Defamation in Digital Era

Defamation is injury to reputation, and reputation, the older treatises insist, is a man's property — often more valuable than the rest of his estate. The digital era does not invent a new tort; it only enlarges the audience. A tweet, a Facebook post, a WhatsApp forward, a Google review or a comment on a news portal travels further and stays longer than the printed columns the classical authorities had in mind, but the cause of action remains the same defamation that the common law shaped through Youssoupoff v. M.G.M. Pictures Ltd. (1934), Cassidy v. Daily Mirror Newspapers Ltd. and D.P. Choudhary v. Manjulata (1997). What changes is the form of the statement, the route of its publication and the identity of the persons through whose hands it passes before reaching the reader.

This chapter applies the doctrinal frame of the classical law of defamation libel and slander to the online context, and sits within the broader Law of Torts notes on this site. The student should keep three working questions in mind throughout. First, when an online statement is made, is it libel or slander, and does the distinction still matter for an Indian plaintiff? Second, when a defamatory post is read by anyone other than the person defamed, is the requirement of publication satisfied, and on whose head does liability fall — author, sharer, platform, or all of them? Third, which of the three traditional defences — justification, fair comment, and privilege — survives unaltered when the statement appears on a screen rather than in a newspaper? Since defamation is one of several wrongs in the wider scheme of tort, it is useful to read this chapter alongside the chapter on negligent misstatement, which polices a related but distinct interest in careful speech.

Libel, slander and the digital form

The English law historically divided defamation into libel — representation in some permanent form, such as writing, printing, picture or effigy — and slander, the publication of a defamatory statement in a transient form, typically by spoken words or gestures. A useful test framed by the classical authorities is that libel is addressed to the eye and slander to the ear. A second test, accepted in Youssoupoff v. M.G.M. Pictures Ltd., treats the photographic part of a cinema film as libel because it is permanent matter to be seen by the eye; the speech synchronised with it forms part of one complex exhibition and is treated as ancillary. Section 1 of the (English) Defamation Act, 1952 went further and provided that broadcasting of words by means of wireless telegraphy is to be treated as publication in permanent form.

Applied to digital communication, the test is straightforward. A post on a social-media wall, a blog entry, an online news article, a comment on a forum, an entry in a discussion thread, a YouTube video description and a hosted image are each addressed to the eye and stored in a form that endures. They are libel. A live audio broadcast over a streaming service or an unrecorded voice call is, on the older test, slander — but the moment the same audio is recorded, transcribed, captioned or embedded in a permanent host, it begins to resemble libel. A live video stream that is simultaneously archived for replay is, like the cinema film in Youssoupoff, properly treated as libel: the photographic part is permanent matter to be seen by the eye. The classical line between libel and slander dissolves rather than disappears in the digital medium.

For an Indian plaintiff the line was never as sharp as in England. The weight of authority has long been to discard the distinction in India and treat both libel and slander as actionable per se without proof of special damage. Parvathi v. Mannar (1885) held that the English requirement of proof of special damage for oral defamation is founded on no reasonable basis and should not be adopted by the courts of British India. Hirabai Jehangir v. Dinshaw Edulji and A.D. Narayana Sah v. Kannamma Bai took the same line in cases of imputation of unchastity to a woman. D.P. Choudhary v. Manjulata (1997) — where a local daily wrongly reported that a 17-year-old college student had run away with a boy at night — confirmed that all defamatory words are actionable per se and general damages will be presumed. The result is that an Indian plaintiff suing over a tweet does not need to ask whether the words are libel or slander; she needs only to show defamatory content, reference, and publication. The distinction retains some academic value but no longer disposes of the action.

