Who Can Hear the Constitution Bench Speak?

Introduction

The Supreme Court of India has always been critiqued as an inaccessible forum, reachable only for the privileged. The location of the Court in the capital city of New Delhi has always added to this inaccessibility. The 229^th Law Commission Report had noted that the Court was geographically inaccessible to litigants residing in distant and remote areas. In addition, the contemporary continuation of certain colonial practices has rendered courtroom spaces inaccessible. Based on an ethnographic study, Khorakiwala observes how the architecture of these courts creates issues of access to justice. Furthermore, a recent account looking at a decade of original empirical research has indicated that a select group of senior advocates attempts to exercise disproportionate influence on the outcome of cases. These cases themselves are also judged by an ‘old boys’ network.’ Hence, the Indian Supreme Court has been considered an isolated, unreachable, and distant temple of justice.

While these concerns exist both geographically and spatially, the Court has taken several steps to ensure that its decisions and proceedings are accessible. In the case of Swapnil Tripathi, the Supreme Court symbolically opened its doors and allowed judicial proceedings to be electronically live-streamed. It was observed that such a measure would make both the law and justice accessible to those who seek them. Moreover, the Court began translating its decisions into vernacular languages. According to official figures, 36,000 (approximately) Supreme Court judgments were translated into Hindi and 42,000 (approximately) judgments in the other 18 regional languages. The Court also undertook the task of transcribing oral arguments for the constitution bench judgments. This would not only act as a permanent record of arguments but also enable people to understand the complex arguments made by their lawyers. While prima facie, these steps have attempted to solve the accessibility problem, it is imperative to take a step back and examine the contemporary effectiveness of these measures to ensure access to justice.

Methodology

In this piece, I examine these measures through empirical data (Diagram 1.1, appended below) collected for the recent thirty-four constitution bench judgments which were rendered during the tenures of Chief Justices Lalit and Chandrachud. This data pertains to the number of total pages of each constitution bench judgment, the number of pages of the concurring judgment, the number of pages spent on facts, the number of pages spent on arguments, the availability of transcripts and the translation of these judgments into vernacular languages. This data was collected by me along with the helpful assistance of Nivedhitha K, Niyati Dhiman, and Syed Sidra Ahmed.

Before I proceed with the roadmap of the blog, it is imperative to discuss the methodology related to the collection of this data. First, Constitution bench judgments have been used here since they deal with matters of substantial legal importance and, in a way, affect a larger number of litigants, people, and cases. Thus, data from these benches would be a perfect indicator in assessing the wide-scale consequences of these measures. Second, some initiatives have been made available only for these benches (such as transcripts for arguments), and accordingly, this data is employed here. Third, judgments during the tenure of Chief Justices Lalit and Chandrachud have been used, as before this, these benches were a rarity and a substantial number of benches were only constituted during this specific period. This data has been collected from the Supreme Court of India website (including the e-SCR database or standalone databases). These figures are updated till 26 May 2025.

In the first part of this blog, I evaluate the practice of translating judgments into vernacular languages and highlight the deplorable state of affairs. In the second section, I examine the transcribing measure in order to test its effectiveness vis-à-vis accessibility. Lastly, I would navigate the issues caused by the lengthy nature of these judgments.

Court’s Tongue or Mother’s Tongue? The Translation Conundrum

The Supreme Court had announced that judgments will be translated into 18 regional languages – Assamese, Bengali, Garo, Gujarati, Hindi, Kannada, Kashmiri, Khasi, Konkani, Malayalam, Marathi, Nepali, Odia, Punjabi, Santali, Tamil, Telugu, and Urdu. While it was announced that judgments would initially focus on Hindi, Tamil, Gujarati, and Odia, it can be seen that most of the judgments have been translated mainly into Hindi. While that act in itself is worthy of critique, it is to be wondered why the Supreme Court chose not to translate their judgments into all the 22 scheduled languages. The Supreme Court has conveniently left out Bodo, Dongri, Maithili, Manipuri, Sanskrit, and Sindhi and instead added Garo and Khasi. It becomes imperative to question why certain languages were left out and why some of them were deemed worthy. Hence, it would have been more appropriate to consider the adoption of scheduled official languages (in a phased, stage-wise manner) or provide objective criteria on the exclusion of some of these languages. While it has been reported that the translation project has come to a halt, the data from the constitution benches reveal a grim picture as well. Out of the 34 judgments, only 13 of them have been translated. However, not all 13 have been translated into all chosen languages. Hindi and Punjabi have been the most chosen language depicting the highest translations i.e., 11 times. This is followed by Gujarati (4 times), and then Odia, Telugu, and Marathi that just feature once. A graphical description for easier understanding has been showcased below:

This clearly indicates that not only are fewer judgments being translated, but also that the translations are being conducted for very few languages. This causes a grave concern since access to the judgments of the constitution bench has been restricted mainly to those who can read English, Hindi, or Punjabi. It can be seen that the beneficiaries most affected by some of the judgments do not have an opportunity to understand a judgment in their regional language. For instance, the Article 370 judgment has not been translated into Kashmiri or Dogri, which is likely to accentuate prevalent inaccessibility. Hence, it becomes imperative that the translation project is conducted in a manner where all constitution bench judgments are translated into all languages in due course in order to increase access to justice.

