You can read any judgment of the Supreme Court of India for free, from the Court itself. The Supreme Court publishes its judgments and daily orders on its own website, and through the electronic Supreme Court Reports (e-SCR) it has made its reported judgments freely available in the official reported version. Since 2023 the Court has also assigned a neutral citation beginning with INSC to its judgments, which means a free copy can be cited without needing a commercial reporter's volume and page. No subscription, no login, no fee.
That is a genuinely strong position by international standards. In several common law jurisdictions the authoritative text of a judgment still sits behind a publisher's paywall. In India it does not. Anyone with a browser can read the same words the bench delivered, in the version the Court itself treats as reported, and cite them in a court filing.
So the interesting question is not whether Supreme Court judgments are free. They are. The interesting question is why lawyers still pay for research, and the answer is precise: free access solves retrieval of a judgment you can name. It does not solve discovery of a judgment you cannot. This piece maps every free route, sets out honestly what each one does badly, and explains where the remaining gap actually is.
The Free Routes, and What Each One Is For
These are not interchangeable. Each was built for a different purpose and each has a different shape of blind spot.
The Supreme Court's own judgment search
The Court's website carries a judgment and order search that lets you retrieve by party name, case number, date range, judge, and free text within the judgment. This is the primary source in the strictest sense: it is the Court publishing its own output. If you know the case, or know enough about it to narrow the field, this is the shortest path to an authoritative copy.
Its weakness is that it is a retrieval interface for a known target, not a research interface for an unknown one. Free text search across judgment bodies is keyword matching, which means it finds documents containing your words rather than documents containing your idea. If the bench that decided your point in 1994 used different vocabulary from the one you typed, the search is silent about it.
e-SCR, the reported judgments
The Supreme Court Reports is the Court's official reporter. The e-SCR initiative put those reported judgments online for free, which matters more than it might sound, because the reported version is the one that carries the official apparatus: the headnote, the catchwords, the counsel appearances, and the pagination that older citations refer to. A great deal of Indian practice is built on citing SCR, and being able to reach that version without a subscription removes a real cost from a young practice.
The limit is structural rather than a defect. Only a fraction of what the Court decides is reportable, and only reportable judgments are reported. Everything else, which is a very large volume of orders and short judgments, exists on the Court's site but not in the reports.
Neutral citation
The Court adopted a neutral citation format in 2023, and several High Courts have adopted their own. A neutral citation identifies the judgment by court, year and a running number, with paragraph numbers built into the judgment itself, so it does not depend on any publisher's volume and page. It is the single most consequential change for free legal research in India in years, because it closes the loop: you can now find a judgment for free, and cite it in a form a court accepts, without touching a commercial product at any point.
eCourts, the NJDG and case status
These answer a different question. eCourts services and the National Judicial Data Grid tell you about the procedural life of matters: what is pending, at what stage, listed when, disposed on what date. For Supreme Court practice this is how you establish whether a High Court judgment you are relying on has been carried up, and whether a special leave petition against it is pending. That is a completeness question the judgment text can never answer about itself.
Third party free aggregators
Several free services aggregate Indian judgments and layer their own search and citation linking on top. They are useful, widely used, and they are not the Court. Text may be reproduced from an earlier upload, formatting can differ, and corrections issued by the Court are not guaranteed to propagate. Use them to discover; go back to the Court's own copy to cite.
| Route | Best for | Weak on |
|---|---|---|
| Supreme Court judgment search | Retrieving an authoritative copy of a judgment you can already identify by party, date or case number | Discovery. Keyword matching finds your words, not your idea, and there is no meaningful concept ranking. |
| e-SCR | The official reported text with headnote and catchwords, and for anything cited by an SCR reference | Coverage of the non reportable output, which is a very large share of what the Court decides. |
| Neutral citation (INSC) | Citing a free copy without a commercial reporter, and pinpointing a paragraph rather than a page | Older judgments. The format applies going forward, so historical authority is still cited in reporter form. |
| eCourts and NJDG | Procedural position, pendency, whether an SLP is pending, next dates | Judgment text and reasoning. These are case management surfaces, not law reports. |
| Free aggregators | Fast lookup, convenient linking between cited cases, wide reach across courts | Provenance and currency. Text is a copy of a copy, and corrections do not always follow through. |
| Commercial reporters and platforms | Headnotes, editorial treatment, citator flags, and search built for discovery | Cost, and the fact that no publisher's coverage or classification is the official record either. |
Order, Judgment, Reported Judgment: Three Different Things
A great deal of confusion about what is and is not available for free dissolves once these three are separated.
- A daily order is what the Court passes on a listing date: issuing notice, granting or refusing interim relief, adjourning, tagging matters together, or dismissing a special leave petition in a line. These are published, they are numerous, and most are not law reports material.
- A judgment is a reasoned decision. Every one is published on the Court's website. Whether it is marked reportable is a decision recorded on the judgment itself.
- A reported judgment is one carried in a law report with editorial apparatus. The Supreme Court Reports is the official series; the well known private series exist alongside it and are not official.
The practical consequence is that a judgment being absent from the reports does not make it unavailable, unciteable or unimportant. It makes it harder to find, because the reporting layer is also the indexing layer, and that is exactly the gap discussed below.
