SC’s translation projects raced ahead in 2023 as retd. HC judges, law clerks help AI

But concerns persist over who will use them and lack of standardised legal vocabulary in all regional languages

December 31, 2023 08:04 pm | Updated 08:04 pm IST - New Delhi

 A view of Supreme Court in New Delhi. File

A view of Supreme Court in New Delhi. File | Photo Credit: SHIV KUMAR PUSHPAKAR

The Supreme Court of India’s monumental project of translating all of its 36,000 judgments into Scheduled Languages achieved unprecedented speed in 2023, with the E-SCR portal starting with just 2,238 translated judgments as of January and ending the year with over 31,000 rulings translated. 

Currently, the highest number of translated judgments are in Hindi at 22,396, followed by Punjabi (3,572), Kannada (1,899), Tamil (1,172), and Gujarati (1,112). 

Chief Justice of India D.Y. Chandrachud in his Republic Day speech had said that the Supreme Court was on a mission to provide SC judgments in all Scheduled Languages - a project that started under the tenure of CJI Ranjan Gogoi in mid-2019 and taken forward by his successors. 

Justice Gogoi had at the time given credit to his successor and then number two judge, Justice S.A. Bobde, for first mooting the idea of translating the court’s judgments into regional languages. Justice Bobde continued the efforts during his tenure as top judge, introducing AI tools and also inaugurating the Supreme Court Vidhik Anuvaad Software (SUVAS).

However, the speed of translations increased under the tenure of CJI Chandrachud, largely due to the court’s decision to put into service an army of retired High Court judges and their law clerks, who were able to correct erroneous interpretations or translations of the AI-based software. This coupled with the system set in place for the translations by the Editorial Section of the Supreme Court has resulted in the jump in pace, say officials. 

But despite the speed of translations picking up drastically in 2023, questions have been raised by lawyers and legal experts on how these translated judgments would be useful when High Courts are not yet permitted to conduct proceedings in regional languages except in Hindi-speaking States. Further, the translated rulings come with a disclaimer that absolves the Supreme Court Registry of any responsibility for incorrect or inaccurate translations. 

And apart from Hindi, there is no standardised glossary of legal terminology in other regional languages to facilitate uniform translations from English, which has already led to incoherent translations in some languages like Tamil, explained Justice K. Chandru (retd), former judge of the Madras High Court.

The Bar Council of India’s Bharatiya Bhasha Samiti chaired by former CJI Justice Bobde is currently working on creating this “Common Core Vocabulary” for all Indian regional languages, the government said earlier this year.

A Supreme Court source said that they started with breaking down the judgments year-wise and page-count-wise. “We found that most judgments were in the range of less than 30 pages and less than 20% of the judgments were 100 pages or more.”

The team also started picking out common orders for petitions that would help train the software for translations. 

As per the system set in place, once the translations are done through the software and by retired judges and law clerks (who are paid a stipend), they come back to the Editorial Section of the SC, where they are examined thoroughly. The whole process is being overseen by a committee headed by Justice S.A. Oka, who liaises with the Chief Justice.

But for common orders such as “SLP leave granted”, the machine translations in Hindi were throwing up an issue, the source explained. “The software was translating ‘leave’ to avkaash - which actually meant ‘holiday’. We needed the human eye to fix these things,” they said, adding that similar challenges were coming up in translations to other regional languages as well. 

As for the Tamil translations, Justice Chandru told The Hindu, “I have seen some of the translations made so far. None can read them. For each word in law in English, we have different usages in Tamil Nadu and there is no standardised version created so far… From where they coined the new Tamil words is anybody’s guess. Besides a shortage of Tamil words, even the construction of sentences is not proper.”

He added that while the AI system for translations had its own problems, there was a lack of expert translators who could get the job done accurately.

Even for the Hindi translations that are most common on the ESCR portal, lawyers practising in States like Uttar Pradesh and Madhya Pradesh have told The Hindu that the absence of headnotes in the translated judgments is another barrier in them being used. 

Given that Hindi is more commonly used for official business in States like Uttar Pradesh, Bihar, Madhya Pradesh, Rajasthan, etc., it has been easier to translate into the language, the SC source explained, adding that the process consumes significantly longer time in other regional languages.  

In the current body of translated judgments, there are 39 Bengali judgments, 872 in Marathi, 334 in Telugu, 245 in Malayalam, 104 in Odia, and five in Assamese. Apart from these, one each is available in Khasi and Garo languages, six in Urdu, as well as 27 in Nepali and 11 in Konkani. 

But even as the project continues, the court has ensured that each translated judgment comes with a disclaimer: “Supreme Court Registry will not be responsible for incorrect or inaccurate translation and for any error, omission or discrepancy in the content of translated text. The translation of judgments is provided for general information and shall have no legal effect for compliance or enforcement.”

The disclaimer also says that anyone using the translations should be responsible for checking the accuracy themselves with the original version, adding, “The translations are being done with the help of various human agencies and software tools. Reasonable efforts have been made to provide an accurate translation.” However, the court noted that it would be obliged to carry out corrections if errors are pointed out to it.

Justice Chandru said one way of ensuring a glossary of legal terminology in regional languages evolves is by allowing its use in court proceedings. He added, “The present move is only to showcase that the Supreme Court was alive to the need of diversity and is closer to the common man, but the fact of the matter is the common man does not even understand the legalese language used in courts.”

0 / 0
Sign in to unlock member-only benefits!
  • Access 10 free stories every month
  • Save stories to read later
  • Access to comment on every story
  • Sign-up/manage your newsletter subscriptions with a single click
  • Get notified by email for early access to discounts & offers on our products
Sign in


Comments have to be in English, and in full sentences. They cannot be abusive or personal. Please abide by our community guidelines for posting your comments.

We have migrated to a new commenting platform. If you are already a registered user of The Hindu and logged in, you may continue to engage with our articles. If you do not have an account please register and login to post comments. Users can access their older comments by logging into their accounts on Vuukle.