The Language of the Modhupur Mandi (Garo), Volume 1

ONE:The Language and the People

The Garo Hills form the western two-fifths of Meghalaya, one of the so-called "hill states" of northeastern India. Outsiders have always called the people who live in these hills "Garos", and their language has always been known as "Garo". The Garos are often included among the "hill people" of northeastern India, but Garos are also found in the low country that lies to the north, west, and south of the hills. The plains area to the south now lies in Bangladesh and it is the dialect of Garo that is spoken in Bangladesh that is the special focus of this book. More than a half million Garos live in India. The majority are found in the Garo Hills but many live in the neighboring districts as well. Well over one hundred thousand live over the border, in Bangladesh.

The word "Garo" is not often used by the people to whom it refers. It is used, instead, by other Indians, Bangladeshis, and foreigners, and for many of the people themselves, the word has acquired somewhat unpleasant connotations. Those living in Bangladesh generally prefer to call themselves "Mandis", which otherwise simply means 'people, human beings', and I will follow their preference here by referring to Bangladeshi Garos as "Mandis", and to their dialect, as "Mandi". This is the local pronunciation of the word that is pronounced "Mande" in the more northern dialects of the language. The difference in pronunciation gives me a convenient way to label the particular dialect with which I am especially concerned. The dialect of Bangladesh, to be sure, grades imperceptibly into the neighboring dialects of India, but so much of my own recent experience has been in Bangladesh, that I need a way to refer to the particular variety of the language that is spoken there.

The Mandis who live in Bangladesh are well aware of the differences between their own dialect and that of the hills to the north, but their ownPage  10use of the word "Mandi" includes everyone whom others call "Garo". They also refer to their own form of the language as "A'beng" and they recognize that it has some affinity with the A'beng dialects spoken in the western part of the Garo Hills. Beyond A'beng, however, they also recognize the special status of the dialect upon which written Garo is based and which has become a sort of standard dialect. Only rarely do people in Bangladesh try to model their own speech upon the northern dialect but they do sometimes suggest that it is in some degree "better" than their own local dialect. They are often able to provide a few stereotypic examples of how their dialect differs from the northern standard. Bangladeshi Mandis refer to this northern standard dialect as "A'chik". A'chik means, literally, 'hill' or 'slope', and to describe people as "A'chik" implies that they are hill people. The Mandis, who live in the flat plains of Bangladesh, do not feel that they are, in any sense, hill people, and they do not feel that "A'chik" can refer to them. Thus they use "A'chik" to refer specifically to Garos other than themselves, and to refer to the form of the language that they perceive to be dominant in the Garo Hills.

For Garos who actually live in the hills, "A'chik" has a wider scope, for it is used to refer to all the people whom others call "Garos", even those who live in the plains. I need a word that can refer to the language of the hills, however, and in particular I need to be able to refer to the dialect that has become, more or less, the standard. Repeatedly, I will need to contrast this northern ("standard") dialect with the dialect of the Bangladeshi Mandis. It will be convenient to follow the usage of Mandis of Bangladesh and refer to the standard dialect as "A'chik", and to contrast it with the southern and more local "Mandi" dialect.

Unfortunately, no generally accepted term has yet emerged that can replace "Garo" as a general term for all the people traditionally called "Garos". Bangladeshi Mandis apply "A'chik" more narrowly than northerners do. Northerners sometimes use "Mande", the northern pronunciation of southern "Mandi", as an equivalent for "Garo", but this is by no means universally accepted. In spite of its negative associations, therefore, I will continue to use "Garo" as a cover term for the full range of mutually intelligible dialects, both in India and Bangladesh, that outsiders have traditionally called by that term. I will also use "Garo" to refer to the written form of the language that many of these people now use. "A'chik" and "Mandi", therefore, are to be understood as two spoken varieties of the "Garo" language.