A dismissal in limine is not an affirmance
This trips people up constantly when they read Supreme Court records. Where a special leave petition is dismissed without reasons, that dismissal does not approve the reasoning of the judgment below and does not convert a High Court's view into law declared by the Supreme Court under Article 141. Where leave is granted and the appeal is decided, the position is entirely different. Reading a one line dismissal as a Supreme Court endorsement is one of the most common research errors in Indian practice.
The Real Gap Is Discovery, Not Access
Here is the point that most articles on this subject miss. Suppose the judgment that decides your matter exists. It was delivered in 1991. It is on the Court's website, fully searchable, absolutely free. You will still not find it, because you do not know it exists and you do not know the words the bench used to describe your problem.
Free access answers the question can I read this judgment. It does not answer the question which judgment should I be reading. Research is the second question, and research is what people are actually paying for.
That gap is unusually wide in Indian case law, for reasons that are worth stating because they are structural and not going away.
Seventy years of drafting convention
A judgment from the 1950s and one from last term are both citable. They do not read alike. The concept is stable across the decades. The vocabulary is not, and keyword search only matches vocabulary.
Renumbered statutes, unrenumbered judgments
When a provision is renumbered or a code is replaced, judgments interpreting the predecessor stay good law and go on citing the old number. Search the current numbering and decades of authority quietly drop out.
Headnotes are editorial
Much research leans on headnotes and catchwords. Those are prepared by editors to varying depth and are absent from a great deal of the raw record. Searching the headnote layer searches an editor's words, not the Court's.
The forward question is unanswerable from the document
A judgment tells you every authority it relied on and nothing at all about what happened to it afterwards. Whether it has been overruled, doubted or referred is scattered across judgments delivered since.
Volume
The corpus is enormous and growing daily across the Supreme Court, 25 High Courts, hundreds of district courts and the tribunals. Free access to all of it is not the same as being able to read any meaningful fraction of it.
Text recovered from print
Anything reaching far enough back was recovered from paper. Character recognition on old print is imperfect, and exact matching is unforgiving of one misread character in the single word you searched for.
A Research Method That Costs Nothing
None of the above means free research is impossible. It means free research requires method, because the tools will not compensate for the absence of one. This sequence works and uses only free sources.
Start from the statute, not the case law
Read the provision as it stands today and as it stood on the relevant date. India Code carries central legislation and amendment history. A great many research dead ends are actually people searching case law for a proposition that a 2019 amendment settled.
Find one good judgment on the point, by any route
A commentary, a colleague, a free aggregator, or the Court's own free text search. You are not looking for the best authority yet. You are looking for an entry point that uses the correct terms of art.
Mine that judgment for vocabulary
Read how the bench phrases the question. Those phrases, not yours, are what the rest of the corpus is written in. Re-run your searches using the Court's language. This single step transforms free text search from useless to workable.
Follow the citations backwards
Every judgment names the authorities it relied on. Pull them. This gives you the line of authority as the Court itself understands it, which is more reliable than any list you assemble by keyword.
Then work forwards, deliberately
Search for later judgments that name your case. This is laborious by hand and it is the step that answers whether your authority still stands. Nothing on the face of the judgment will tell you, and no free source flags it for you.
Check the procedural position
For a High Court judgment, check on the case status surfaces whether it was carried up and whether anything is pending. For a Supreme Court judgment, check whether the question stands referred to a larger bench.
Cite from the Court's own copy
Whatever you discovered the judgment through, take the text and the citation from the Supreme Court's own publication. Where a neutral citation exists, use it, and pinpoint the paragraph rather than a page.
Free does not mean verified
This is general information about research sources, not legal advice. Anything you find on any aggregator, including this platform, is a view of what the Court published. Confirm the citation, the date, the bench and the text against the Court's own record before it goes into a pleading. The single most damaging research error is not failing to find a case. It is citing one that does not say what you think it says, or does not exist at all.
Where CourtMesh Fits, and Where It Does Not
CourtMesh covers roughly 310 million cases across the Supreme Court, all 25 High Courts, the district judiciary and tribunals, all sourced from official government portals with no intermediary in between. Because access to Indian judgments is genuinely open, that corpus is built on the same public record you can reach yourself. The platform is not selling you access to something withheld.
What it sells is the second question. Semantic retrieval finds judgments that express your idea in vocabulary you did not type, which is precisely the failure mode of free keyword search. Citation relationships let you traverse forward from a judgment to what came after it. Case status from official portals tells you whether a High Court judgment has been carried up.
And the boundaries matter as much as the capabilities. Semantic search and citation analysis cover a subset of that corpus rather than all of it, and that subset is expanding rather than complete. A clean result means the tool found nothing, not that there is nothing. The Court's record remains the authority, and no software is enrolled under the Advocates Act 1961. The duty to the court is personal and does not transfer to a tool.
The honest bottom line: if your question is whether you can practise law in India without paying for judgments, the answer is yes, and that is worth celebrating. If your question is whether you can research well without method, the answer is no, and no amount of free access changes it.
Access is solved. Finding is not.
Every Supreme Court judgment is free, official and citable, and that is a genuine achievement. What remains hard is finding the judgment whose existence you have no reason to suspect, written by a bench that described your problem in words you would never type. CourtMesh searches roughly 310 million cases from official government portals by meaning as well as by keyword, and tracks citation relationships across the subset that carries deep analysis, without pretending that subset is the whole corpus. The reading, the judgment and the citation are still yours.
Explore CourtMesh
