Introduction.

FEW early notices of the people of Khandesh have been traced. According to Lassen the Sanskrit-speaking tribes passed down the Gujarat coast and up the Tapti valley through Khandesh into the  Deccan. [Indische Alterthumskunde, I. 181.] Another wave of settlers would seem to have entered by the east, as, according to local tradition, Rajputs of different tribes ruled from Asirgad as far back as the sixteenth century before Christ.

The first known historic reference to the people of Khandesh [The Mahabharat (H. H. Wilson's. Works, VII. 164) places, next to the people of Vidarbha or Bedar, Khandas or Shandas who may possibly have given their name to Khandesh. Khandesh has also been thought to be the Khandav forest of the Mahabharat, which was burnt down and brought under tillage by Arjun the brothel of Krishna. These identifications are doubtful.] seems to be Ptolemy's (A.D. 150) mention of the Phyllitae and Kondali or Gondali, probably the Bhils and Gonds, whom he places south of the Narbada not far from its source. [Bertius' Ptolemy, Map X. and 204. Wilson (VII, 139) is of opinion that the Bhils are included under the Pulindae whom Ptolemy places further to the west. Another of Ptolemy's tribes the Tabassi have been referred to Khandesh and supposed to be the Buddhist ascetics of the Ajanta and other Satmala cave temples. Yule in Ind. Ant. IV. 282.] The Bhils, still the most characteristic and one of the largest classes in Khandesh, seem to show by the varieties of their dialect, Nemadi, Marathi, and Gujarati, that they have been pushed back into Khandesh by later arrivals, from the east, the south, and the west. Many of the changes that have narrowed the limits of the Bhil country have taken place since Ptolemy wrote. But in his time, as at present, Khandesh was probably one of the leading Bhil settlements. Of the Gonds traces remain in a Gond tribe of herdsmen found in Chalisgaon and in a Gond sub-division of Mhars.

Since Ptolemy's time, the first great change -in the population of Khandesh seems to have been the arrival, apparently up the Tapti valley from Gujarat, of a detachment of the great tribe or nation of Ahirs or Abhirs. [Some of the Ahirs, apparently later arrivals, came from northern India.] The origin- of the Ahirs, who, besides in Khandesh, are found in the North-West Provinces, Bengal, Central India and the Central Provinces, in Cutch and Kathiawar in Gujarat, and in Nasik, Ahmednagar and other parts of the north Deccan, is doubtful. They have been thought to be the Abars, one of the Skythian tribes who, in the second and first centuries before Christ, entered India from the north-west,[ Cunningham's Archaeological Report, II. 23,33.] or, and this is more likely, they are supposed to be an old Indian or half Indian race who were driven south and east before the Skythian invaders. [Compare V. de St. Martin, Geog. Grec. et Latine de 1'Inde, 230; Cent. Prov. Gaz. LXIII.] In either case the bulk of the nation seems to have passed south during the time of Inde-Skythian ascendancy (B.C. 200 - A.D. 200) in north India. Before the Christian era, they were near the north-west frontier;[ V. de St. Martin, Geog. Grec. et Lat. de 1'Inde, 230.] in the second century after Christ they were in Upper Sind; [Bertius' Ptolemy (A.D. 150), Map X.] and in the third century in Lower Sind and north Gujarat. [McCruidle's Periplus (A.D. 247), 113. The expression is ' Inland from Surastrene'] Next they appear south of the Tapti, ' between the Tapti and the Konkan,' or ' between the Tapti and Devgad. [ Purans quoted in Ward's Hindus, III. 450, and Wilford's As. Pes. VIII. 336.] They are spoken of as settled in Khandesh. [Langlois Harivansh, II. 401.] And an inscription in one of the Nasik Buddhist caves shows that early in the fifth century (419) the country was under an Ahir king.[ Second International Congress, 354.] The Ahir dynasty is said to have lasted for only sixty-seven years. But as local tradition centres in an early Ahir or Gauli rule, it seems probable that, as was the case in Kathiawar, the Khandesh Ahirs were closely connected with the Yadav's who were in power in the eighth, and again appear as the. rulers of Devgiri or Daulatabad in the twelfth and thirteenth centuries. [Details are given below in the " History" chapter.] Their present strength has not been ascertained. The 1872 census returns do not show any Ahirs. But a special inquiry, carried on by Mr. J. Pollen, C.S., has brought to light, besides Ahir or part Ahir husbandmen and Ahir Kolis, an Ahir branch in almost all crafts and other middle class castes, and proved them to be so large an element in the population, that Ahirani is the local name for the Khandesh dialect of Marathi. [Besides Ahirs proper and Abhir Brahmans, there are Ahir Sutars, Ahir Lohars, Ahir Shimpis, Ahir Salis, Ahir Guravs, and Ahir Kolis.] In some of these classes, as among the carpenters, blacksmiths, and goldsmiths, the Ahir element has remained distinct. Ahir carpenters and Ahir blacksmiths marry together, but neither of them marry with the other sub-divisions of carpenters and blacksmiths. In other cases the Ahir element has merged into the general class, and Ahir has come to be little more than a surname.

