While reading the historical debates leading to the passing of the first post-independence Copyright Act in India, I came across some interesting passages that are relevant to a debate on a theme that seems to have lost its place in Indian Copyright jurisprudence – the Reversionary Right.
On 14 May 1957, during the Rajya Sabha’s debate on the Bill that became the Copyright Act, Professor A. R. Wadia recorded his surprise that clause 18 of the Bill as introduced in 1955 had been removed. The clause would have entitled an author who had assigned his copyright to have it reassigned to him, between the seventh and the tenth year after the assignment, on repaying what he had received. The Government’s note had described it as a safeguard for authors who, on account of poverty or other reasons, part with copyright for a meagre amount and lose the fruits of their intellectual labour. Wadia was told that the Joint Committee had deleted the clause at the instance of the authors who sat on it. He cited a Marathi work sold outright for forty rupees that had run to twenty editions, the publisher enriched and the author left with nothing. The clause had a long history, and so did the argument that defeated it, and the history is the subject of what follows.
The right at issue is the reversionary right, in its modern form the right to terminate a transfer. It limits the transferability of copyright, the feature that allows the right to be assigned in gross and to circulate apart from its author. Reversion returns to the author, after a period, a right she may have assigned when young and unknown, for a fraction of the work’s eventual value.
The transferable right and the author who sells it
Copyright in its modern form is a transferable right to exclude, and the transferability is central to its commercial use. The same transferability exposes the author, who is the party that must sell, and sells at the outset, before the value of the work is known. A publisher who acquires many works sets the failures against the successes, however, an author with one work cannot. Reversion answers it by giving the author a second occasion to bargain once the value is known. Hanoch Dagan and Molly Shaffer Van Houweling identify a second rationale, the author’s autonomy, which the reverter serves by releasing the author from a transfer made by a younger and poorer self.
An argument as old as copyright itself
The case for reversion was old before India met it. In 1710, the journal – The Observator wrote on the upcoming Statute of Anne which was assumed to be “design’d as a Kindness to us Authors” that:
“That the Bookseller shall have a Property in the Copy only for a limited Time, after which it shall revert to the Author or his Assignees. This they say will be an Encouragement to Learning, and a Security to Authors against being ill-treated or impos’d upon by Booksellers, who run away with the Profits of their Labours …; so that Authors not being able to foresee this, because Copies are like Ships put to Sea, whose prosperous or unfortunate Voyage is not to be foreseen, they have nothing more than their first Copy-Money, let the Book sell ever so well.”
The Statute of Anne of 1710 acted on the thought and gave the author a second fourteen-year term if he survived the first, returning the right to him rather than the bookseller who had bought the manuscript. The Anne reversion right failed because the author could assign the second term before it began, and the booksellers required him to do so, so the term passed to the publisher with the first. A Bill drafted in part by Jonathan Swift in 1737 would have cured the failure by making the reverter unassignable, and it did not pass.
The United States kept a reversion-like renewal term through its 1909 Act, and it failed for the same reason, because the courts let the author assign the renewal before it vested, so the publisher took it with the original term. The 1976 Act replaced the renewal with an inalienable termination of transfers, on the rationale the eighteenth century had stated, the impossibility of valuing a work before it is exploited. The argument against reversion was as constant, that the prospect of losing the work deters the disseminator from investing in it.
The reversionary right India inherited and abandoned
India came to independence holding a stronger reversionary right than the Statute of Anne had given. The Imperial Copyright Act of 1911, applied to India by the Indian Copyright Act of 1914, provided in the proviso to section 5(2) that an author who was the first owner could not, by an assignment made otherwise than by will, vest in the assignee any right beyond twenty-five years from his death, that the reversionary interest in those years devolved on his legal representatives notwithstanding any agreement to the contrary, and that any agreement to dispose of it was null and void. The inalienability met the defect that had defeated the Anne reversionary right. The flaw of the 1911 right was that it vested in the author’s estate, so the executors could sell it to meet his debts, and the family it protected held it for a short time or not at all, its benefits described in the practitioner works as illusory.