Statement must be defamatory

A defamatory statement is one which tends to lower the plaintiff in the estimation of right-thinking members of society generally, or which tends to make them shun or avoid him. An imputation that exposes the plaintiff to disgrace, humiliation, ridicule or contempt is defamatory. The standard is that of a right-minded citizen of fair average intelligence, not that of a special class whose values are not shared by fair-minded members of the society generally — the test taken by the courts in Mst. Ramdhara v. Mst. Phulwatibai (1969) and applied to the printed page in Capital and Counties Bank v. Henty & Sons (1882). The same test holds when the words are pixels rather than ink. A Facebook post calling a teacher dishonest, a tweet imputing fraud to a businessman, a Google review imputing professional incompetence to a doctor, or a WhatsApp forward suggesting that a young woman is of loose character — each is to be measured by the impression it leaves on the right-minded member of the audience reached.

Two further rules from the classical authorities, drawn from the same body of doctrine summarised in the chapter on libel, slander and the classical defences, must be carried over without alteration. First, mere hasty expression spoken in anger, or vulgar abuse to which no hearer would attribute any set purpose to injure character, is not actionable, but words that ridicule and humiliate are. The general rule from Parvathi v. Mannar and Venkata Surya Rao v. Nandipati Muthayya — that mere insult does not give rise to an action — applies equally to coarse abuse on a comment thread. Second, what looks like a compliment may be defamatory in context. The famous illustration from S.N.M. Abdi v. Prafulla Kumar Mohanta (2002), where allegations of misuse of muscle power against a deposed Chief Minister were held defamatory and damages of five lakh rupees awarded, shows that the court reads the publication as a whole and locates the imputation in the impression it leaves upon a substantial respectable group of readers, even if that group is a minority of the total audience.

The innuendo and the digital context

An online statement is often defamatory not on its face but through innuendo — a latent or secondary meaning derived from context, captioning, hyperlinks, hashtags or pre-existing rumour. The classical learning on innuendo applies almost word for word. Capital and Counties Bank v. Henty & Sons taught that the words of a circular taken in their natural sense may be innocent and the supposed imputation may not be reasonably understood by readers. Tolley v. J.S. Fry & Sons Ltd. (1931) taught that an apparently innocent advertisement — there, a caricature of an amateur golf champion next to the defendants' chocolate — may, on the innuendo proved, become defamatory because the audience is led to believe that the plaintiff has prostituted his amateur status for advertising gain.

Translate these holdings to the screen. An online image showing a public figure standing next to a celebrity, with a caption announcing their engagement, repeats the structural facts of Cassidy v. Daily Mirror Newspapers Ltd. — where a photograph captioned with the announcement of an engagement was held defamatory of the man's existing wife by innuendo, the obvious innocence of the publishers being no defence. A meme that pairs the plaintiff's image with text that seems flattering but is understood by the in-group as denigrating may be the modern Tolley. The plaintiff must plead and prove the latent meaning; once proved, the apparently innocuous post becomes the foundation of a libel suit. The audit of meaning is no different on screen than on paper, but the plaintiff must show that the audience he points to actually understood the secondary meaning.

Reference to the plaintiff

The plaintiff must show that the statement was understood as referring to him. Intention of the writer is immaterial; the test, drawn from Hulton & Co. v. Jones and confirmed by Lord Loreburn in the House of Lords, is whether ordinary sensible readers, knowing the plaintiff, would form the opinion that the statement referred to him. A fictitious name is no shield: in Hulton, the imaginary 'Artemus Jones, Churchwarden at Pekham' caught a real barrister of that name. Newstead v. London Express Newspapers Ltd. (1939) caught a Camberwell barber when the report concerned a Camberwell barman of the same name. The Indian view, taken in T.V. Ramasubha Iyer v. A.M.A. Mohindeen, is that liability for innocent publication is limited — the Madras High Court held that the rule in Hulton was contrary to justice, equity and good conscience in India, especially after the (English) Defamation Act, 1952 had itself altered the position in England.