The next recorded addition to the population of Khandesh is the arrival, partly, direct up the Tapti valley, and partly it would seem through Malwa and the Central Provinces, of the great body of Gujarat Kunbis who now hold the bulk of the Cultivated land to the north of the Tapti. According to the chronicles of the Reve Kunbis, they arrived about the eleventh century in a. large body, in whose vanguard alone were 2000 carts. It seems not unlikely that this account is correct, and that the Kunbis were forced to leave Gujarat by the encroachments of Rajput tribes, driven south before the early Musalman invaders of north India.

About this time, and on till the arrival of the Musalmans (1310), it seems probable that while the Devgiri Yadavs held Khandesh, the southern castes, of whose arrival no tradition remains but who are still known to have come from the south, first settled in the district. [Rane or Maratha Rajputs.] During the latter part of the fourteenth century, by the establishment. (1370) of a line of Arab kings, a new foreign element was introduced into Khandesh. About the same time the district was visited by so severe a famine that, according to Ferishta, its whole people, except a few Bhils and Kolis, disappeared. [Two great Khandesh famines are reported, one about 1370, the other the great Durga-Devi famine from 1396 to 1407. As he makes no mention of the Durga-Devi famine, it seems probable that Ferishta's 1370 famine should be placed some thirty years later.] Under the Faruki kings, Khandesh rose to much wealth and prosperity, and its population received many additions both of neighbouring Hindus and of foreign Musalmans.

In 1600, when it passed to the Emperor Akbar, parts of Khandesh were highly tilled and well peopled, and its Bhils, Gonds, and Kunbis were specially noticed as hardworkers and dutiful subjects. [Gladwin's Ain-i-Akbari, II. 54.] Under the Moghals, during the seventeenth century, prosperity continued. The disturbances in the Deccan, that ended with the fall of Ahmednagar (1638) and Bijapur (I680), must have driven numbers into the more peaceful district of Khandesh, and during all his reign, Aurangzeb (1660-1707) was converting Hindus to Islam and Musalmans were Hocking from north India into Khandesh.