Court of Record or Court of Complexity? The Transcribing Issue

It is common knowledge that judgments of the Supreme Court are complex to understand for even people with a legal bent of mind. Hence, for any ordinary person, the complexity rises manifold. This could be attributed to the extensive usage of legal terminologies, Latin phrases, complex structures, or incoherent sentence formations. Furthermore, the arguments made by lawyers add another level of complexity since they tend to be unnecessarily lengthy, verbose, and technical. While some technical legal arguments might be contended to be necessary for the case, the true art of advocacy is to make your pleadings in a simple, rational and accessible manner. In order to make the arguments accessible in the first place, the Indian Supreme Court decided to start transcribing them using artificial intelligence. However, several issues plague such an initiative.

I argue that we need to question whether the availability of transcripts makes constitution bench judgments accessible in the first place. The transcribing measure only publicises the copy of the lawyers’ arguments, but the legal arguments themselves are dense and inaccessible. A substantive perusal of the transcripts showcases that the arguments made by lawyers in these cases are extremely complex, verbose, and difficult to follow for any reasonable reader. Therefore, mere availability of legal material cannot be equated to substantive accessibility of the law to the public.

 The observations from the dataset highlight that not all 19 judgments have the complete set of transcriptions. Out of the existing 19 – 14 judgments have the complete set, 3 judgments have no arguments which were recorded at all, and 2 judgments have an incomplete record  (out of the 34 judgments, only 19 of the decisions have the transcripts of the arguments available since the decision to transcribe was taken after the arguments in 15 of the benches were already over). This indicates that even out of the few decisions whose arguments are being recorded, there seems to be no complete record of all of them. The absence of complete transcripts highlights not only a lackadaisical attitude adopted by the judicial administration but also the ineffectiveness of such a questionable accessibility measure. In order to ensure that the transcripts are an effective accessibility measure, I propose that first, the transcripts for all constitution bench cases must be made available at the earliest; second, the transcripts must be designed in an easy to understand manner; and third, active steps must be taken to ensure that complex legal arguments made by lawyers are deconstructed and simplified.

Does Size Always Matter?: The Length Problem

In the case of Ajit Mohan v. Legislative Assembly, Delhi, the Supreme Court stated that it is the need of the hour to not only write clear judgments but also shorter decisions, which would enable the litigants and the people to understand. While such an observation is laudatory, it is imperative to question if the Supreme Court itself writes clear, short judgments. In this piece, I will refrain from critiquing the lack of clarity in the judgments and focus on the length of the judgments. A perusal of the data from the constitution benches indicates the continuing trend of writing lengthy judgments.

The judgments of the constitution benches have been, on average, an alarming 210 pages long. It could very well be counter-argued that constitution benches deal with substantial and complex questions of law and thus, must be lengthier. However, the length of these judgments is not due to the observations and analysis but rather because of the reiteration of facts and arguments. In the case of both Joseph Shine and RR Kishore, almost half of the judgment merely summarises facts and arguments. In the Aligarh Muslim University case,  the facts and arguments stretch for 33 and 32 pages respectively! Moreover, concurring judgments also have pages dedicated to detailed facts and arguments. In the case of Association for Democratic Reforms v. Union of India, Justice Khanna’s concurring judgment uses 12 pages to reiterate facts even when the majority opinion by Justice Chandrachud already mentions the same facts over 19 pages.

Since lengthy judgments take away from making the decision accessible, shorter judgments must be the norm. This will ensure that in multilingual and stratified societies like India, large segments of the population, especially those who are not fluent in legal English or did not have access to formal education, can still understand the governing laws of the land. While detail and reasoning are essential in judicial writing, clarity and conciseness are critical to making justice truly accessible.

Conclusion

This piece is an attempt to test the viability of accessibility measures adopted by the Supreme Court, i.e., translation of judgments and transcription of arguments. The analysis of empirical data from constitution bench judgments shows that these steps have been similar to installing a ramp to a library where all the books continue to be in Latin. Moreover, the practice of writing lengthier decisions has been critiqued from an access to justice perspective. The failure of the Court to effectively maintain and uphold institutional accessibility measures indicates the absence of administrative will to ensure that the doors of the Supreme Court are truly open for all. Even if it could be argued that the door has been kept open, it is really difficult to make sense of what is going on inside. While the Supreme Court’s measures have been a step in the correct direction, it’s a bit like switching from cursive to calligraphy — prettier, perhaps, but still hard to follow. It is unfortunate to conclude that only the elite – regionally, linguistically, and socially – continue to hear the constitution bench speak.

Data from the 34 Constitution Bench Judgments:

 Diagram 1.1. (Table of all 34 Constitution Bench Judgments and necessary details as on 25 May 2025)