘day’ is rendered either as təʃnu or ʃnu in Gyarong, terms, which are most likely related to one another (Prins 2016, 77). Interestingly, ʒakma is also attested, which carries the additional meaning of ‘time’ (Prins 2016, 745).

Furthermore, ʃnu surfaces as a suffix in pəʃur or pəʃurʃnu ‘yesterday,’ compounded in ʃnu pəʃnu ‘today,’ pəʃurtɽə‘ the other day; a few days ago,’ and soʃnu or so ‘tomorrow’ (Prins 2016, 730, 762). Intriguingly, soʃnu bears some resemblance to Ersu su+ȵo ‘tomorrow,’ which might indicate that <so-> functions as a prefix meaning ‘next’ if these two words are indeed cognates.

Gyarong speakers also can specify time to the half hour (thirty minutes) with unmarked morphemes.

To describe slightly broader periods, the locative particle <tʃe> is used to connect time with specific events or provide emphasis to them within a time frame. The particle <-j> is an allomorph of <tʃe>, but only attested when used for larger temporal concepts like years (Prins 2016, 265). Days, periods in days, days of the week, and months use only <tʃe>.

Gyarong <-j> and <tʃe> can also be used for festivals and seasons, and are usually preceded by təʒak ‘time, day’ (Prins 2016, 266). The morphemes no, ro, and mo, ‘at the latest,’ ‘later than,’ and ‘just, recent, just at that time’ respectively, also combine to modify expressions and thus make them more relative (Prins 2016, 268-269).

Thangmi

Thangmi also known as ‘Thami,’ is divided into two major dialects—Dolakha and Sindhupalcok—both of which have a notable and extensive assortment of temporal adverbs (Turin 2011).

It should be noted that the exact measurement of what constitutes ‘morning’ compared to ‘afternoon’ is not analogous to Western temporal norms. Rather, within a South Asian context, these terms correlate more closely to meals or one’s daily routine. Thus, 2:00 PM could constitute the ‘evening’ in the correct context (Turin 2011, 336). Like other Tibeto-Burman adverbs, Thangmi morphemes can be further modified with the suffix ka ‘throughout,’ as in unise-ka meaning ‘all day’ (Turin 2011, 337).

Thangmi beryaŋ ‘the time at which / at that time’ is likely a derivation of the Nepali ber ‘period of time,’ as a common feature of temporal expressions in the language, and can precede and hence modify a clause (Turin 2011, 339). Thangmi beryaŋ also appears in question forms that seek to ask ‘at what time’ something is occurring:

The recurring presence of <kal> in the above Thangmi forms relating to years raises interesting questions about its semantic meaning and origin. It is possible that <kal> is a reflex of an Indo-Aryan substrate, since kāl can mean ‘time’ or ‘tomorrow’ in Nepali (Turner 1997, 90). The compounded presence of yaŋ ‘today’ with kal to signify ‘next year,’ as in kalyaŋ, further complicates this query (Turner 1997, 341, 345; Turin 2011, 340).

Regarding specific time, Thangmi expressions calque from Nepali and appear to have been meta-linguistically influenced by the introduction of the analog clock (Turin 2011, 345). When asking for the time in Thangmi, the question itself is morphologically structured with the verb ‘to ring, strike, sound’ syaksa, as in hani syak-Ø-an? ‘What time is it?’ (Turin 2011, 345). Although there are no Indigenous intervals for quarters, the lexeme bakoṭek ‘half’ is used for thirty-minute increments (Turin 2011, 345).

Kulung

Kulung (UNESCO: vulnerable, 18,686 speakers, ISO 639-3: kle), an Eastern Rāī language, has its own calendar system and time-related lexicon.

Kulung system is not analogous to the Gregorian calendar, but does roughly correlate to an evenly distributed twelve months. Tolsma (2006) notes that Kulung la translates as ‘moon,’ but does not draw on comparative evidence that shows that la also means ‘month’ in related languages (253). While there is no clearly observable pattern in the names for Kulung months, at least lonamma does appear to correlate with lonam ‘dry season,’ which is discussed below (Tolsma 2006, 255).