Clause 18 of the 1955 Bill on Copyright in independent India would have supplied reversionary right that the author could exercise himself between the seventh and tenth year of the assignment, so that in 1955 India was positioned to hold a stronger reversionary right that the UK one.
Two years later it held neither. The 1955 bill debates reproduced the older argument on both sides.
One member supported reversion with the example of a novel refused by seven publishers that succeeded once it appeared, and proposed that literary contracts run for a limited period so that no absolute right would be given to the publishers to exploit the poverty of the authors. P. N. Sapru defended the author on the ground that he would not have social institutions founded on the acquisitive instinct, and that the writer’s motive is not reducible to the pound, shilling and pence he will earn. The Minister piloting the Bill, K. L. Shrimali, put the case that prevailed, the case the booksellers had stated against Swift in 1737. The reassignment clause, he said, would work against the authors’ own interests, because a publisher who makes an investment, spends on advertisement, and knows he will see no return after seven years will not push the book. Business is business, he told the House, and to kill the goose that lays the golden egg leaves no goose.
The Joint Committee deleted clause 18 at the instance of the author-members, the fact that surprised Wadia. The 1957 Act dropped the inherited section 5(2) as well, and adopted a freedom-of-contract model under which an author may assign for the whole term, subject to a discretionary revocation under section 19A hedged by a five-year embargo. India refused the reverter twice at one sitting, the inherited form and the proposed one.
The human cost was set out by Prof. Basheer where he cited Ghulam Mohammed, the composer of Pakeezah, who died in penury while his works earned lakhs for the producer who marketed them, and the widow of the music director Khemchand Prakash, found begging in Mumbai after his death, and stated that artists were required to assign entire copyrights to producers for a lump sum while their works earned crores. Had the reverter survived, the composer could have recovered the assigned song after seven years and shared in its value. Without it the song remained the producer’s for the full term.
The judicial substitute and its limits
Without a statutory reverter, the exploited author was left to the courts and the law of contract, and the fullest body of that law is the English music litigation of the late twentieth century. In Schroeder Music Publishing v Macaulay the House of Lords in 1974 refused to enforce a publishing agreement against an unknown songwriter who had assigned the worldwide copyright in everything he might write over five or ten years for fifty pounds and a contingent royalty, the publisher undertaking no duty to publish and the writer keeping no right to terminate. Lord Reid asked whether the bargain was fair and observed that the writer’s output would be sterilised if the publisher withheld it. Lord Diplock identified the policy as the protection of those whose weak bargaining power forces them into unconscionable bargains. Lord Reid added that a clause permitting the writer to terminate and recover his copyrights would have altered the case, the feature whose absence rendered the contract unconscionable being a reversion right. The Court of Appeal applied the principle in Clifford Davis Management v WEA Records, setting aside the assignment of two composers of Fleetwood Mac who had transferred all their works for ten years for one shilling, and the line ran on through the recording industry, voiding the contracts of the Stone Roses and of Holly Johnson of Frankie Goes to Hollywood as unreasonable restraints of trade.
India holds the same doctrine. Its contract law voids the unconscionable bargain made under unequal bargaining power, as the Supreme Court held in Central Inland Water Transport Corporation v Brojo Nath Ganguly, and section 19(3) of the Copyright Act requires an assignment to state the royalty payable to the author. The judicial route, however, is partial. It reaches the author able to litigate, which is the author who has become valuable, and leaves the contracts of the rest in force. A statutory reversionary right would apply the same principle to every author by operation of law.
India’s refusal to have a reversionary right is unusual. French and German law retain for the author an inalienable core. Canada retains the 1911 reverter. The United Kingdom abolished it for the future alone, and the European Union introduced a use-it-or-lose-it revocation right in 2019.
Why this history now?
The history bears on the current demand that those who train generative models on protected works should pay for the use. The demand is made in the author’s name and assumes that the author holds the copyright in her work. On the present distribution of rights, the author who assigned her work holds nothing a training license would compensate. Therefore, before we get to an AI debate with the author as the face, it is important we fix our copyright system in the first place to actually enable those who it claims to!
The debates can be found here! Specifically important pages are 98, 136, 154 (PDF). Huge thanks to Prashant Reddy and Sumathi Chandrashekharan for creating this database with all debates.