All major Munda languages of Jharkhand now each has its own script. It all started with Raghunath Murmu creating Ol Chiki for Santali, then came Lako Bodra's Warang Chiti for Ho, Ruidas Nag's Mundari Bani for Mundari, now Bhumij language also has a script called Ol Onol. I recently learnt that a smaller branch of Mundas who now reside in West Bengal, called the Kodas, too have developed a script called Nag Chiki. It seems to me that invention of scripts has become a hobby of the Munda people, not to mention the numerous scripts devised for Santali and Ho before they were rejected in favor of the currently used scripts.
A common characteristic of these scripts is that they are alphabetic, except for Warang Chiti which functions as an alphasyllabary in some cases. Each of these scripts have some or the other deficiencies when employed to write Munda languages. As Abhishek Bilkan Aind has pointed out, alphabetic scripts like Ol Chiki and Mundari Bani are not perfect systems for writing Munda languages, because these scripts only represent the phonetics of the language, these symbols do not carry any other linguistic information, they are practically as good as any other phonetic writing system, be it Latin, Devanagari, Odia or Bengali.
One may argue that Ol Chiki letters are modelled after the objects they phonetically represent, for instance, Ol Chiki letter ᱛ ot is shaped a circle representing the earth, ot in Santali means earth, the letter ᱫ ud is shaped like a mushroom and ud means mushroom. Surely, Ol Chiki letters have pictographic origins, but so does the Latin script. Pictographic association of Ol Chiki letters to actual objects only serves as a mnemonic for new learners of the script. Once the script is learnt, the letters ot and ud being in the shape of the earth and mushroom serves no purpose, the letters acquire a purely phonetic function.
In this regard, Warang Chiti went a little further, using the letters not just as phonetic symbols, but also as pictographs. For instance, the Warang Chiti letter 𑢵 ot, which is exactly the same as Ol Chiki ot both graphically, phonetically and meaning wise, is also used to write the word for earth. So, instead of writing 𑣉𑣕𑣈 ote in Warang Chiti, one would simply write upper case ot. Similarly, the Warang Chiti letter 𑢨 a:e represents the sound e as well as the word a:e meaning 'he/she'. So the sentence nir tana:e, is written as 𑣓𑣂𑣜 𑣕𑣓𑣁𑢨, the upper-case e is a pictograph of a finger pointing to a third person.
Use of 𑢶 and 𑢨 as pictographs in Saar Hora Part I.
Use of 𑢪, 𑢭 and 𑢵 as pictographs for oŋ, biŋ and ote in Saar Hora Part I.
But Warang Chiti pictograms are limited in number, only around 10 letters out of 32 can be used as pictographs.
What is wrong with the existing Munda writing systems? In the following sections, I will be looking at various problems with Munda scripts as I discover them while analyzing specific writing systems.
Santali Latin Alphabet
Looking at it from phonological point of view, this system represents the sounds of Santali perfectly. It distinguishes 9 vowel sounds (only 8 of which occur frequently). It distinguishes between checked and unchecked final consonants. This system, which is based on Lepsius Standard Alphabet developed by Karl Richard Lepsius, was first used for Santali by LO Skrefsrud. This system uses the letters of the Latin alphabet with various diacritics, which Bodding himself calls "troublesome"¹, and writes:
Diacritical signs are a nuisance; but until we get separate and easily written characters for all the different sounds, they cannot be avoided. ²
The major problem with this system is its extensive use of diacritics. I took a random post written in Latin Santali from Facebook, the post contained 969 letters, and 186 diacritics, i.e., about 20% of the characters had diacritics on them. Though the sample size is quite small, but still, the frequency of diacritics is quite high, one in every five characters written has a diacritic attached to it. This system is perfect for academic purposes, but it is really "troublesome" for general use.
The use of diacritics rather than separate letters often leads to people omitting the diacritics while writing, as Bodding points out. ³ Hence, in my opinion, it would have been better if separate letters were used for the unique vowels rather than modified letters. Separate vowel signs would ensure that the identity of the vowel sounds remains intact as the younger generation of Santali speakers have a tendency to merge the open and closed vowels.
