Custom is the oldest source of law, yet a courtroom will not enforce a practice merely because it is old. To displace the general law a custom must clear a cumulative battery of tests: it must be ancient, continuous and invariable, reasonable, certain, and not opposed to statute, morality or public policy. Each requirement is a separate hurdle, and failure on any one is fatal. This chapter dissects those five essentials—drawing on the classic Privy Council learning in Ramalakshmi Ammal v. Sivanantha Perumal and Collector of Madura v. Moottoo Ramalinga and the modern restatements in Gokal Chand v. Parvin Kumari and Bhimashya v. Janabi—and explains how they are now codified in Section 3(a) of the Hindu Marriage Act, 1955. For the wider framework, read this alongside our notes on the definition and nature of custom and the Customary Law hub.
Why a Custom Must Pass a Test
A custom is, in the classic words of the Privy Council in Hurpurshad v. Sheo Dyal (1876) 3 IA 259, "a rule which in a particular family or in a particular class or district has from long usage obtained the force of law." The phrase "force of law" is the point of difficulty: countless practices are observed in Indian society, but only a fraction of them are legally binding. The essentials of a valid custom are the filter that separates a genuinely law-creating usage from mere habit, fashion or local convenience.
The essentials matter most because a valid custom is not subordinate to text—it may actually override it. In The Collector of Madura v. Moottoo Ramalinga Sathupathy (1868) 12 MIA 397, the Judicial Committee laid down the foundational proposition that "under the Hindu system of law, clear proof of usage will outweigh the written text of the law." Because a proven custom can defeat the smritis and the codified general law, the courts insist that the proof be rigorous and that the usage satisfy every essential before it is given that elevated status. The burden is correspondingly heavy and rests squarely on the party who pleads the custom, as we explore below. These essentials are not academic ornaments—they are the gatekeepers that prevent self-serving or anti-social practices from masquerading as law. For the conceptual background to how usage hardens into law, see our companion note on the definition and nature of custom.
The Five Essentials at a Glance
Indian courts, drawing on both the English common law and the Hindu texts, have settled on a recognised list of attributes that a custom must possess. The classic formulation is that a custom must be (1) of antiquity—ancient or immemorial; (2) continuous and invariable in its observance; (3) reasonable; (4) certain as to its content and the class it governs; and (5) not opposed to law, morality or public policy. To these primary five, the texts add that the custom must be obligatory (observed as of right, not by mere permission or goodwill) and must not be immoral—the last often folded into the reasonableness and public-policy heads.
The Supreme Court compactly restated the list in Bhimashya v. Janabi (2006) 13 SCC 627, observing that for proof of a custom "all that is necessary is that the custom or usage has been acted upon in practice for such a long period and with such invariability" that it has been accepted as the governing rule, and that "certainty and reasonableness" are indispensable. The English starting point is the seventeenth-century Case of Tanistry (1608) Dav 28, where the Court of King's Bench struck down the Irish custom of tanistry as "unreasonable and void" and "void for uncertainty"—an early demonstration that even an ancient practice fails if it cannot satisfy the other heads. The essentials are therefore cumulative, not alternative: a usage must clear all of them simultaneously.
Essential 1 — Antiquity
Antiquity is the first and most famous requirement. The custom must be ancient. In English law the test was rigid: a custom had to date from "time immemorial," fixed by the Statute of Westminster I (1275) at the year 1189 (the accession of Richard I). Indian courts wisely declined to import that arbitrary date. The Privy Council in Hurpurshad v. Sheo Dyal held that it is not essential in every case for antiquity to be carried back to a period beyond the memory of man; what is necessary is that the usage "has been acted upon in practice for such a long period and with such invariability as to show that it has, by common consent, been accepted as the governing rule."
How long is "long"? There is no fixed numerical rule, but courts have often treated a usage demonstrably observed for around a century as sufficiently ancient, while shorter periods may suffice where the proof of acceptance is otherwise strong. The orthodox texts demand only that the custom be of "long usage"; the question is one of fact in each case. The relaxation of the rigid English date is sensible for India, where written records of practice are often sparse and where many communities have followed their usages for generations without documentary trace. What the court ultimately looks for is not a precise birth-date but the unmistakable mark of a settled, accepted rule whose origin is lost in time. Crucially, antiquity alone is never enough. The locus classicus is the Privy Council's warning in Ramalakshmi Ammal v. Sivanantha Perumal Sethurayar (1872) 14 MIA 570, that "it is of the essence of special usages, modifying the ordinary law of succession, that they should be ancient and invariable"—antiquity and invariability are yoked together, and a practice that is old but has been departed from will still fail.