The Kulung have their own defined seasons, either marked at their beginnings or referred to in their entirety.

At least some Kulung seasonal prefixes correlate to attributes for which each period is characteristically known, namely: buŋ means ‘flower’ in observation of blooming in the spring, while cuŋ means ‘coldness’ as winter is generally cold (Tolsma 2006, 232, 234). The suffix <-nam> is polysemous in meaning, including ‘sky, sun, weather’ and ‘time’ itself (Tolsma 2006, 258).

For the description of past, present, and future time increments, and rather like Thangmi, Kulung possesses many single constituent terms to situate events in a timeline.

Kulung prefixes in chindi, dokthum, khetthum, nokthum do not exhibit surface similarities with their numerical equivalents in the language: <nit-> (2),<sup-> (3), <liː-> (4), and <tuk-> (6), respectively (Tolsma 2006, 40). It is also noteworthy that there does not seem to be a term equivalent to ‘in five days’ attested in the language, although it is of course possible that such constructions were simply not documented.

Additionally, Kulung contains phrases that represent relative time references whose bases are actually incremental time terms. In some of these phrases, reduplication is evident. For example, le:pa le:pa ‘day by day’ is a reduplicated form of the noun le:pa, which means ‘afternoon’ (Tolsma 2006, 254). Moreover, jisna translates as ‘in a minute’ despite there being no Kulung word for ‘minute.’ It is possible that jijis ‘a very little,’ jis ‘a little,’ and jisna are all related and variants of jis, which would make ‘in a minute’ the best possible gloss (Tolsma 2006, 264).

There does not appear to be a discrete pattern in the etymology of Kulung terms for years; and the existence of a four-year interval marker, without any accompanying terms, is also noteworthy. With such a rich array of seemingly unrelated and specific terms, Kulung certainly merits further inquiry into its temporal systems and related lexicon. The apparent lack of many distinguishable cognates with related languages is also a source of sustained interest.

Chepang

Chepang, spoken in the Tarai region of Nepal and India, contains an array of specific and diverse lexemes to describe temporal constructs and observations. The dictionary from which this data has been sourced mostly derives its lexicon from the Eastern dialect of Chepang. However, the dictionary does include additional entries from Northern and Far-Eastern sub-dialects that are noted when relevant to time terminology (Caughley 2000).

In keeping with related languages, Chepang attests an extensive range of sequential time-keeping through days and years. This system extends eight units into the past and eight into the future: the largest span in our survey. These constructs occur with respective numerals or with other modifiers as affixes. 

For the most part, modifiers precede the temporal head of the phrase. Chepang -nəm means ‘day’ as a time reference and may derive from the PTB *nəy of the same meaning (Caughley 2000, 157). Moreover, Chepang-nəm is a flexible noun; it can pattern with syaŋh to produce the temporal meaning of ‘[on the] following day’ or with other modifiers to provide the general meaning ‘X day’ or ‘the day of X:’

Nevertheless, the morphemes that precede -nem in the paradigm shown in Table 26 are still worthy of scrutiny. The numeric modifiers present in the constructions do not appear to be related to their general, countable 

forms. For example, while pləy.(jyoʔ) glosses as ‘four,’ there is little visible connection between this morpheme and the morphemes attested in sequential terminology that relate to ‘four’ (Caughley 2000, 367). Similar lexical discrepancies arise with other sequential terms. In this way, the Chepang paradigm is similar to the Kulung sequential system in that the numerals present in these compounds do not outwardly resemble their typical nominal surface representations. More research is needed to understand these Chepang terms and their own internal construction, in particular the underlying forms of numeric modifiers.