Another problem I find with Santali Latin alphabet is its representation of
checked consonants as ḱ ć t́ and ṕ with hard (unvoiced) sounds as base letters.
While this is good enough for academic transcription, in reality, since these
sounds change into g j d and b when
conjugated, it would have been better to use these as the bases for checked
consonants, thus ǧ ǰ ḓ b̭ ( or g̒ j̒ d̦ b̦),
this would have eliminated the need to change the entire letter for another
letter when conjugating, just removing the diacritic from the base letter would
have been enough. In the present system, when words like saṕ are
conjugated with oḱ, it results in saboḱ, the ṕ changing
to b. If ǧ ǰ ḓ b̭ were used instead, then the
conjugation would have been sab̭+oǧ=saboǧ.
One might argue that using soft consonant letters for hard consonants is
phonologically incorrect. But the same argument can be made for using k
for ʔ sound. Santali ʔ is not close to k,
Skerfsrud used it to represent the glottal stop sound just to be consistent.
"...but as we are compelled to use t as the basis for
the dental, and p for the labial, it seems to lie more
consistent also to use k as the basis for the guttural,
and ch for the palatal; for if p, which changes
into b, when succeeded by a vowel, is the right basis for the
labial..." ⁴
I believe that the speakers of Munda languages perceive the hard (voiced)
checked stops as checked forms of soft (voiced) consonants, rather than of hard
consonants. This is no doubt true for Southern Santali, Mundari and Ho
speakers. But I have seen people using Santali Latin and yet unconsciously
write ṕ as b, as in instances presented below:
These pictures show ṕ mistakenly written as b. It is most probably because the native speaker's mind subconsciously associated the checked consonants with soft consonants rather than hard consonants (perhaps due to anticipation of a conjugation). In my opinion, these subconscious associations that run in a native speaker's mind matter more than what a nonnative speaker linguist might hear or perceive.
This becomes very relevant when looking at the present discourse in Jharkhand regarding the use of Ol Chiki for Santali in Santal Paragana. Those opposing Ol Chiki have been relying on the fact that it uses soft sounds to render the hard checked stops. While better arguments can be made against Ol Chiki, somehow 'water' being written as dag in Ol Chiki has become a prominent argument, even though it is made clear that Ol Chiki g ᱜ represents a checked stop unless followed by ohot ᱽ (i.e., ᱜᱽ). Bodding and Hoffman were two foreigners who encoutered the exact same set of sounds but decided on different ways of transcribing them, Bodding went with phonetic accuracy, while Hoffman went with a practical spelling.
Anyway, it is up to Santali speakers to decide whether to base their orthography on perception of a native speaker not.
An interesting phenomenon I noticed among some Santali youth is that they are unable to distinguish between certain checked consonants in addition to not being able to distinguish o from u. I had spoken to a Santali speaker who could not tell apart the final consonants in words like oṕ, ot́, and ut́. He pronounced all three words as uḱ!
Ol Chiki
This system has 6 vowels which have their own letters, additional vowels can be produced by addition of găhlă ṭuḍăg. But the Ol Chiki writers do not generally distinguish between open e̠, o̠ and closed e, o. Ol Chiki writers use u in words where Santali Latin would use o. This is either because Ol Chiki writers are unable to perceive the difference between them or because o has merged with u in their respective varieties of Santali. Ol Chiki does not have a separate letter for e, it can be written by adding a modifier to e̠ (ᱮᱹ), but that is not done by Ol Chiki writers. It seems e too has merged with e̠ or i like o.
