Konkani is the official language of the state of Goa, located on the western coast of India. This language has faced many political threats such as four hundred and fifty years of Portuguese colonization and contention with Marathi in order to be recognised as the official language, post-Liberation in 1961. It finally entered the Eighth Schedule of the Constitution of India in 1992. These hardships have diversified the nature of Konkani. It is spoken in several dialects in the state of Goa and elsewhere, and is written in five different scripts, owing to the migration of Konkani people from Goa over the centuries. There are Konkani communities in the neighbouring states of Karnataka, Kerala and Maharashtra, which are heavily influenced by the dominant local culture. Hence, Konkani is written in Devanagari, Roman, Kannada, Malayalam and Perso-Arabic scripts. This phenomenon creates a linguistic and literary gap in the community of Konkani speakers. Persistent efforts to bridge the gap have been made, and one of them is by taking the assistance of technology. The World Konkani Centre situated in Mangalore, Karnataka, has developed a transliteration tool, Konkanverter, which transliterates Unicode text between four of the five writing systems of Konkani, namely Devanagari, Roman, Kannada and Malayalam. This paper reports the attempts to digitize, and transliterate an available performance play text in Konkani, from one script to another. It also explores digitization as a way of preserving Konkani texts in multi-script formats.
www.unigoa.ac.in
The official language of Goa, Konkani, also spoken in pockets of Karnataka, Kerala
and Maharashtra, showed a negative growth rate according to the Census of 2011[
In addition to this apprehension, The Telegraph, a noteworthy Indian national daily
published the interview of language expert Professor G N Devy in August 2017 after
he had published several volumes of the People‟s Linguistic Survey of India, wherein
he observed that, “Nearly 400 of India's 850-odd languages face the threat of
extinction because of an erosion of traditional jobs that is fuelling migration to cities,” and
that “the languages spoken in the coastal areas will be the worst-affected” [
Goa has been through the perils of colonization, mechanization, migration and so forth. In addition, the Konkani language carries a linguistic peculiarity of being written in five different scripts, with two or more scripts being used in the same geographical region. The literature created in these five scripts remains restricted to readers who are familiar with the scripts. Their geographical location decides their readability of the scripts. Therefore, there will be very few who can read in three scripts, let alone four or five.
However, it is remarkable that its cultural and linguistic identity has been preserved by its people quite fervently, despite the grave threats that appeared time and again to erase it. The Konkani people carried their language, their deities and their culture to the places they migrated to, and planted their cultural heritage in alien soil. However, in a scenario where technological tools greet the end-users in global languages, a minority language can easily be bypassed. Although close to nine of the twenty-two scheduled languages recognized by the Constitution of India are featured by Google, so far, Konkani has not been one of them. However, under Google Indic Keyboard, Konkani-English interface is provided, with a limited corpus, in Android smart-phone keyboards.
If a small language like Konkani is to survive the invasion of technology and its mission to standardize communication, it needs to adopt technology to preserve its diversity, and subsequently its culture and heritage. This study aims to explore the possibility of having a cross-orthographical readership of Konkani using the transliteration tool known as Konkanverter, which will not only bridge the orthographic disparity and increase readership and production of literature, but also contribute to the creation of Konkani digital archive. 2
Due to the migration of people from Goa over the centuries, Konkani diasporic communities exist in Karnataka, Kerala, and Maharashtra. It is not surprising that their spoken dialect of Konkani carries the flavour of the regional languages in the respective states, which are Kannada, Malayalam and Marathi respectively. Hence the local scripts have been adopted by these Konkani communities in the respective regions.
The two dominant scripts of Konkani in Goa are Devanagari and Roman or Romi as it is colloquially known. Devanagari orthography is considered the official one; Romi is used by select weeklies/ magazines and Tiatr (drama) scripts.
Rajan terms this phenomenon of multiple scripts as “synchronic trigraphia” and
goes on to elucidate it as “a major issue of political contention inside the community,
each group favouring the usage of a particular script as the official orthography.
Different orthographic communities exist in isolation with minimal interaction and with
its [sic] own literary tradition, as very few people are fluent in multiple
orthographies” [
Although, these communities may have elementary proficiency in reading in the
other writing system, this multiplicity of scripts creates an obstacle for wider reading.