While I will be concerned primarily with Mandi as spoken in Bangladesh, and secondarily with the relatively "standard" A'chik, it should not be imagined that Garo has only two dialects. Mandi speakers recognizePage  11their own dialect as a form of A'beng, but "A'beng" is also used for the dialects that are spoken in much of the western part of the Garo Hills, although these are not exactly the same as the dialect used in Bangladesh. The dialect of the northern part of the district is sometimes called A'we, and the standardized A'chik derives especially from eastern A'we. Chisak is found in the northeast just south of A'we. Matchi and Dual are spoken in the east-central portion of the hills and Ganching (also known as Gara) and Chibok are spoken in the south. I will have nothing to say about most of these dialects. All of them are mutually intelligible, although speakers who are unfamiliar with a dialect from an area distant from their own may need some patience and an occasional explanation in order to understand. All these dialects grade into one another without sharp borders. People listen to other dialects with curiosity and amusement but they do not correspond to deep divisions within the larger Garo community. The research that would tell us exactly how they differ from each other has yet to be carried out.

All of them are dialects of the Garo language, and most people who consider themselves to be ethnically Garo speak one of them. A few people who are accepted as Garos, however, speak distinct languages. The most important of these people are the A'tong, whose villages are found along the Simsang (Someswari) River in the southeastern corner of the Garo Hills. The A'tong consider themselves to be Garos and they are accepted as such by other Garos, but they speak a language that is different from and not mutually intelligible with Garo. Indeed, the language of the A'tongs is closer to the languages called "Koch" and "Rabha" than it is to Garo. Ethnic classification and language classification simply fail to correspond in this case. Outsiders sometimes find it confusing to be told that some Garos speak Garo while others speak A'tong, but this should be no more mysterious than to recognize that some Irish people speak Irish while others speak only English. Garo and A'tong are enough alike to allow A'tong speakers to learn Garo fairly easily, and many A'tong are bilingual. I have been told that the A'tong language is now losing ground to the numerically dominant Garo.

The language known as "Ruga" is, or was, spoken in a small area in the south central part of the Garo Hills. Like A'tong, Ruga is closer to the Koch and Rabha languages, and also to A'tong, than to the language of most Garos, but the shift to Garo has gone further among the Rugas than among the A'tongs. Indeed the Ruga language now appears to be moribund. A few older people can still speak what they consider to be Ruga, but the speech of those I have met is heavily influenced by Garo, and the remaining speakers admit that they are more comfortable with Garo than with Ruga. They lack the fluency that they remember in their parents. Their children,Page  12they tell me, speak only Garo, though they may understand the Ruga of the older people. I doubt if anyone born after 1950 would claim to be able to speak Ruga.

Finally there is a group known as Megam to the Garos and as Lyngngnam to the Khasis. These people live in the western-most part of the Khasi hills, just to the east of the Garo Hills, and their culture is enough like that of the Garos that they are sometimes considered to be a subgroup of Garos. Their language, however, is similar to Khasi and thus a member of the Austro-Asiatic language family. It is utterly unrelated to Garo.

The Garo language belongs to a subgroup of the Tibeto-Burman family of languages known as the "Bodo" group or sometimes as "Barish, Bodo-Garo, or Bodo-Koch". In addition to Garo, these Bodo languages are spoken in pockets up and down the Brahmaputra valley in Assam, in the Cachar hills to the southeast, and as far south as the state of Tripura. The non-Garo languages of this Bodo group are themselves divided into two subgroups, as different from each other as they are from Garo. One of these subgroups includes languages known at various times and in various places by such names as Boro, Mech, Dimasa, Kachari, and Hill Kachari. Some of these various forms of speech may be sufficiently distinct from one another to count as separate languages, although some of their speakers now prefer to call themselves "Boro" rather than "Mech", "Dimasa", or "Kachari". Tiwa (formerly known as "Lalung"), also in Assam, and Kokborok (formerly "Tipra" or "Tripuri"), which is spoken in Tripura state to the south, are also closely related to Boro.