For phrases meaning ‘≥ 2 year(s) ago,’ numeral modification appears through suffixation rather than through prefixation. The function of the morpheme -koʔ, which precedes the numeral, is unclear. Similarly, the presence of yat is intriguing. Chepang yat glosses as ‘one’, a numeral, with alternative forms of ʔat and ya(ʔ)– (Caughley 2000, 217), and it modifies other time terms, as in the compound yat lah ‘(one) month’ and yat.ʔaŋ, an archaic form for ‘year’ (see Hodgson 1848; as cited in Caughley 2000, 217).

Regarding general lexemes for temporal units, Chepang boasts various Tibeto-Burman cognates as well as many Indo-Aryan loans. While -nəm is used for ‘day’ in phrases that reference time, din glosses as ‘day’ as a period or unit and ten as ‘today’ (Caughley 2000, 539-540). While -nəm is related to other day-related terms such as nāt ‘day’ in Sunwar (Borchers 2008, 91) and ni and nani in Dhimal (King 2009, 89, 566), din is indisputably a loan from Nepali. While ‘month’ is documented as mə.hi.na (also a loan from Nepali mahinā), lah.həw means both ‘month’ and ‘moon’ (Caughley 2000, 217, 240, 539). In isolation, Chepang lah carries the meaning of ‘moon,’ and is cognate with other Tibeto-Burman terms that mean ‘month’ like lā in Dolakha Newar (Genetti 2009, 69). To generate this meaning, a numeral must precede it: 

Chepang lah is also used to describe the phases and characteristics of the moon, i.e., toko.toyh.lah ‘full moon’ and si.lah ‘dark moon’ (Caughley 2000, 240). Chepang uses a twelve-month calendar for the names of months and follows the Hindu system in terms of segmentation. Some loans from Sanskrit, also attested in Nepali, such 

as pha.gun for approximately ‘mid-February to mid-March,’ are found in Chepang (Caughley 2000, 540). The Chepang term for ‘year’ is bər.sə (Caughley 2000, 540), most likely a loan from Nepali barṣa ‘year,’ which is also found in Dhimal as bare, a suffix for ‘year.’ Additionally, in Chepang, there is also a term for the ‘present year’ nek (Caughley 2000, 540). In general, Indo-Aryan loans are well-documented in the temporal lexicon of Chepang and are also found in terms for days of the week. Smaller increments of time, such as ‘morning,’ ‘afternoon,’ and ‘night’ are also documented, and increments of a day may be further subdivided into specific temporal periods.

Many Chepang lexemes for elements from dawn through to morning exist. The morpheme nyam is polysemous, can be glossed as ‘sun, sunlight, storm-cloud, weather’ depending on where it occurs, and resurfaces in many morning-related terms (Caughley 2000, 164). Kulung <-nam> carries virtually the same meaning.

Chepang terms for other parts of the day also generally discriminate between early and late periods. Chepang kaʔ.syurh and hiʔ.diŋ both gloss as ‘early afternoon’ while yo.ha.re surfaces as ‘mid-afternoon’ (Caughley 2000, 539). Compared to morning terminology, evening terms in Chepang are less specific, but many still do carry detailed and metaphoric meaning, as shown in Table 28 below.

The Chepang morpheme yaʔ or yah relates to darkness or night, albeit in a metaphorical manner. For example, yah.ram.diŋ carries an additional meaning of ‘spirit of the evening’ (Caughley 2000, 219). Chepang yah– is also an intransitive verb that means ‘climb up (like a vine)’ or also ‘move up ([like] shadows of [a] setting sun)’ (Caughley 2000, 218). Nevertheless, this sub-genre of terms deserves further research, as some relevant compounds like yaʔ.diŋ.ro have uncertain morphologies.

Some Chepang time terminology interacts with locative meanings and by extension, specifies locations related to time periods. Examples include how the sun affects certain terrain, as in nyam yah ‘to benight [as in become night], set (sun) on someone, end (day), move up (edge of sun on hillside)’ (Caughley 2000, 218). In such lexemes, the broader interconnectedness between time and space becomes clearer. Such metaphoric and

extended meanings move into more complex territory in the example of yah.yam: both a time of darkness and a location in the underworld (Caughley 2000, 219). In sum, these examples as well as the numerous Chepang temporal increments reveal how time is both measured and understood as something that remains uncountable and is rooted in human imagination and cultural experience.