Essential 2 — Continuity and Invariability
A custom must have been observed continuously and without interruption. If a usage has been abandoned and then revived, or if it has been observed only fitfully, it loses its character as law. The leading authority is the Privy Council's decision in Mst. Subhani v. Nawab AIR 1941 PC 21, which holds that continuity and invariability are essential and that "a custom, however old it may be, in the absence of clear proof of its continuous observance without any variation, would not obtain legal existence." Interruption in the enjoyment of a right is less fatal than interruption in the right itself: a temporary failure to exercise a customary right (because, say, no occasion arose) does not destroy it, but a discontinuance of the right by the community does.
Invariability is the twin of continuity. The custom must have been observed in the same form; if instances pull in different directions, the alleged custom is really no custom at all but a series of inconsistent acts. This is why courts insist on consistent instances. In Saraswathi Ammal v. Jagadambal AIR 1953 SC 491, the Supreme Court refused to recognise an alleged custom of succession among dasis in the Tanjore district, holding that the oral evidence of instances—where better documentary proof had been withheld—was too meagre and unsatisfactory to establish a continuous and invariable usage. Where a special family custom is set up, the courts demand cogent and consistent instances over time, as in Harihar Prasad Singh v. Balmiki Prasad Singh, where the claimants sought to prove dozens of instances to establish a special rule of succession in a Bhumihar family. The contrast with mere assertion shows why this essential is the practical battleground of most custom litigation. For how local, family and class customs differ in their continuity demands, see our note on the kinds of custom.
Essential 3 — Reasonableness
A custom must be reasonable. The test is not whether the custom appears reasonable to a modern judge applying contemporary values, but whether it was reasonable at its inception and is not so manifestly unjust, oppressive or absurd that no court could enforce it. The rule was crisply put in the Case of Tanistry and is echoed throughout the Indian authorities: an unreasonable custom is "void ab initio." The burden of showing unreasonableness lies on the party challenging the custom; the law leans, at first, toward upholding a usage that is otherwise ancient and certain, and will strike it down only where its unreasonableness is clearly made out.
Reasonableness operates as a brake on customs that work pure mischief. A usage that is merely inconvenient, or that a court would not itself have devised, survives; a usage that is positively iniquitous does not. The Madras High Court applied this thinking in Deivanai Achi v. Chidambaram Chettiar AIR 1954 Mad 657, refusing to recognise the so-called self-respecters' (suyamariyathai) marriage as valid on the strength of an alleged custom, because the practice was too recent and uncertain and could not be shown to be an ancient, definite and reasonable usage of the community—a defect later cured for that community only by statute. Reasonableness thus overlaps with antiquity (a recent practice has had no time to prove its reasonableness through acceptance) and with the requirement of conformity to public policy, discussed below.
Essential 4 — Certainty
A custom must be certain. Two kinds of certainty are required: certainty as to the content of the rule (what exactly does the custom require or permit?) and certainty as to the class, locality or family to which it applies (who is bound by it?). A vague, fluctuating or indefinite practice cannot have the force of law, because the courts could not apply it predictably. The Case of Tanistry again furnishes the principle: the impugned custom there was held "void for uncertainty" as well as unreasonable.
Certainty is closely linked to invariability. If the instances pleaded vary so much that no single rule can be extracted from them, the custom fails for uncertainty as much as for want of continuity. Indian courts have repeatedly rejected alleged customs on this ground where the witnesses gave inconsistent accounts of the supposed rule. The Supreme Court in Bhimashya v. Janabi expressly listed certainty, alongside reasonableness, as an "indispensable" element. The practical lesson for litigants is that the custom must be pleaded with precision—what it is, whom it binds, and within what territorial or communal limits—and then proved by instances that all point unambiguously to that single, settled rule. Uncertainty is fatal for a practical reason: a court administers a custom as a rule of decision, and a rule that cannot be stated with precision cannot be applied consistently from case to case. A pleading that asserts a vague "tradition" without defining its precise content or its beneficiaries invites dismissal, however genuine the underlying practice may be.
Essential 5 — Conformity with Law and Public Policy
A custom must not be opposed to any statute, nor to morality and public policy. This is the most important limit in the modern era, because the legislature can always abrogate a custom by enacting to the contrary. Where statute and custom collide, statute prevails: a custom contrary to an express enactment is simply void. Thus the codifying Hindu statutes of 1955–56 swept away many customs that conflicted with their provisions, preserving only those customs which the statutes themselves expressly saved.