The languages of other main Bodo subgroup are sometimes known as the "Koch" languages. They include Rabha, spoken north of the Garo Hills and as far west as the northern part of West Bengal, several dialects or languages variously known as Tintinkiya Koch, Wa'nang Koch, or Pani Koch that are spoken to the west and north of the Garo Hills, and probably on the southern fringes as well. It is to this sub-group that the A'tong and Ruga languages belong. Finally, one other isolated language called "Deori" (not to be confused with the unrelated "Chutiya") which is spoken further to the east in the upper part of the valley of Assam, also belongs to the wider Bodo group. When speakers of these Bodo languages encounter one another, they easily recognize the similarity of their speech, but few, if any, of them are mutually intelligible. Perhaps they differ from one another to about the same degree as the Romance languages of Europe.

The Bodo languages, in turn, are related to a group of languages spoken in the northeastern part of Nagaland and in the adjacent southeastern corner of the state of Arunachal Pradesh. From north to south, these languages include Tangsa, Nocte, Wancho, Phom, Konyak, Chang, andPage  13Kiamungan. Konyak probably has the largest number of speakers of any of these northeastern "Naga" languages. Other closely related languages are spoken across the border in Myanmar. All these languages form one subgroup of Tibeto-Burman. Almost all of the languages spoken in the hills of northeastern India belong to the Tibeto-Burman family, but Khasi is an outstanding exception. To recognize languages as Tibeto-Burman is to state that they are distantly related to Tibetan, Burmese and to many other languages of Nepal, Myanmar, and parts of Southwest China. Finally, the Tibeto-Burman languages are even more distantly related to Chinese, and together they form the Sino-Tibetan language family.

The written form of Garo was developed by American Baptist missionaries during the final decades of the 19th century. These missionaries' first contact with Garos was in the northeastern corner of the Garo Hills, so the dialect of that area was used as the basis for the written language. This was also the area where schooling began and where the first substantial number of educated and literate Garos lived, and it was this historical accident that has made the northeastern-most dialect of Garo so influential for the speech of educated Garos. When the government of British India completed the occupation of the Garo Hills in 1873, the political headquarters was established in the west-central part of the Garo Hills, in a town that came to be called "Tura". A number of educated Garos from the northeast moved to Tura where they formed the nucleus of a growing body of educated speakers, and their dialect became established in Tura as well as in the northeast. A hundred years later Tura people still regarded the dialect of the surrounding villages as distinctly rustic. The northeastern dialect, or the standard that has grown out of it, is the subject of virtually all earlier descriptions of the Garo language, including my own.

Brief word lists of the language were collected in about 1800 by British officials (Eliot 1794, Hamilton 1940 [1820]) but almost a century went by before more American Baptist Missionaries began to produce the first real dictionaries and short grammars. They first used the Bengali alphabet but early in the 20th century they switched to Roman, and it is with the Roman alphabet that Garos have written ever since. The bible, hymn books and other Christian publications are widely available in the Garo Hills, though somewhat less so in Bangladesh. The language has been used as the medium of elementary education in the Garo Hills for many decades, and school books covering the entire range of elementary school subjects are prominent among Garo publications. Collections of folk tales, and a scattering of other publications in Garo have appeared from time to time, and two or three thin weekly newspapers appear with some regularity. People in the Garo Hills easily keep records and write letters in their language, but if anyone could be found who could read Garo as easily as many Westerners readPage  14their own languages he or she could surely read the entire corpus of Garo printed literature within a few months.

The position of the written language is quite different in Bangladesh. The largest number of the 100,000 or so Mandis who live in Bangladeshi are found in the districts of Jamalpur, Mymensingh, and Netrakona, most of them within 10 or 15 kilometers of the Indian border just south of the Garo Hills. Another, though smaller concentration is found a bit further south in what only a few years ago could legitimately be described as the Modhupur forest, half way between the cities of Tangail and Mymensingh. The forest is a remnant of a larger forested area that once stretched so far to the southwest that it reached to within 20 or 30 miles of Dhaka, and it seems likely that Mandis were once found throughout much of this forested area, a considerably larger area than they now occupy. A few Mandis still live around Bhaluka, an area of slight elevation that lies about half way between Dhaka and the city of Mymensingh, and within recent memory Garos are reported to have lived in the neighborhood of Jaydebpur, only 25 miles north of Dhaka. The present work refers especially to the dialect of Modhupur, but I make some reference to the dialect of the border areas as well, and I point out some of the differences between A'chik, as spoken in the Garo Hills, and the Mandi dialects of Bangladesh.