Under the Marathas (1760-1817) the Hindus again rose to importance. But any additions from the south must have been more than met by the losses in the disturbances that marked the close of the eighteenth century. The terrible famine of 1803 laid the district waste, leaving for the time but a few Bhils and Kolis. When the famine was over' some of the old inhabitants returned. But so disturbed were all, except some favoured parts of the district, that numbers still staid away. [The present special prosperity and. populousness of Savda is believed to be owing to the proteotion afforded at this time (1803-1818) by the power of the Nimbalkar and Rastia. Bom. Gov. Sel. XCIII.203.] In those troubled times three bodies of foreign mercenaries rose to importance; Arabs, north India or Pardeshi Brahmans, and Mysor or Karnatak troops, apparently partly Hindu partly Musalman. With the establishment of British power (1817 -1820) these three classes of mercenaries disappeared. Part of the Arabs were sent to Arabia, and the rest found their way to Haidarabad in the Deccan; almost all the Karnatak troops returned to their' own land; and most of the Pardeshi Brahmans settled as husbandmen. [Compare Capt. Briggs (1821) in MS. Sel. 157 (1821 -1829).] On the establishment of order, the old inhabitants returned in numbers from Berar and Gujarat, and crowds of strangers flocked from the Nizam's and Sindia's dominions. [Mr. Chaplin's Report, 20th August 1822; East India Papers, IV. 515.] Still the country was very empty. The 1821 census showed a population of only 418,021 souls or 31¼ to the square mile. In 1837 it was still ' miserably populated', large tracts being held by Bhils. The population was estimated at 478,457 souls, about 60,000 more than in 1821, and it was calculated that of the whole number, Brahmans represented 5.40 per cent, Rajputs 347, Shudras and Marathas 69 58, low and depressed tribes 14.72, and Musalmans 6.38.[Col. Sykes in Rept. of Brit. Assoc. (1837), 258, 264.] Fifteen years later (1852) a fresh census showed a total strength of 686,003 souls, or an increase, since 1837, of nearly forty percent. [Bom. Gov. Sel. I 4.] Still great part of the district was empty, and five-sixths of its arable land lay waste. An attempt was made to supply the want of people by bringing settlers from crowded Ratnagiri. But the most liberal offers of rent-free land and money to buy bullocks and tools, failed to tempt a single settler. [Bom. Gov. Sol. I. 13 k Mr. A. T. Davidson, Superintendent of Survey (1861). Bom. Gov. Sel. XCIII. 432.] Still population was steadily increasing, and with the rise of produce prices (1856), the introduction of a lighter and more even assessment (1860-1866), and the opening of the railway (1863), large numbers came to Khandesh. Compared with those of 1852, partly no doubt because they were more complete, the census returns for 1872 showed a total of 1,028,642 souls, or an increase in twenty years of nearly fifty per cent. Since 1872, though Khandesh has passed through several trying years, the population is known to have considerably increased. The east and centre are populous, but the south is thinly peopled, and in the west great arable tracts are still empty. In spite of recent increase, Khandesh remains one of the thinnest peopled parts of the Bombay Presidency.

Of the immigrants under British rule, exclusive of those who came back on the first establishment of order (1817-1820), the chief classes are, among traders, Marvadi Vanis, Bhatias, and Bohoras; among craftsmen, Rangaris. and Telis; and among husbandmen and labourers, Marathas and Kunbis.

At present the most interesting section of the population are the Vanjaris, the owners of pack-bullocks, who since cart roads have been opened to the coast (1834), and still more since they have had to compete with the railway, have been forced to give up their old wandering ways and settle in fixed villages. [Another older division -of Vanjaris is established as a regular part of the Kunbis. See below, p. 69.] Their leaders, men of capital who always did some trafficking on their own account, have started as traders, and the bulk of the people, who have probably always been in the habit of raising crops during their long halts in the rainy season, have taken to tillage. Some still earn their living as carriers. But none have taken to the practice of crafts, probably because, as they travelled with bullocks and with no carts, they had no need of the services of carpenters or blacksmiths.

This Vanjari and Ahir element in the Khandesh people has the high value of showing, what can hardly be traced in most parts of the country, the chief process by which the bulk of the present Hindu population was probably formed. A succession of tribes of northern herdsmen have been driven south, and by changes in the government or trade of the country, have been forced to settle. Among their leaders would be men of capital if not trained traders; the mass of the people would have a rough knowledge of. tillage; and as many tribes travelled with carts and sheep, some among them would be familiar with carpentry, smithwork, and weaving. When they settled the leaders would become traders and landholders, the bulk of the people would start as husbandmen, and of the rest, some would take to the practice of crafts and others would continue as herdsmen or carriers. Among the husbandmen, some of the poorer or lower sort, forced to take up lands in the outskirts of the settlement, cut off to some extent from their own people and straitened from want of labour, would associate with the earlier tribes, and taking their women as wives or slaves, would raise a mixed class. As the new tribe settled many of their special beliefs and practices would cease, peculiarities of dress would be given up, and if they had not them before, they would engage Brahmans to conduct their ceremonies. In time sameness of work would become a closer bond than a common origin. Husbandmen would begin to marry with the older settled husbandmen and other craftsmen with men of their own calling. [ This is the probable explanation of the large number of sub-divisions among the craftsmen classes in Khandesh. Many of them the Telis, Salis and Koshtis, Lobars, Rangaris, Nhavis, and Mhars have the stock number twelve and a half, the half being apparently a mixed or inferior class.] Each division would come to be known by the name of its calling, and the tribal title would sink to the name of a sub-division or to a surname. Of the old tribe the only apparent traces would be those who had kept to the original calling of herdsmen or carriers; the offspring of the low class women proud of their strain of higher blood; and perhaps the Brahmans, who known among their caste fellows by their, patrons' 'name, had come to form a distinct sub-division.