The digital twist is identification by handle, screenshot, tag or context. A tweet that names no one but reproduces the plaintiff's photograph identifies him. A blog post that hyperlinks the plaintiff's profile, or that contains enough biographical detail to make the subject identifiable to ordinary followers, satisfies the requirement of reference even though the legal name is absent. Where the words refer to a class of persons — "all lawyers in city X are fixers", "every revenue clerk in this taluka takes a bribe" — the rule from Knupffer v. London Express Newspapers Ltd. applies: no individual member can sue unless the words could reasonably be considered to refer to him in particular. The same constraint that defeated the appellant in Knupffer defeats the social-media plaintiff who points to a generalised online attack on a profession or community without showing pointed reference.

Publication on the open web

Publication is the third essential. It means making the defamatory matter known to some person other than the person defamed. The classical authorities collected several propositions on this point. Communication to the plaintiff himself is not publication; defamation is injury to reputation, and reputation consists in what others think of him. Dictating a defamatory letter to one's typist is publication, because the typist is a third person. A defamatory postcard or telegram is publication even without proof that someone read it, because the postal officials in Williamson v. Freer (1874) handle the message and a postcard is highly likely to be read in transit. If the defamatory letter is written in a language the plaintiff does not understand, and this is known to the writer, publication arises when the letter is read by a third person — the rule in Mahendra Ram v. Harnandan Prasad (AIR 1958). The classical learning thus already turned on the foreseeable audience and the likelihood of third-person access — concepts that map cleanly onto digital communication.

On the open web, publication is in most cases obvious. A tweet, a public Facebook post, a comment on a news article and an open Telegram channel are each addressed to a class of readers larger than the plaintiff alone. The act of pressing 'post' is the equivalent of dictating to the typist multiplied by the number of followers. A direct message, by contrast, that travels only between the defendant and the plaintiff resembles the sealed letter in Powell v. Gelston (1916) or Huth v. Huth (1915) — there is no publication by the defendant unless he ought reasonably to have foreseen that a third person would read it, an inquiry close in spirit to the standard of foreseeability discussed in the wider law of general defences such as volenti non fit injuria. The principle from Theaker v. Richardson (1962) — that there is publication when the defendant knew or ought to have known that the letter, although sent to the plaintiff, would be read by a third person — applies with full force when a private message is sent through a shared device or a family group.

Repetition, retweet and the fresh cause of action

The liability of the person who repeats a defamatory matter arises in the same way as that of the originator, because every repetition is a fresh publication giving rise to a fresh cause of action. The classical authorities applied this rule to the editor, printer and publisher of a newspaper, all of whom were strictly liable along with the author. The rule applies as cleanly to the share, retweet, repost, and forward. Each new dissemination is itself a publication. A user who retweets a defamatory tweet to her own followers makes the imputation known to a fresh group of readers; a WhatsApp user who forwards a defamatory message to a family or work group does the same. The plaintiff acquires a new cause of action against each fresh publisher, subject only to the established defences and to the lenient rule for innocent disseminators discussed below. Where multiple defendants are joined — author, editor, sharer and host — apportionment of liability proceeds on the principles governing joint and several tortfeasors.

Two complications deserve mention. First, the principle of repetition does not transform an isolated comment into mass publication unless someone actually transmits it further; mere accessibility is not the same as publication. Second, the originator continues to be liable for the original act and may also be liable, by ordinary principles of foreseeability, for the foreseeable repetition by others. Where the message is structured in a form designed to encourage forwarding — a viral graphic, a meme, a captioned video — the original author cannot easily disclaim responsibility for the spread that follows.

Innocent dissemination — booksellers, librarians, intermediaries

The classical authorities recognised a class of persons — booksellers, newspaper-vendors, librarians — who disseminate defamatory matter without knowing its content. The law adopted a lenient attitude toward them. They are not liable if they did not know, and could not, by the exercise of reasonable diligence, have known, that what they were circulating was defamatory. The leading authority is Emmens v. Pottle, where large-scale news-vendors who sold copies of a publication containing libellous matter were held not liable because they neither knew nor were negligent in not knowing the defamatory content. The standard is twofold — actual knowledge or constructive knowledge by reasonable diligence.