Even apart from statute, a custom that offends public morality or public policy will not be enforced. The classic illustration is Mathura Naikin v. Esu Naikin (1880) ILR 4 Bom 545, where the Bombay High Court (per West J.) refused to recognise the custom of dedicating girls as naikins or dancing-girls for an immoral purpose, holding such a usage opposed to public policy and morality and therefore incapable of legal recognition. Customs sanctioning practices repugnant to justice or decency—however ancient—fall at this hurdle. The point connects to the constitutional order too: a custom inconsistent with fundamental rights or constitutional morality cannot stand. We develop the statutory-override theme in the notes on custom as a source of Hindu law and the special treatment of custom under Muslim law.
Supplementary Requirements — Obligatory Force and Morality
Beyond the canonical five, the texts insist that a custom be obligatory—observed as a matter of right and felt to be binding, not merely tolerated as a matter of grace or convenience. A practice followed because individuals choose to follow it, and which they feel free to abandon at will, lacks the opinio necessitatis that converts usage into law. The community must regard the rule as compulsory. This is why courts probe whether the instances reflect an obeyed rule or just a recurring choice.
The morality requirement, though often folded into reasonableness and public policy, is sometimes stated separately because of its prominence in Indian custom litigation. A custom that is immoral cannot be enforced regardless of its age. The line of "dancing-girl" cases—of which Mathura Naikin v. Esu Naikin is the leading example—rests on this principle, as does the refusal of the courts to lend their aid to customs sanctioning practices the law regards as against good morals. The supplementary requirements reinforce, rather than dilute, the central message: a custom earns the force of law only when the community treats it as a binding, decent and settled rule.
Burden and Mode of Proof of Custom
Because a custom displaces the general law, the party asserting it must plead and prove it. The Supreme Court's leading statement is Thakur Gokal Chand v. Parvin Kumari AIR 1952 SC 231, holding that a custom must be proved to be ancient, certain and reasonable, and that even where cohabitation raised a presumption of marriage, the special custom relied upon could not be enforced unless it was firmly established by evidence of long, invariable usage having the force of law. The onus is on the propounder, and it is discharged only by clear and unambiguous proof.
How is custom proved? Custom is fundamentally a question of fact, established by instances—judicial decisions recognising the usage, documentary evidence such as deeds and records, and the testimony of persons likely to know. The Supreme Court in Bhimashya v. Janabi reiterated that it is "incumbent on the party setting up a custom to allege and prove the custom on which he relies." Once a custom has been judicially recognised in a series of decisions, however, the courts may take judicial notice of it under Section 57 of the Indian Evidence Act, 1872, so that it need not be proved afresh in every case—a point illustrated by Ujagar Singh v. Mst. Jeo AIR 1959 SC 1041, where the question was whether the customary law of the Punjab regarding a sister's succession had become so well established as to be judicially noticed. Until that stage is reached, the party must prove the custom by cogent evidence of repeated, consistent instances. Saraswathi Ammal warns that oral evidence of instances will be viewed with suspicion where better documentary evidence has been withheld without explanation.
Statutory Codification — Section 3(a), Hindu Marriage Act, 1955
The judge-made essentials have been crystallised in statute. Section 3(a) of the Hindu Marriage Act, 1955 defines "custom" and "usage" to signify "any rule which, having been continuously and uniformly observed for a long time, has obtained the force of law among Hindus in any local area, tribe, community, group or family"—subject to two provisos: that the rule is "certain and not unreasonable or opposed to public policy," and that, in the case of a rule applicable only to a family, "it has not been discontinued by the family."
This single clause captures every essential discussed above. "Continuously and uniformly observed" embodies continuity and invariability; "for a long time" embodies antiquity; "obtained the force of law" embodies the obligatory character; "certain" embodies certainty; "not unreasonable" embodies reasonableness; "not... opposed to public policy" embodies conformity with law and morality; and the family proviso codifies the rule that a family custom dies once the family abandons it. An identically worded definition appears in Section 3(a) of the Hindu Succession Act, 1956 and cognate codes, so the same essentials govern across the codified Hindu law. The statutory definition is therefore best understood not as a fresh test but as a faithful condensation of the case law from Ramalakshmi Ammal to Subhani v. Nawab.
The Essentials in Muslim and Tribal Contexts
The essentials of a valid custom apply across personal-law systems, but their practical scope differs. In Muslim law the role of custom was deliberately narrowed by the Muslim Personal Law (Shariat) Application Act, 1937. Section 2 of that Act provides that "notwithstanding any custom or usage to the contrary," in the listed matters—intestate succession, special property of females, marriage, dissolution of marriage, maintenance, dower, guardianship, gifts, trusts and wakfs—the rule of decision where the parties are Muslims shall be the Muslim Personal Law (Shariat), with questions relating to agricultural land originally excepted. The Act thus abrogated many customs that had previously displaced the Shariat among Indian Muslim communities, illustrating the supremacy of statute over custom. For the fuller picture, see our note on custom as a source of Muslim law.