Language.

Gujarati is in use among the higher class husbandmen to the north of the Tapti, and it is the language of trade throughout the district; and Marathi, the speech of the people of the south and west, is the language of Government offices and schools, and is gradually gaining the ascendant.

Ahirani

But in their homes the bulk of the people speak a dialect known as Khandeshi, Ahirani, [The name Ahirani is probably derived from that of the Abhiri mentioned by the latest Prakrit grammarians as a dialect spoken by cowherds and others in the country near the western coast about Gujarat. Prof, R G. Bhandarkar, M.A., Hon. M.R.A.S.] or Dhed Gujari, a curious mixture of Gujarati, Marathi, Nemadi, and Hindustani. Though from its construction it looks like a compromise between the modern Gujarati,. Marathi, and Hindi, this dialect is the offspring of several old Prakrit varieties, Magadhi, Saurashtri, Shaurseni, Lati, Maharashtri, Paishachi, and Apabhransha or Bhakha. Except a few ballads and songs recorded by the Khandesh Bhats, this is entirely a spoken language. It discards the cerebral l and substitutes for it the palatal semivowel y ; thus black is kaya not kala. Resembling; Marathi and Gujarati in the more general grammatical forms, Ahirani presents several peculiarities in declension and conjugation. The plural is formed by adding the suffix s, not as in Gujarati and Marathi by a change in the word itself; thus bhit, wall, becomes bhitas, walls. [This s is peculiar. When the Sanskrit plural termination s was lost, some of the vernaculars were obliged to have recourse to a new mode of expression. Thus Bengali marks plurality by adding a noun expressive of 'all,' 'a collection,' or 'a class,' such as sarva, gana, and diga, and Uriya by affixing the word mana meaning measure or kind. The Ahirani's is therefore not unlikely the old Prakrit savva, Sanskrit sarva, and Gujarati sau. Prof. R. G. Bhandarkar, M.A., Hon. M.R.A.S.] Cases are formed by suffixes. Thus, for the accusative and dative le is added; for the instrumental, ni vari, ghai; for the ablative, thin; for the genitive, na, ni, na; and for the locative, ma and mazar. [The le of the accusative and dative is the Marathi la. Of the instrumental suffixes ni seems to be the Marathi ne, and the Sanskrit ena ; and vari and ghai are probably corruptions and abbreviations of the instrumental forms of the Sanskrit words vritti action and gati motion, which seem to have dwindled into case terminations. The ablative thin seems to be a corrupt form of the Sanskrit affix tas. The final becomes h in Prakrit, and, joined with the previous t, sounds like th. The genitive na and the locative ma are much like the corresponding Gujarati terminations, while the optional mazar of the locative is a form of the current Gujarati word' mozar meaning within, inside.] Masculine and neuter nouns ending in a, change a into e in the plural when Gase terminations are applied. [The following is an example:

Feminine nouns preserve, however, the final a in the plural before case terminations. Masculine nouns ending in a, i, u, and o, do not undergo any change, and the case terminations are applied to the final letter of the original word. A notable exception in a is bhingota, a bee, which changes its ta to tya before any case terminations are affixed to it. Another exception in u is ju, louse, which becomes juva and keeps the double plural form juvas to which case terminations are affixed. Pronouns both personal and relative are the same as in Marathi, but their case forms differ. Except for the difference of case forms, the first and second personal pronouns form their plurals in the same way as the Marathi. On the other hand, the third personal pronoun and the relatives insert an s in the plural before the case terminations.[

PRONOUNS.