This branch of doctrine is the natural home for the online intermediary. A platform that hosts user-generated content, a search engine that indexes a third party's defamatory page, an internet service provider that carries packets without inspecting them, and an online marketplace that lists a defamatory review are all, in functional terms, modern booksellers. The classical test asks whether the intermediary knew of the defamation or should reasonably have known. Until knowledge is imputed — typically by a complaint or notice — the intermediary stands in the position of Emmens and is not liable. After knowledge, the position alters: the intermediary that fails to take reasonable steps to remove or disable access to the defamatory content begins to look like the editor in Gurbachan Singh v. Babu Ram (AIR 1969), who could not claim indemnity from his news supplier and who, on the classical reasoning, was expected to check the news before publishing it. The law is content with a calibrated standard: passive carriage attracts the lenient rule, active editorial selection or knowing retention attracts ordinary publisher liability. The rule itself has not changed since the nineteenth century; only its field of application has expanded.

Defences — justification, fair comment, privilege

The three defences open to a defendant in a defamation action are justification of truth, fair comment and privilege. The defences are unaffected by the medium of publication, but the digital context puts each of them under fresh pressure.

Truth. In a civil action for defamation, truth of the defamatory matter is a complete defence. The reason, in the classical formulation, is that the law will not permit a person to recover damages in respect of an injury to a character which he either does not or ought not to possess. The defence is available even though the publication is made maliciously, and it survives minor inaccuracies if the statement is substantially true — the principle of Alexander v. North Eastern Ry., where a notice that overstated the alternative term of imprisonment from fourteen days to three weeks was held substantially accurate and not actionable. Radheshyam Tiwari v. Eknath (AIR 1985) is the warning. The editor and publisher of a Marathi weekly had alleged corruption and bribery against a Block Development Officer; unable to prove the truth of those facts, he lost the defence of justification and the related defences too. An online publisher who alleges fraud on Twitter and cannot prove it stands in the same position as Tiwari.

Fair comment. Fair comment on a matter of public interest is the second defence. Three essentials must be satisfied: the words must be a comment, that is, an expression of opinion rather than an assertion of fact; the comment must be fair; and the matter commented on must be of public interest. The boundary between comment and fact, on which Silkin v. Beaverbrook Newspapers Ltd. (1958) lays so much stress, is delicate online — a tweet that begins as opinion can slide into a factual allegation when it names dates, sums, and persons. The test of fairness is the honest opinion of the commentator, not the tribunal's own view: as Diplock J said in Silkin, the question is whether it was an opinion, however exaggerated, obstinate or prejudiced, which was honestly held by the writer. Malice destroys the defence — Gregory v. Duke of Brunswick (1843) shows that hooting actuated by malice is not fair comment on a performance; equally, a coordinated online campaign to bring down a public figure does not become fair comment merely because the underlying subject is of public interest. R.K. Karanjia v. Thackersey (AIR 1970) is the clearest Indian authority — a comment based upon invented and untrue facts is not fair, and the Bombay High Court rejected fair comment along with the other defences.

Privilege. The third defence is privilege, absolute or qualified. Absolute privilege attaches to parliamentary proceedings under Article 105(2) and to the corresponding state-legislature privilege under Article 194(2), to judicial proceedings, and to certain state communications. The privilege follows the occasion, not the medium. A statement made in Parliament is absolutely privileged whether reported on a printed page, a television broadcast or a livestream. Qualified privilege attaches to statements made on a privileged occasion — in discharge of a duty, legal, social or moral, or in the protection of an interest — and to fair reports of parliamentary, judicial or other public proceedings, provided in each case the statement is made without malice. The reciprocity of duty and interest, articulated in the classical learning, must actually exist; Chapman v. Lord Ellesmere (1932) confirms that the privilege protects the publication only to those who have an interest in it, not to the world at large. The challenge in the digital era is that platforms make every statement potentially publishable to the world; a publication that would have been privileged when sent to a circle of professional colleagues may lose privilege when posted on an open feed. The Bombay High Court in R.K. Karanjia v. Thackersey rejected the newspaper's claim of qualified privilege precisely because, beyond the matter being one of general public interest, no duty to communicate to the general public could be spelt out — and because the publication was tainted by malice. Online publishers must read that pair of holdings together: the duty must be there, and the publication must be without malice.