By contrast, custom retains vigorous force in tribal and certain agrarian settings, where the general personal laws are often expressly excluded and customary law governs succession, marriage and land. Even there, however, the essentials still operate: a tribal custom must be ancient, continuous, reasonable and certain before a court will enforce it, and it must not offend a binding statute or constitutional guarantee. The tests are constant; only the field of operation expands or contracts depending on how far statute has occupied the ground.
Common Pitfalls and Exam Pointers
Three errors recur in answers on this topic. First, students treat the essentials as alternatives—reasoning that an ancient custom must be valid. They are cumulative; an ancient custom that is unreasonable, uncertain or illegal still fails, as the Case of Tanistry shows for an ancient practice struck down for unreasonableness and uncertainty. Second, students confuse the English "time immemorial" rule (fixed at 1189) with Indian law; Hurpurshad v. Sheo Dyal rejected any fixed date and asks only for long, accepted usage. Third, students forget that custom is a question of fact carrying a heavy burden of proof, governed by Gokal Chand v. Parvin Kumari and Bhimashya v. Janabi.
A strong answer states the five essentials, anchors each to a leading case—antiquity to Hurpurshad and Ramalakshmi Ammal, continuity to Subhani v. Nawab and Saraswathi Ammal, reasonableness to the Case of Tanistry and Deivanai Achi, certainty to Tanistry and Bhimashya, and legality/public policy to Mathura Naikin—and then closes with the statutory codification in Section 3(a) of the Hindu Marriage Act, 1955 and the priority of Collector of Madura's principle that clear proof of usage outweighs the written text. For first-principles grounding, revisit the definition and nature of custom and the Customary Law hub.
Frequently asked questions
What are the five essentials of a valid custom?
A valid custom must be (1) ancient (antiquity), (2) continuous and invariable, (3) reasonable, (4) certain as to content and the class it binds, and (5) not opposed to statute, morality or public policy. The texts add that it must also be obligatory (observed as of right) and not immoral. These requirements are cumulative—failure on any one is fatal, as the Case of Tanistry illustrates.
Does an Indian custom have to date from 'time immemorial' like in English law?
No. English law fixed immemoriality at the year 1189, but Indian courts rejected any fixed date. In Hurpurshad v. Sheo Dyal (1876) 3 IA 259 the Privy Council held it is not essential to carry antiquity back beyond the memory of man; it is enough that the usage has been acted upon for such a long period and with such invariability that it has, by common consent, become the governing rule.
Can a valid custom override the written Hindu law?
Yes. In The Collector of Madura v. Moottoo Ramalinga Sathupathy (1868) 12 MIA 397 the Privy Council held that under the Hindu system clear proof of usage will outweigh the written text of the law. This is precisely why the essentials are applied so strictly and the burden of proof is so heavy—a proven custom can defeat the general law.
Why does continuity matter, and which case establishes it?
Continuity ensures the usage has never been abandoned; an interrupted custom loses legal existence. Mst. Subhani v. Nawab AIR 1941 PC 21 holds that continuity and invariability are essential and that a custom, however old, will not obtain legal existence without clear proof of continuous observance without variation. Saraswathi Ammal v. Jagadambal AIR 1953 SC 491 refused a custom on inadequate, inconsistent instances.
When is a custom struck down as unreasonable or against public policy?
A custom is void if it is manifestly unjust, oppressive or absurd, or if it offends morality or public policy. In Mathura Naikin v. Esu Naikin (1880) ILR 4 Bom 545 the Bombay High Court refused to recognise the custom of dedicating girls as dancing-girls for an immoral purpose. The Case of Tanistry (1608) struck down an ancient custom as 'unreasonable and void' and 'void for uncertainty.'
Who must prove a custom and how is it proved?
The party asserting the custom must plead and prove it, as held in Thakur Gokal Chand v. Parvin Kumari AIR 1952 SC 231 and reaffirmed in Bhimashya v. Janabi (2006) 13 SCC 627. Custom is a question of fact, proved by consistent instances, judicial decisions, deeds and the testimony of knowledgeable persons. Once judicially recognised repeatedly, it may be judicially noticed under Section 57 of the Indian Evidence Act, 1872, as discussed in Ujagar Singh v. Mst. Jeo AIR 1959 SC 1041.