] The demonstrative this differs from the Marathi. The ha, hi, and he of tlie latter become hau in the masculine, and hai in the feminine and neuter. [This hau is nearer than the Marathi ha to the Sanskrit asau (the a being dropped and s changed to h). It is declined as follows:

PRONOUNS, HAU, HAI, this.

] The interrogative or indefinite kon suffers no change in its crude form, but the indefinite pronominal particle kay takes kasa as its crude from to which the case terminations are added. Though curious, the conjugation of verbs is, with but few exceptions, regular.[

The following are the brief details of the leading Irregular verbs:

Irregular verbs.

Bol to speak is conjugated like kar except in the past tense which has bolnu, bolnut, for the first person, and bolna and bolnat for the second and third person singular und plural.] The present tense has one form for the singulars of all persons and another for the plurals, the terminations being s and tas respectively, thus kar to do has karas and kartas. These seem to be derived from the old Prakrit present participle karant, further corrupted to karat. In Marathi, also, the present tense is formed from the present participle, but to distinguish the persons the old personal terminations are appended to it, while in Khandeshi they are not. In Gujarati the old present participle is used to form the past conditional, and no terminations are applied to distinguish the persons. The t of the participle is however softened to s in the Khandeshi. In the plural kartas which corresponds to the Marathi karitat the second t is softened. The past tense is formed by adding n, the terminations being nu and nut for the first person, and na and nat for the second and third persons, singular and plural respectively. To Marathi this n is unknown, but it is used in northern Gujarati, as in bandhano for bandhayo 'bound', dithano for dekhayo 'seen', and in some verbs in the Braj Bhasha, as "kina, ' done', dina 'given', and lon, ' taken'. The n is generalised from such old Prakrit forms as-dinna ' given' for ' datta' (Sansk.), luna' cut' for luna (Sansk.), and bhinna 'divided' for bhinna (Sansk).

The future has su and sut for the first, shi and shal or shat for the second, and i and tin or til for the third persons singular and plural. The s of the first and second persons seems to be the old Prakrit ss * and Sanskrit sya * . In Gujarati it is preserved in all the persons. The Marathi l is dropped or rather not affixed, except in the optional plural forms of the second and third persons. [Prof. R. G. Bhandarkar, M.A., Hon. M.R.A.S.] The irregular karmani prayoga of the Marathi and Gujarati, which requires the verb to agree in gender and number with the object, is found in Ahirani. Thus, ' a house was built by Ram,' Ramani ghar bandh; a book was read by a Brahman, Brahmanni pothi vachi. Adjectives vary in form according to the number and gender of the nouns they qualify, but not according to their case inflections as in Marathi. The Ahirani conjunctions an and na correspond to the Marathi ani and va and the Gujarati ana and ne. There is a peculiar word for moreover, akhor. The adverbs of place are: here, athe, ithe (Sansk. atra); there,, tathe, tatha (Sansk. tatra); where, kathe, hatha, and koth (Sansk. kutra). The adverbs of time are: when (relative), javhaya, jadhaya, and jadhal, corresponding to the Sanskrit yada; then, tavhaya, tadhala (Sansk. tada) ; when (interrogative), kavhaya, kadhaya (Sansk. kada). The adverbs of manner correspond to the Marathi- forms. Some peculiar words are in use derived neither from Gujarati nor from Marathi. Among these may be noticed dndor a boy, ander a girl, bak towards, ibak hither, tibak thither, pan near, dhura until, and mayav alas! [Contributed by Rao Saheb K. B. Marathe, B.A., LL B]

Among themselves the Western Satpuda Bhils speak a dialect of Gujarati, while those further east use a form of Nemadi. Most of them know a little Hindustani or Marathi and employ it in speaking to Europeans or men from the plains, to whom their own dialect is unintelligible. A very marked local tendency is to drop every possible consonant. Liquids go first as in Koi for Koli, Mai for Mali; they are often followed by one goes in the social scale the as Vaijo for Vaghdev. The lower probably due to the more marked is the provincialism. It is is most. influence of the aboriginal races among whom the marked. [Mr. Sinclair in Ind. Ant. IV. 100.]