Injunction and take-down

The classical authorities accept that a court may restrain by injunction the publication of defamatory statements. Prameela Ravindran v. P. Lakshmikutty Amma is illustrative — the Madras High Court restrained the respondent from sending letters of a defamatory kind to various persons concerning the validity of a marriage, on prima facie evidence and balance of convenience. The restraint was an injunction against publication; it was not contingent on the medium being a letter rather than a notice or a poster. The same remedy is available in respect of online publications. A court that is satisfied that the statement is defamatory and that the balance of convenience lies with the plaintiff may direct the defendant to remove the post and restrain him from making further statements of a similar kind. Take-down, in this sense, is the digital form of the classical injunction restraining further publication, and the remedies on the civil side are part of the wider doctrine considered in the chapter on remedies judicial and extra judicial. The relief is discretionary and is normally calibrated, like the classical injunction, to restrain only the defamatory matter and not the wider freedom of speech.

Defamation of a class, the deceased and intra-spousal communication

Three rules carry over without alteration. First, when words refer to a group of individuals or a class of persons, no member may sue unless he can show that the words could reasonably be considered to refer to him in particular, the rule from Knupffer and applied in India in Dhirendra Nath Sen v. Rajat Kanti Bhadra. The online attack on "all officers of department X" or "every member of community Y" therefore generally does not give an individual member a cause of action. Fanu v. Malcolmson, by contrast, held that where a statement though general can reasonably be considered to refer to a particular plaintiff — there, the Waterford factory of the plaintiff — the action will succeed. Second, defaming a deceased person is no tort, although it may amount to a criminal offence under the relevant penal provision if the imputation harms the reputation of the deceased and is intended to be hurtful to the feelings of the family or near relatives. Third, communication of defamatory matter from husband to wife or vice versa is no publication, the eyes of law treating them as one person — but communication of matter defamatory of one spouse to the other is sufficient publication, the rule in Theaker v. Richardson. The Supreme Court in T.J. Ponnen v. M.C. Verghese reversed the Kerala High Court and held that where the defamatory letters from a husband to his wife had fallen into the hands of the father-in-law, the letters could be proved by other means notwithstanding Section 122 of the Indian Evidence Act. The principle survives in the digital age — a defamatory message sent to one spouse about the other, when in fact it is read by a third person on the same device, will be treated as published by the defendant if a reasonable defendant ought to have foreseen the third-person reading.

Damages and the digital amplifier

Damages in defamation are presumed where the statement is libellous or otherwise actionable per se. D.P. Choudhary v. Manjulata awarded ten thousand rupees as general damages against a local daily; S.N.M. Abdi v. Prafulla Kumar Mohanta awarded five lakh rupees in respect of an article in a national weekly; R.K. Karanjia v. Thackersey reduced damages on appeal from three lakh rupees to one and a half lakh, but confirmed that defamation calls for substantial pecuniary relief where the imputation is grave. The wider law of damages general special nominal and exemplary supplies the framework: general damages compensate the presumed injury to reputation; special damages, where pleaded and proved, compensate identifiable financial loss such as lost contracts or lost employment; exemplary damages mark the court's disapproval of conduct that is malicious or oppressive. The digital era does not alter the heads of damage, but it does affect quantum. The reach of an online publication is often much wider than that of a small-circulation print piece; the durability of the digital trace, indexed by search engines and resurfaced years later, is greater; the potential for re-publication by share or retweet multiplies the original injury, a problem that the older treatises on discharge of torts never had to contemplate at scale. The court will take the digital amplifier into account when setting the figure, just as the older courts took into account the circulation of the offending newspaper.

