The usual character employed for writing Bihari is that known as Kaithī, a cursive form of the well-known Nagari character of Upper India. The name of the character is derived from the Kāyath or Kāyasth caste, whose profession is that of scribes. Kaithi is widely spread, under various names, all over northern India, and is the official character of Gujarati. The Nagari character is commonly employed for printed books, while the Brahmans of Tirhut have a character of their own, akin to that used for writing Bengali and Assamese. In the south of the Bihari tract the Oriya character belonging to the neighbouring Orissa is also found.
Bihari has to its east Bengali, also a language of the Outer Band. To its west it has Eastern Hindi, a language of the Intermediate Band (see [Indo-Aryan Languages]). While it must decidedly be classed as an Outer language, it nevertheless shows, as might be expected, some points of contact with the Intermediate ones. Nothing is so characteristic of Bengali as its pronunciation of the vowel a and of the consonant s. The first is sounded like the o in “hot” (transliterated o). In Eastern Bihari the same vowel has a broad sound, but not so broad as in Bengali. As we go westwards this broad sound is gradually lost, till it entirely disappears in the most western dialect, Bhojpurī. As regards s, the Māgadhī Prakrit pronounced it as ś, like the sh in “shin.” The Prakrits of the West preserved its dental sound, like that of the s in “sin.” Here Bengali and Eastern Hindi exactly represent the ancient state of affairs. The former has the ś-sound and the latter the s-sound. At the present day Bihari has abandoned the practice of the old Māgadhī Prakrit in this respect, and pronounces its s’s as clearly as in the West. There are political reasons for this. The pronunciation of s is a literal shibboleth between Bengal and Upper India. For centuries Bihár has been connected politically with the West, and has in the course of generations rid itself of the typical pronunciation of the East. On the other hand, a witness as to the former pronunciation of the letter is present in the fact that, in the Kaithi character, s is always written ś. In the declension of nouns, Bihari follows Bengali more closely than it follows Eastern Hindi, and its conjugation is based on the same principles as those which obtain in the former language.
The age of Bihari as an independent language is unknown. We have songs written in it dating from the 15th century, and at that time it had received considerable literary culture. Bihari has three main dialects, which fall Language. into two divisions, an eastern and a western. The eastern division includes Maithilī or Tirhutiā and Magahī. Magahi is the dialect of the country corresponding to the ancient Magadha, and may therefore be taken as the modern representative of the purest Māgadhī Prakrit. Its northern boundary is generally the river Ganges, and its western the river Son. To the south it has overflowed into the northern half of Chota Nagpur. It is nearly related to Maithili, but it is quite uncultivated and has no literature, although it is the vernacular of the birthplace of Buddhism. Nowadays it is often referred to by natives of other parts of the country as the typically boorish language of India. Maithili faces Magahi across the Ganges. It is the dialect of the old country of Mithilā or Tirhut, famous from ancient times for its learning. Historically and politically it has long been closely connected with Oudh, the home of the hero Rāma-candra, and its people are amongst the most conservative in India. Their language bears the national stamp. It has retained numerous antiquated forms, and parts of its grammar are extraordinarily complex. It has a small literature which has helped to preserve these peculiarities in full play, so that though Magahi shares them, it has lost many which are still extant in the everyday talk of Mithila. The western division consists of the Bhojpuri dialect, spoken on both sides of the Gangetic valley, from near Patna to Benares. It has extended south-east into the southern half of Chota Nagpur, and is spoken by at least twenty millions of people who are as free from prejudice as the inhabitants of Mithila are conservative. The Bhojpuris are a fighting race, and their language is a practical one, made for everyday use, as simple and straightforward as Maithili and Magahi are complex. In fact, it might almost be classed as a separate language, had it any literature worthy of the name.
(Abbreviations: Mth. = Maithili, Mg. = Magahi, Bh. = Bhojpuri, B. = Bihari, Bg. = Bengali. Skr. = Sanskrit, Pr. = Prakrit. Mg. Pr. = Magadhi Prakrit.)
Vocabulary.—The Bihari vocabulary calls for few remarks. Tatsamas, or words borrowed in modern times from Sanskrit (see [Indo-Aryan Languages]), are few in number, while all the dialects are replete with honest home-born tadbhavas, used (unlike Bengali) both in the literary and in the colloquial language. Very few words are borrowed from Persian, Arabic or other languages.
Phonetics.—The stress-accent of Bihari follows the usual rules of modern Indo-Aryan vernaculars. In words of more than one syllable it cannot fall on the last, whether the vowel of that syllable be long or short, pronounced, half-pronounced, or not pronounced. With this exception, the accent always falls on the last long syllable. If there are no long syllables in the word, the accent is thrown back as far as possible, but never farther than the syllable before the antepenultimate. Thus, ki-sȃ-n(a) (final a not pronounced); pȃ-nī, há-ma-rā; dé-kha-lả-hū. In the last word there is a secondary accent on the penultimate, owing to the following imperfect vowel (see below). When the first syllable of a word has not the main stress-accent, it also takes a secondary one, as in dè-kha-li-ai-nhi. When the letter a follows a syllable which has the accent (secondary or primary) it is only half pronounced, and is here denoted by a small a above the line. In Mth. (but not in Mg. or Bh.) a final short i or u is often similarly very lightly pronounced, and is then represented by the same device. Before such an “imperfect” i or u the preceding syllable has a secondary accent, if it has not already got the main one.