Though supporters of Santali Latin like to point out that daḱ is written as dag in Ol Chiki, this is an important question when it comes to Munda languages, this question is not limited to Santali, similar problem exists with Ho and Mundari. The problem is that ǧ and ǰ of Old Munda behave differently than ḓ and b̭. For instance, in Ho, sab̭+oǧ=saboǧ, geḓ+oǧ=gedoǧ, but daǧ+oǧ=daǧaao (not dagoǧ), sareǰ+oǧ=sareyoǧ (not sarejoǧ). The conjugations are regular in case of ḓ and b̭, but very irregular in case of ǧ and ǰ (of all Ho verbs that ended in ǰ only 'goǰ' is conjugated regularly). This raises problems when devising a writing system where conjugations are desired to be regular. The solution I propose to this irregularity is introduction of non-phonetic letters that represent these checked consonants but are detached from full consonantal pronunciation.
I believe that a perfect script for Munda languages should reflect the historical pronunciations, this has to be done to distinguish homophones (and also to maintain a link with linguistic heritage). This is especially true for language like Ho where many homophones occur due to phonetic changes, for instance, the words sūr, uku, uḓ, bā, tū, horo all have two meanings because two different sounds have merged to one. The distinction has to be shown in orthography, else a purely phonetic script will just be function as a form of a tape recorder.
On aesthetics of Ol Chiki, Raghunath Murmu does an impressive job on simplifying shapes of real-life objects in the form of an alphabet.
Warang Chiti
Current Warang Chiti orthography has no set rules to distinguish between homophones like sūr, uku, uḓ, bā, tū and horo. Has two letters for two affricates, but all eastern Munda languages only have one sibilant sound /s/. The existence of this letter in the alphabet has paved way to Warang Chiti writers to misspell various words with /s/. Resulting in words like esu, sim, sirma being wrongly written as eshu, shirma and shim. No regular orthography, verb conjugations are haphazard with many random and irrational spelling rules.
It seems to me that the creator of the script did not know that the endings of words like /ʌʔᵋ/, /uʔᶦ/, /reʔᵋ/, /lɑːʔᶦ/ has evolved from the final ć in the old Munda language. Therefore, words like /ʌʔᵋ/ and /uʔᶦ/ are each given separate letters that also represent vowels e̠ and i. It thus becomes very difficult to adapt this script to reflect older phonetic system of Munda languages.
I believe it is most probable that among the languages of eastern branch of North Munda languages, i.e., Santali, Mundari, Ho, Bhumij, Koda, Asuri, Birhori, Korwa, Birjia etc, Santali is the most conservative variety in terms of preserving Old Munda phonology. Although there is some evidence of certain vowels and checked consonants merging in the speech of younger generation, written Santali of Santal Paraganas preserves vowels and checked consonants of Old Munda. But Santali also happens to be one of the languages most receptive of Aryan phonology, for instance, it has developed aspirated consonants which were most probably absent in old Munda language. These aspirated consonants mostly occur in Aryan loanwords, but some native words like diri are pronounced with as dhiri.
In this regard, Ho has been least influenced as it has not developed aspirated consonants, and also one of the least receptive of Aryan vocabulary (this might also be true for Hasadaʔ Mundari). Yet, among the Munda languages, it is Ho which has diverged the furthest from old Munda phonology. I believe this divergence is not caused by influence from Aryan languages but is rather an independent development. The reason I believe this is that changes like ć > ʔᵋ/ʔᶦ, muruḓ/baha/tuṛu > mūr/bā/tū, cannot be attributed to influence of Aryan languages.
PO Bodding gives 9 vowels in his dictionary, a ạ e e̠ ẹ i o o̠ ọ u. Among these, ẹ does not occur in any common Munda vocabulary, it is specific to Santali occuring in few places. And ọ occurs in the word ọl, Mundari and Ho form of this word is ul. I am of the opinion that this word might originally be ol, the distinction might have developed because of the Aryan loan word ol (Amorphophallus campanulatus). This is called hadaʔ in Ho. So, ọ might actually have been an o in this word.
Based on comparison of Munda languages, I believe that Old Munda langauge had 8 vowels. These vowels can be found in written Santali of Santal Paragana. But what happened to these vowels in other Munda languages? I have recently realized that Ho language spoken in areas surrounding my village actually has 7 vowels and, in some areas, has 8 vowels.