Defamation across torts — siblings to study together

Defamation occupies its own corner of the law of torts but it sits next to several allied wrongs that the student should hold in view. The intentional torts to the person — trespass to person, assault, battery and false imprisonment — protect bodily integrity and liberty; defamation protects the social self. Nuisance, public and private and trespass to land and trespass to goods protect interests in land and chattels; defamation protects the standing of the person in the eyes of the community. The tort of careless misstatement, dealt with in its own chapter, protects against words negligently uttered which cause financial loss to a person who relies upon them; defamation, by contrast, requires no reliance and protects reputation as such. Within the same chapter family, the foundational article on defamation libel slander and defences sets out the doctrinal core; this article extends the same frame to the digital medium without altering its underlying structure.

General defences and capacity

The three classical defences to defamation — justification, fair comment, privilege — are specific to the tort. They sit alongside the cross-cutting defences in the wider law of torts, including private defence. Statutory authority may, in particular cases, bar an action; inevitable accident rarely arises in defamation since the action is on the imputation, not on physical harm. Vicarious considerations are present whenever the defamatory post is made by an employee in the course of employment — the employer may be answerable on principles of vicarious liability of master and servant; the State, through its officers, may be answerable in cases that engage the principles of the vicarious liability of the State, especially where official communications are involved and the question is whether the occasion was absolutely or qualifiedly privileged. Capacity rules — the position of minors, persons of unsound mind and corporations as plaintiffs and defendants — apply to defamation as they apply to other torts, and are gathered in the chapter on capacity in tort. Liability among multiple defendants — author, editor, sharer, platform — engages the principles of joint and several tort liability already cross-referenced above.

Jurisdiction, conflict of laws and the borderless post

An online publication is, in principle, available wherever there is an internet connection. A defamatory tweet posted from one country may be read by readers in many. The classical law of defamation presupposes that publication occurs at the place where the words are read — the postcard is published where it is read by the postal officials, the letter where it is opened by the third person. A statement posted online is therefore published, in this older sense, in every jurisdiction where it is read. For the Indian plaintiff, this means that a court at the place where the post was downloaded and read, and the plaintiff's reputation injured, may have territorial jurisdiction; for the foreign defendant, it means that the act of posting may attract liability in jurisdictions far removed from the server. The courts must balance the openness of the internet against the territorial limits of process, and they typically do so by inquiring whether the publication had a real and substantial connection with the forum. The conflict-of-laws layer is beyond this chapter, but the principle that publication occurs where the matter is read is a direct extension of the classical authorities on the postcard and the telegram.

Reading the digital case — a checklist

For a problem question or a practical opinion on an online defamation matter, the student should work through the same checklist that the classical authorities require. First, identify the defamatory imputation and ask whether it is defamatory by the standard of the right-thinking member of society generally, or only by some innuendo that the plaintiff must plead and prove. Second, ask whether the statement refers to the plaintiff in the eyes of ordinary readers who know him, including by handle, photograph or biographical detail, and screen out attacks on undefined large classes by the rule in Knupffer. Third, identify the publication: who posted, who shared, who retweeted, who archived; each is a fresh act of publication and each may attract a fresh cause of action, subject to the lenient rule for innocent disseminators on the model of Emmens v. Pottle. Fourth, run the three defences: justification, fair comment, privilege. Fifth, consider remedies — damages presumed, special damage where pleaded, and injunction or take-down where appropriate, on the model of Prameela Ravindran. The full set of background ideas is gathered together in our wider tort syllabus, and the student is encouraged to read this chapter alongside the foundational article on the tort, the chapter on remedies, and the chapter on damages.