Except of Bohoras who speak Gujarati, the home tongue of almost all Khandesh. Musalmans is Hindustani.

Details. 1872.

The following tabular statement gives, for the year 1872, details of the population of each sub-division according to religion, age, and sex:

Khandesh Population, 1872. Sub-divisional Details,

continued.

Khandesh Population, 1872. Sub-divisional Details—continued.

From the above statement it appears that the percentage of males on the total population was 51.58 and of females 48.41. Hindu males numbered 489,429, or 51.62 per cent, and Hindu females numbered 458,850, or 48.38 per cent of the total Hindu population; Musalman males numbered 40,604, or 51.16 per cent, and Musalman females 38,755, or 48.84 per cent of the total Musalman population. Parsi males numbered 42, or 68.85 per cent, and Parsi females numbered 19, or 31.15 per cent of the total Parsi population. Christian males numbered 314, or 60.73 per cent, and Christian females numbered 203, or 39.27 per cent of the total Christian population. Other males numbered 221, or 51.87 per cent, and other females numbered 205, or 48.13 per cent of the total Other population.

Health.

The number of infirm persons was returned at 7298 (males 4672 females 2626), orseventy per ten thousand of the total population. Of these 382 (males 279, females 103), or four per ten thousand were insane; 618 (males 438, females 180), or sixteen per ten thousand idiots; 1009 (males 681, females 328), or ten per ten thousand deaf and dumb; 3757 (males 2063, females 1689), or thirty-seven per ten thousand blind; and 1532 (males 1206, females' 326), or fifteen per ten thousand lepers.

Age.

The following tabular statement gives the number of the members of each religious class of the inhabitants according to sex at different ages, with, at each stage, the percentage on the total population of the same sex and religion. The columns referring to the total population omit religious distinctions, but show the difference of sex.

Khandesh Population by Age, 1872.

continued.

The Hindu population of the district belongs, according to the 1872 census, to the following sects:

Khandesh Hindu Sects, 1872.

From this statement it would seem that of the total Hindu population the unsectarian classes numbered 889,981 or 93.78 per cent; the Shaivs 32,845 or 3.41 per cent; the Vaishnavs 21,273 or 2.24 per cent; and the Shravaks 5280 or 0.55 per cent. The Musalman population belongs to two sects, Sunni and Shia; the former numbered 73,088 souls or 92.09 per cent, and the latter 6271 souls or 7.90 per cent, of the whole Musalman population. The Parsis are divided into two classes, Shahanshai and Kadmi; the number of the former was fifty seven or 93.44 per cent, and of the latter was four or 6.55 per cent. In the total of 517 Christians there were 3 Armenians, 70 Catholics, and 444 Protestants, including 6 Episcopalians, 72 Presbyterians, 15 Wesleyans, and 351 native Christians. Other religions were represented by 5 Brahmos, 59 Sikhs, and 36 Jews. Besides these, 326 persons, under the head Others, remained unclassified.

Occupation.

According to occupation the 1872 census divided the whole  population into seven classes:

I.—Employed under Government, or municipal or other local authorities, numbering in all 16,256 souls or 1.58 per cent of the whole population.

II.—Professional persons, 5896 or 0.57 per cent.

III.—In service or performing personal offices, 9106 or 0.88 per cent.

IV.—Engaged in agriculture and with animals, 216,975 or 21.09 percent.

V.—Engaged in commerce and trade, 17,708 or 1.72.

VI.—Employed in mechanical arts, manufactures, and engineering operations, and engaged in the sale of articles manufactured or otherwise prepared for consumption, 195,294 or 18.98 per cent.

VII.—Miscellaneous persons not classed otherwise, (a) wives 201,933 and children 845,638, in all 547,571 or 53.23 per cent; and (b) miscellaneous persons 19,836 or 1.93 per cent; total 567,407 or 55.16 per cent.

For descriptive purposes the different Hindu classes group themselves most conveniently under the heads of Brahmans, Writers, Traders,- Husbandmen, Craftsmen, Labourers, Early or Unsettled Tribes, Depressed Classes,; and Beggars.