Discussion

This comparative temporal survey highlights multiple lexical and morphological trends evident in related Tibeto-Burman languages, albeit languages that are oftentimes genetically and geographically distant from one other. In many documented cases, the terms for ‘day’ and even ‘month’ are common across the family and appear to be reflexes of well-attested Tibeto-Burman forms. Most unassimilated Tibeto-Burman words for ‘day’ are reflexes of PTB *nəy, which also carry the meaning ‘sun.’

This tendency, however, is by no means uniform: Kham in particular attests multiple suffixes for ‘day’ that do not appear to be reflexes of this PTB form. ‘Month’ often manifests as la (Darma), lā (Dolakha Newar), or in a modified form, as in Gyarong tsəla. These terms consistently pattern with the word for ‘moon’ in these languages. In some cases, as in Dhimal, there appears to be no Indigenous terms for ‘month.’ The connection between the moon and the term for ‘month’ is unsurprising given our natural observations of lunar cycles and phases as a tool for analysing the passing of time, particularly in a region of the world where the lunar calendar remains so culturally important.

By contrast, there does not appear to be a singular term for ‘year’ within the data presented and analysed in this contribution. Wambule, Jero, Dhimal and Dolakha Newar stray quite far from PTB *s-niŋ ‘year,’ whereas languages like Ersu and Gyarong incorporate considerable amounts of Chinese lexicon with the result that original, Indigenous terms may have been replaced in the process. Strong cultural and religious influences may account for the presence of Indo-Aryan lexicon such as sāl(Nepali or Hindi) ‘year’ in Kham and Darma discourse alongside various Chinese loans in Gyarong and Ersu, respectively.

Despite noted differences, there are still many general similarities between Tibeto-Burman languages in terms of how they lexicalise and grammaticalise time. For example, the presence of temporal constructs to address future and past scopes (e.g. ‘hence’ in English) is common across Lepcha, Jero, Wambule, Ersu, Kulung, Chepang, and Thangmi, to mention but a few. There are variations and apparent gaps between and within some paradigms, and we can only hope that further research will explore these differences and absences, and perhaps shed light on how they may have arisen.

Compounding is another general trend in lexicalisation, as exemplified by terms for ‘this year’ that combine the temporal adverb ‘now’ with the lexeme for ‘year’ in a number of languages under review. Wambule and Jero construct ‘this year’ in such a manner, and this pattern is also attested in Dhimal and Darma using different lexical sources. A modern reflex of PTB *asniŋ ‘this year’ is attested in Kham as aĩhsi (or related forms depending on the dialect). The existence of such compounding lexicalisation is interesting and represents an extended and deictic usage of scope. ‘Now’ expands to encompass not the specific time referenced (e.g. ‘the here and now’ so to speak, or even ‘the current era’), but rather a clearly segmented period of the present. In such cases, segmental time—our focus—and referential time intersect to generate a new semantic concept that is at once specific and broad, and that can reference both the near-past and near-future (e.g. a full year). Such a handling reminds us of Sunwar adisā, a term that simply references an unspecified time in the near-future that is distinct from ‘tomorrow.’

The most salient, unifying observation of note is the general morphological distinction that exists between the distribution of time constructs in Tibeto-Burman and Indo-Aryan languages. Where Indo-Aryan languages tend to nominalise time, Tibeto-Burman languages grammaticalise time through adverbial constructions. Even in smaller temporal inventories such as Kham and Sunwar, time is largely encoded through adverbs, as it is in Thangmi, Darma, Wambule, and Jero. In Chepang, there are sometimes even synonyms or related terms that function in different grammatical environments, as in din ‘day (as a period),’ a loan from Nepali, and <-nəm> ‘day (as time reference and modifier).’ Other common lexemes of time manifest as classifiers. In a purely analytical sense, the use of classifiers is unsurprising, as time is inherently susceptible to measurement.