Conclusion

Defamation in the digital era is not a new tort. It is the same tort, applied to a faster and broader medium. The three essentials — defamatory content, reference to the plaintiff, publication to a third person — remain the controlling concepts, drawn from Cassidy, Hulton, Newstead, and the long line of Indian authorities running through Manjulata and S.N.M. Abdi. The three defences — justification, fair comment, privilege — remain the controlling answers, controlled by the statements of principle in Alexander, Silkin, and Karanjia. The lenient rule for innocent disseminators in Emmens v. Pottle is the natural home for the platform that hosts content without knowledge. The injunction in Prameela Ravindran is the natural ancestor of the take-down order. The exam-aspirant who has internalised the classical learning has, on this view, also internalised the digital. Practice will refine the application; doctrine has already supplied the frame.

Frequently asked questions

Is an online post libel or slander, and does it matter for an Indian plaintiff?

An online post is libel because it is in a permanent form addressed to the eye, on the test taken in Youssoupoff v. M.G.M. Pictures Ltd. that treats the photographic part of a cinema film as libel. Section 1 of the (English) Defamation Act, 1952 treated even broadcast words as publication in permanent form. For an Indian plaintiff, the distinction is largely academic. The weight of authority since Parvathi v. Mannar (1885) and confirmed in D.P. Choudhary v. Manjulata (1997) is that both libel and slander are actionable per se without proof of special damage. The plaintiff need only show defamatory content, reference and publication.

Is retweeting or forwarding a defamatory message itself a fresh publication?

Yes. The classical rule that every repetition is a fresh publication giving rise to a fresh cause of action applies without alteration to the digital medium. Each retweet, share, repost or forward is a new act of publication, and the person who repeats the matter is liable in the same way as the original author. Liability is strict for the editor, printer or publisher of a defamatory matter on the older authorities, and the same logic applies to the user who consciously amplifies the post. The only safe harbour is the lenient rule for those who disseminate without knowledge and could not, by reasonable diligence, have known of the defamatory content.

When is an online platform liable for user-generated defamatory content?

On the classical principle from Emmens v. Pottle, large-scale disseminators who neither knew nor were negligent in not knowing the defamatory content of what they distributed were not liable. Applied to the platform, the test asks whether the intermediary had actual knowledge or constructive knowledge by reasonable diligence. Until knowledge is imputed, typically by a complaint or notice, the platform is in the position of the innocent newsvendor. After knowledge, failure to take reasonable steps to remove or disable access to the defamatory content engages ordinary publisher liability, on the reasoning of Gurbachan Singh v. Babu Ram (AIR 1969).

Can I get a court order to take down a defamatory post?

Yes. The Madras High Court in Prameela Ravindran v. P. Lakshmikutty Amma restrained the respondent by injunction from sending letters of a defamatory kind, on prima facie evidence and balance of convenience. The injunction is not tied to a particular medium and is available against online publications too. A court satisfied that the impugned statement is defamatory and that the balance of convenience lies with the plaintiff may direct removal of the post and restrain further publication. The relief is discretionary and is normally calibrated to restrain only the defamatory matter and not the broader freedom of speech.

Does the defence of fair comment protect a critical online review?

Fair comment protects a critical online review only if three conditions are satisfied. First, the statement must be a comment, an expression of opinion rather than an assertion of fact. Second, the comment must be fair, that is, an opinion honestly held by the writer, however exaggerated, obstinate or prejudiced, on the test of Silkin v. Beaverbrook Newspapers Ltd. (1958). Third, the matter must be of public interest. A comment based upon invented or untrue facts is not fair, the principle of R.K. Karanjia v. Thackersey (AIR 1970). Malice on the part of the writer destroys the defence.

Where can the plaintiff sue when the defamatory post is read across many places?

The classical authorities on the postcard and telegram place publication where the matter is read. By extension, an online publication is published in every jurisdiction where it is downloaded and read, and the plaintiff's reputation is injured. An Indian court at the place where the post was read and the reputation injured may have territorial jurisdiction. Conflict-of-laws considerations apply to foreign defendants, and the courts typically inquire whether the publication had a real and substantial connection with the forum. The principle, however, is a direct extension of the older publication rule, not a departure from it.