Even in Bangladesh, Mandi is by no means uniform, but all its dialects are easily mutually intelligible, and Mandis move about freely enough to give everyone experience with a variety of dialects. Though they are most familiar with the Bangladeshi dialects, many people have visited the Garo Hills in India and have become familiar with northern dialects as well. Even those who have never traveled north have met occasional Northerners on visits to Bangladesh, so everyone has some awareness of the full range of Garo dialects. People move freely about the Garo Hills, and a considerable number of Garo refugees from Bangladesh have settled in the Garo Hills, so that people in the hills have also had wide exposure to other dialects.

Elementary schools in Bangladesh, even those located in Mandi villages, use Bengali as the medium of instruction, and as a result, Bangladeshi Mandis are more often literate in Bengali than in Garo. Many people study enough English to be able to associate speech sounds with the letters, and this is enough to allow them to sound out written Garo, though few Bangladeshi Mandis do so easily. Every Mandi in Bangladesh, whether literate or not, is well aware that the language has a written form that is widely used in India. Except for a few Bibles and hymn books, however, printed Garo rarely reaches the Mandis of Bangladesh.

The written language was originally based upon a dialect that is spoken far from Bangladesh. This dialect is as different from the Mandi dialects asPage  15is any dialect of Garo, and it strikes Bangladeshi Mandis as a bit peculiar. The distance between Mandi and written Garo may have helped to make Bengali acceptable as a written language to the Mandis, and given them less incentive to push for using written Garo, and in fact there has been little interest among Mandis in promoting their own language as a substitute for Bengali in the schools. However, the minority status of the Mandi in Bangladesh and the overwhelming dominance of Bengali might well have been enough to assure the dominance of written Bengali even if written Garo were ideally suited to the local dialect.

In addition to other dialects of Garo, Mandi speakers have extensive experience with other spoken languages, especially Bengali and English. All Bangladeshi Mandis whom I have met are at least partially bilingual in Bengali and some are fluent. A very few Mandis have learned Bengali so well that it has become their dominant language, but for most, Mandi remains the first and most comfortable way of talking. It is the language of intimacy and of family life. Still, much that is new in the world comes to the Mandis through the medium of Bengali, and Mandis feel absolutely free to incorporate any Bengali word that they feel will be understood into their Mandi sentences. As a result, Bengali words flood into Mandi. These new words affect every corner of the language.

For those Mandis who receive sufficient education, English also becomes a significant influence, and English words join those from Bengali in invading Mandi. The prestige of English is even greater than that of Bengali, and those who are capable of doing so like to sprinkle their language with English words and phrases especially when speaking in relatively formal situations. As with Bengali words, Mandi speakers feel free to use any English word that they think others will understand, but since the level of skill in English is much lower and less widespread than skill in Bengali, the direct impact of English is less. Many English words come to Mandi indirectly through Bengali, however, and English must be placed beside Bengali, when considering the present linguistic environment of the Mandis, and when considering the changes that their language is undergoing.

Garos living in the Garo Hills are less often bilingual in any language than are those who live in Bangladesh, and no Indian language has as powerful an influence on Garo in India as does Bengali on the Mandi of Bangladesh, although even the dialects of the most remote corner of the Garo Hills have plenty of borrowed words. English is used as the medium of education beyond the elementary level in India, and even some elementary education is conducted in English. This means that English has a somewhat stronger role than it does in Bangladesh, but only a minority of Garos use English with any ease, and many remain effectively monolingual. ThePage  16impact of both Bengali and English can only grow in the decades to come.