I. Introduction

The Nepali language is descended from Indo-Aryan, which is in turn descended from Proto-Indo-European. Nepali is spoken by over eleven million people in Nepal alone. It is spoken by an additional 156,000 people in Bhutan and six million people in India (Ethnologue). It is the national language of Nepal, where it is the first language of approximately 58% of the population (Acharya, 1). Since it is the “exclusive language of government, administration, primary and secondary education, the radio and official press in Nepal” (Slusser, 394), the vast majority of residents of Nepal who speak some other language as their mother tongue also speak Nepali as a second language (or lingua franca). Worldwide, the Nepali language is spoken by over seventeen million people (Ethnologue).

Nepali was born in the Kathmandu Valley (the region we today call Nepal) during the time the area was becoming politically united and residents of the region were being driven to communicate more than ever before. At the time, the area was inhabited by speakers of many different Indo-Aryan dialects, as well as a relatively small population of speakers of Tibeto-Burman languages. Information regarding the specific dialects and languages spoken in the area at the time of the birth of Nepali is very difficult (if not impossible) to find. However, it is known that among the Indo-Aryan dialects in use in the area at the time were Sanskrit, Prakrits, and Maithili (Slusser, 392).

II. Historical Timeline of the Language

It is commonly estimated that Proto-Indo-European (PIE) was spoken approximately 6,000 years ago in Eastern Europe (Trask, Historical Linguistics, 181). Around 2000 B.C., as speakers of PIE immigrated into southern Asia, the Indo-Aryan language was born. As Indo-Aryan speakers continued migrating and settled into different areas of southern Asia, the Indo-Aryan language diversified into a large number of dialects. It is likely that Nepali was born from a mixing of several Indo-Aryan dialects, but it is not possible to know how many or which dialects contributed directly to the birth of Nepali, or if one dialect in particular formed the base of the Nepali language.

Sanskrit was originally brought into the Kathmandu Valley by a group of people who migrated to the area from northern India. It is first attested in the region at about 250 B.C. (Malla, 101). At the time, Sanskrit was imposed as the prestige language of religion and scholarly pursuits in the region, and it remained such until the 18th century (Slusser, 392). Prakrits is believed to have been brought into the valley by Buddhist missionaries around the first century A.D. It is likely that Prakrits was used as the spoken language of Buddhist missionaries in the area and spread throughout the Kathmandu Valley as a large portion of the population converted to Buddhism (Malla, 102). Once the Buddhist religion gained a firm social standing in the area, Sanskrit continued to be used by the upper-classes while Prakrits was used by uneducated population. The Maithili language was brought into the Kathmandu Valley sometime before the 12th century A.D. It continues to be spoken in Nepal today.

From about 300 B.C. through the fourteenth century A.D., it is quite difficult to trace the history of both the people and the language of the Kathmandu Valley due to a lack of texts treating these subjects. We can assume, however, that throughout the Middle Ages the political and cultural milieu in Nepal included influence from both India and Tibet (Slusser, 42). During this period, various Indo-Aryan tribes moved into the region and established independent kingdoms that lasted for various amounts of time. People living in the Kathmandu Valley continued to use a wide variety of Indo-Aryan dialects, as well as some Tibeto-Burman dialects, while the ruling classes of the region used Sanskrit for scholarly, religious and academic purposes. Sometime between the 14th and the 17th century, as the region was becoming politically united, several of these Indo-Aryan dialects merged to form the language that is today called Nepali (ibid, 74).

In its earliest stages, Nepali was known as Khas Khura. In the late eighteenth century, the king of a tribe called Gorkha took it upon himself to politically unite the Kathmandu region. During this time, people began to refer to the Khas Khura language as Gorkhali. Shortly after the Gorkha conquest of the Kathmandu Valley region (in 1769), the language first began to be used for scholarly and administrative purposes (Malla, 111). However, because it was not until 1913 that the first publishing house which printed in Gorkhali was established (ibid.), texts which document the language in its earliest states are very difficult to find. It was during the 1920s (after the UK officially recognized Nepal’s independence) that people from outside Nepal began to refer to the language spoken by the people of Nepal as Nepali.

II. Aspects of the language to be investigated

Nepali is an inflectional language which uses a system of case markers to convey the grammatical function of noun phrases in a sentence. An analysis of the distribution of case markers in Nepali reveals an unexpected split in which transitive verbs seem to assign case to their subjects differently than do intransitive verbs. In an intransitive verb construction, the subject of the sentence (whether it is the semantic agent or experiencer) is realized in the Nominative case, where case is reflected by the absence of an overt marking. In a transitive verb construction, the agent/experiencer (i.e. subject) of the sentence is realized with the instrumental case marker -le, while the complement (i.e. direct object) of the verb is reflected through the use of the accusative case marker, -lai.

The use of the instrumental marker with the subject of a transitive verb is particularly surprising, as it seems to be out of line with both the syntactic and semantic properties assigned to the noun phrase that takes it. Examples of basic Nepali sentences follow:

     

Intransitive:	Subhadra	roin
	Subhadra-NOM	cried-3rdSing
	“Subhadra cried”
	Jane	nachchha
	Jane-NOM	dances-3rdSing
	“Jane dances”
Transitive:	Smriti-le	Jane-lai	dekhchhe
	Smriti-INSTR	Jane-ACC	sees-3rdSing
	“Smriti sees Jane”
	Smriti-le	Jane-lai	angalo halchhe
	Smriti- INSTR	Jane-ACC	hugs-3rdSing
	“Smriti hugs Jane”

A close analysis of the behavior of Nepali constructions containing a subject marked with the instrumental case reveals some clues as to how this split may have developed. The first clue that must be considered is the simple fact that the use of the instrumental marker with a subject only occurs with transitive verb constructions. The second clue is that it is the instrumental case marker in particular that shows an unexpected distribution. The third clue is that the distribution of the instrumental case marker extends to the semantic agent/experiencer of a transitive verb construction. When considered jointly, these facts point to a comparison between the Nepali transitive construction and the passive voice construction outlined in traditional grammars, since passive constructions are restricted to those containing a transitive verb, and in many languages, the agent or experiencer of a passive construction is realized in the instrumental case.

What can be made of these similarities? A diachronic analysis of the Nepali language allows us to see that the distribution of the instrumental case marker in Nepali is a likely remnant of an earlier ergative construction which had developed out of a still earlier passive voice construction. An analysis of this type raises several questions. First, if the case marking system of Nepali developed out of sytactic ergativity, why is it that verb inflection in Nepali constructions containing a transitive verb does not agree with the verbal complement (as it does in both passive and ergative constructions)? Second, should the case marking system of Nepali be expected to change over time, or does it appear stable?