When a word ends in a preceded by a single uncompounded consonant, the a is not pronounced; thus, kisȃna, sounded kisȃn. This vowel is sometimes pronounced with a drawl, like the a in “ball,” and is then transliterated å. When a has this sound it can end a word, and in this position is common in the second person of verbs; thus, dēkhå, see thou. This sound is very frequently heard in Bhojpuri, and gives a peculiar tone to the whole dialect, which at once strikes the casual hearer. The usual short form of the letter ā is a, but when this would lead to confusion it is shortened in Mth. and Mg. to a sound like that of a in the German Mann, and is then transliterated ả. In Bh. it is always shortened to a. As an example, from pānī, water, is formed the word paniyā, but (in Mth. and Mg.) from the word mārab, to strike, we have Mth. mảralī, Mg. mảralī′, I struck, because maralī (-lȋ) would mean “I died.” In Bh. maralȋ actually has both these meanings. The letters e and o may be either long (ē, ō) or short (e, o). In Skr. the diphthongs āi and āu (here transliterated āī, āū) are much longer than the Bihari ai and au, which are contractions of only a + i and a + u respectively. We may compare the Sanskrit, or tatsama, āī with the English “aye,” and the tadbkava ai with the English “I.” In counting syllables in Bihari, ai and au count each as two syllables, not each as one long syllable. The Skr. ṙ appears only in tatsamas. Nasalization of vowels is extremely frequent. In this article it is represented by the sign ~ over the vowel, as in mūh, mảralȋ and dekhalahū.
As regards consonants, ḍ and ḍh, when medial, are pronounced as strongly burred ṛ and ṛh, and are then transliterated as here shown. There is a constant tendency to change these to an ordinary dental r and rh; thus, ghōḍā, pronounced ghōṛā or ghōrā. The semivowels y and v are always pronounced like j and b respectively, unless they are simply euphonic letters put in to bridge the hiatus between two concurrent vowels; thus yāūvana pronounced jāūban, and maliyā for mali-ā, ghoṛawā for ghoṛa-ā. The sibilants ś and s are both pronounced as a dental s, but (a relic of the old Mg. Pr.) are both invariably written as a palatal ś in the Kaithi character. Thus, the English word “session” (seśun) is written śeśan and pronounced sesan. The cerebral ṣ, when uncompounded, is pronounced kh. When compounded, it generally has its proper sound. Thus, ṣaṣṭha, sixth, is pronounced khaṣṭh. As a general statement we may say that Bihari spelling is not fixed, and that there are often many ways of writing, and sometimes two or three ways of pronouncing, the same word.
The main typical characteristics of Mg. Pr. are that western Pr. s becomes ś, and that western Pr. r becomes l. We have seen that the change of s to ś occurs in Bengali but not in Bihari, and have given reasons for the change back to s in the latter language, although the Mg. Pr. ś is retained in writing. In both Bengali and Bihari, a western r is not now represented by l, but is represented by r. This deviation from the Mg. Pr. rule is only apparent, and is due to the letter r representing two distinct sounds. In Skr., in the western Prakrits, and in the modern western languages, r is a cerebral letter, with a cerebral sound. In the modern eastern languages, r is a dental letter, with a dental sound. Everywhere, both in old times and at the present day, l was and is a dental letter. The meaning, therefore, of the change from western Pr. r to Mg. Pr. l was that the western r lost its cerebral sound, and became a dental letter, like l. That dental character is preserved in the r of the modern eastern languages. In fact, in Bihari r and l are frequently confounded together, or with n, another dental letter. Thus, we have kālī or kārī, black; phar or phal, fruit; Skr. rajju-, B. leju-rī a string; Lakhnaur, the name of a town, quite commonly pronounced Nakhlaul; and the English names Kelly and Currie both pronounced indifferently karī or kalī. Compare Assamese saril for Skr. śarīra-.
The genius of the Bihari language is adverse to the existence of a long vowel in a tadbhava word, when it would occupy a position more than two syllables from the end. Thus, ghōṛā, but ghoṛawā; mārel, but mảralī. This is subject to various subsidiary rules which will be found in the grammars. The principle is a most important one, and, indeed, pervades all Indo-Aryan vernaculars of the present day, but it is carried out with the greatest thoroughness and consistency in Bihari. The whole system of declension and conjugation is subject to it. When ā preceding i or e is shortened, the two together become ai, and similarly a shortened ā + u or o become au.