This might come as a surprise for Ho speakers who only recognize 5 basic vowels (I am not counting long vowels here, as they are variations of basic vowels). In order to understand where these additioinal vowels occur, we will have to look at words such as era, hola, menaʔ, ena, joka etc. The word era, though written with an e, is not actually pronounced as /ɛrʌ/, the actual pronunciation sounds a bit like /ɪrʌ/, but it would be wrong to write it as ira as that would be wrong too. The vowel in this word is distinct from both e (as in edel) and i (as in ipil). I think it is either the vowel /e̞/ or /e/. It is this vowel that Bodding transcribed as e (distinct from e̠). In the word hola होला, the vowel is not the same as in the word hon होन, the vowel in hola होला sounds more like u, but it is not a proper u either, it is a vowel in between o and u, Bodding transcribed this vowel as o (distinct from o̠). In the Ho spoken in my village and surrounding areas, I have only heard only these 2 additional vowels apart from the basic 5. But there are some people who pronounce the word kula कुला differently, in my village it is pronounces as kule कुले /kulɛ/, but there are people who pronounce it close to /kulə/, the vowel in the end is neither a nor e, but somewhere in between (most likely an /ə/). Similarly, the word nida निदा is pronounced nide निदे /nɪdɛ/ in my village, but /nɪdə/ in other areas. This vowel might be the Ho equivalent of Santali ạ, though Santali ạ occurs much frequently.
To properly represent all sounds of languages like Santali and Ho, distinct letters would be needed for all the 8 vowels. This 8-vowel system would also explain the homophones such as uku and uḓ.
Ol Onal and Nag Chiki
Looking at the elaborate and rich strokes and curves of Ol Onal script, the question of associating symbols with sounds, objects and culture comes to my mind. I think it is important to consider why we associate certain shapes to certain sounds. I am of the opinion that graphically, the written symbols should reflect some or the other aspect of the culture and heritage of Mundas. When selecting an ideal script, its graphic and visual component has to be taken into account as well, a proper justification and explanation has to be provided as to why those symbols are chosen.
Speaking of shapes and about linking them to our cultural heritage. I think Ol Chiki has very good aesthetics and I think it does a good job of graphically representing Santal culture with all its geometric properties. I am saying this because the Santal villages I have seen are all well planned out and geometrically aligned. I have never seen a Ho village with such great geometric alignment.
Another issue that none of the extant scripts address is that of the prefixes and infixes. Infixes are a unique feature of all Munda languages. Unfortunately, only a few of the Old Munda infixes are productively used today, in case of Ho, only the infixes -n- and -p- are productively used, -t- is quite rare. In Santali, though infixes are used more frequently than in Ho, the issue is quite the reverse, words like hopon and botor are more common than the root forms i.e., hon and boro, this too renders the infixes unproductive because the infixed forms are replacing the root forms rather than diverging in meaning. In both these cases, it is clear that the knowledge of infixes is diminishing among Munda speakers. Infixes like -r-, -m-, -r.- have become unproductive and a vast majority of speakers have no knowledge of them. This I think is a tragedy, stripping these languages of any chance of lexical expansion.
It is for this reason that I advocate for having separate and distinct letters to represent these affixes. I believe this would ensure their survival and might even lead to them being productive again. When I say distinct letters, I mean that they must not be variations of other letters (for example, infixed n being represented with an n modified by a diacritic). I have gone through the problem with diacritics, they are easy to be omitted, which is not ideal if we want to ensure the survival of infixes.
***
¹ Preface. A Santal Dictionary, Paul Olaf Bodding.
² Preface. A Santal Dictionary, Paul Olaf Bodding, pp. vii.
³ Preface. A Santal Dictionary, Paul Olaf Bodding, pp. viii.
⁴ A Grammar of the Santhal Language, L O Skrefsrud, pp. 9.
Comments
Post a Comment