Hence, literature and language both become less accessible to the native speakers of a
single region: “This disparity in scripts creates intra-lingual barriers which make
writings in Konkani inaccessible even to a native Konkani speaker who has limited or no
knowledge of all these five scripts” [
Rajan suggests that a “statistical machine transliteration engine with reasonable
accuracy would greatly enable cohesion and interaction among the greater linguistic
community. Facilitating the usage of multiple scripts would also encourage more
linguistic diversity among the community” [
The methodology used in this study is exploratory in nature. As Dudovskiy puts it,
“Exploratory research, as the name implies, intends merely to explore the research
questions and does not intend to offer final and conclusive solutions to existing
problems” [
In the context of this study, exploratory research helps discuss benefits, and challenges that come in the way, of solving the major research question, i.e. whether digitization can pave a way to language preservation. The present study adheres to this methodology in a way that is applicable to the discipline. The study serves to explore the possibilities of using technology to safeguard the linguistic-cultural heritage of a certain language community, and paves way for further research in this area. The data and analysis presented in this study are first-hand; revision can be taken up as part of the next level of research in the same area. 4
World Konkani Centre [
The present research involved transliteration of a well-known Konkani play “Shree
Vichitrachi Jatra”, written by Pundalik Naik. This play belongs to national award
winning drama collection, Chourang (1982) [
The digital copy of the play was obtained in Devanagari. This text had lexical
errors in it, although not largely phonetic ones. Furthermore, it was found that the font
used was not in Unicode, and since Konkanverter only accepted fonts in Unicode, an
intermediate software tool known as Baraha FontConvert [
Table 2 explores ten random errors found in the transliterated text, and discusses the consistency of errors in the output. It shows column-wise progression, from source text, copied and pasted into font conversion tool (see Table 2 column „Source Font‟) to receive the Unicode output (see Table 2 column „Unicode Output Text‟). This Unicode output text was copied and pasted into Konkanverter to give its transliteration (see Table 2 column „Transliterated Output Text‟). The content of the last column of Table 2, i.e. „Expected Transliteration‟, is a result of directly typing the corresponding words of the column „Source Text‟ of Table 2 in Unicode into Konkanverter input text box, and the output received is not influenced by font conversion and hence is the expected machine transliteration. As given in Table 2, one can observe that the letters फ (fa), म (ma) seem to have been converted erroneously throughout the text; other major erroneous conversions are the joint half letters such as म्हळ्यार - म्हळार ; गाड्याक - गाडाक; वााट्यचें - वाटाचें. Some seem to be random errors (UT: आसला, तळु , काणेतरी) which were not repetitive.
The entire play was thus transliterated into Romi, and errors like the above (see
Table 2) had to be dealt with manually. At the time of writing this paper, it is
unknown whether these discrepancies would get transferred into Kannada or
Malayalam, as the present researchers are not familiar with the spoken dialect or the scripts
of these regions. Rajan [
The above transliteration accuracy rates are to be considered in the case of a text which is lexically accurate. A limitation of this study was that the source text was not proofread for spelling errors, which led to rise in the inaccuracy of transliteration. Therefore, care has to be taken by Konkani language experts to have it perfectly proofread after digitization, and also have the font converted to Unicode, which will ensure faithful transliteration.
For more accuracy to be obtained in translation, ST should be proofread and converted into Unicode, which will render the intermediate font conversion redundant. With only one conversion engine between the input and the output text, greater accuracy will easily be achieved.
Such an analysis of Konkanverter is incomplete without references to Google‟s
contribution to the field of transliteration. Transliteration is featured as one of Google
Input Tools [
The acceptance of only the Unicode text limits the usage of Konkanverter, as it discourages input of any other kind, such as optically recognized characters in general and voice input. Voice input is further deterred by not having the text on-the-fly.
Nevertheless, Konkanverter pushes the boundaries of learning of different scripts of the same language, and makes multi-script archiving possible; it not only bridges the gap between the two dominant orthographic Konkani communities, but establishes a kind of digital footprint for a language which could face the risk of becoming endangered. Such a tool can become an inspiration for minority languages across the globe to enhance their digital presence, and hence safeguard their identity in the digital revolution.