We must recognise that Tibeto-Burman temporal adverbs and classifiers and Indo-Aryan nominalisations do not operate independently from one another. There is a great deal of intermingling between Tibeto-Burman and Indo-Aryan words, and the dual usage of both Indigenous and foreign constructs is a promising area for future inquiry for those interested in code-switching and the incorporation of highly specified time constructs (e.g.

days of the week, hours, minutes) in everyday speech. With the encroachment of Indo-Aryan time-keeping concepts as a byproduct of industrial expansion and national education, it is possible—indeed even likely—that such forms will over time become ever more frequently observed and further lexicalised in Tibeto-Burman languages, in the manner already noted in the lexicons of Qiangic languages that have undergone extensive cultural assimilation to Chinese.

Given the endangered and precarious state of many of the Indigenous languages covered in this review, we welcome more research to be directed towards the unique and somewhat idiosyncratic calendrical systems attested in the Greater Himalayan region as well as toward documenting time constructions in vernacular use and naturally occurring everyday dialogue.

Conclusion

The documentation of historically-marginalised and increasingly endangered languages of the Greater Himalayan region, is one—but only one—step towards their revitalisation and reclamation. Languages that differ in temporal expressions and attest unique monthly and seasonal paradigms—we think here in particular of Kulung, Darma, and to a lesser extent Ersu—can only be studied through respectful and collaborative partnerships with speakers of these languages to better understand the use of temporal concepts, how they function, how they are measured and what they mean for the speakers themselves. Speaking of time, the current moment may be the only opportunity that we have as a scholarly community to work together with speakers of Indigenous languages to collaboratively document, protect, preserve and strengthen these distinct and unique linguistic expressions in support of community goals for sovereignty and self-determination.

Endnotes

[1] We are grateful to the anonymous peer reviewers whose generous feedback has strengthened our article in both structure and direction, and to the University of British Columbia and its many libraries, which when taken together, have provided both funding for this project and made accessible many of the resources that we have consulted. We would also like to extend our gratitude to JALA’s Head of Communications, Nhan Huynh, for her ongoing and open communication with us through this entire process. This article was written on traditional, ancestral and unceded territory of the hən̓q̓əmin̓əm-speaking xʷməθkʷəy̓əm (Musqueam) people.
[2]  The underlining shown here is present in the original transcription and denotes the use of ‘state-setting’ or episodic topics in storytelling. These elements are actually nominalised verbs.
[3]  For this example and for a few others, we are not able to provide a three-tiered gloss or transcription (including morphological breakdown). In examples where morphological transcriptions are absent, this is because the authors of the sources we consulted did not provide them, and we do not have the language-specific expertise to introduce them without error.
[4]  Story told to an audience by Sanu Laxmi Joshi in 1989.
[5]  Zhang does not indicate morphological glosses for these terms. Since we are not in a position to offer an informed segmental morphological analysis of these entries, these spaces are left blank.

[6]  The symbol [ɿ] is an apical vowel used in Sinitic linguistics, but not recognised in IPA. It is rendered unchanged in this paper to reflect the original source document from which we have drawn our data for comparative analysis.
[7]  tə is a nominal contrast marker in Gyarong that alternates with ki (indefiniteness marker) (Prins, 2016, pp. 136, 734). Furthermore, only tə can co-occur with numerals or demonstratives. Gyarong tə also contrasts with ta, the latter in this case denoting a nominal that is more distant metaphysically or less frequently present in one’s life, while tə relates to oneself and indicates closeness (Prins 2016, 142). Both morphemes appear prefixed on temporal (and spatial) noun phrases.
[8]  Chepang jyoʔ is a numeral classifier that appears with other numbers (Caughley 2000, 110).

References