Rakesh and Dinesh enter into an agreement wherein Rakesh agrees to sell to Dinesh a brand new Wagon-R car of LXI model which is red in colour. It turns out that Rakesh has two cars in his possession, a Wagon-R LXI of white colour and another Wagon-R VXI of red colour. Determine whether evidence may be given in order to determine the meaning of the clause. Also: A agrees to sell to B for Rs. 1,000/- 'my white horse'. A has two white horses. Can evidence be given to show which of the horses was meant?
The Nature of the Problem: Equivocation
When parties reduce their agreement to writing, the law insists that the written word alone speaks for them. But occasionally, a document that reads with perfect clarity on its face reveals, upon application to the facts of the real world, that the language fits more than one subject with equal accuracy. This is what the old masters of the common law called equivocation — aequivocatio — speaking equally to two things at once. Lord Bacon explained the distinction with characteristic precision: latens ambiguity is that which seemeth certain and without ambiguity for anything that appears upon the deed or instrument, but there is some collateral matter out of the deed that breeds the ambiguity. Equivocation is the most perfect form of latent ambiguity, for the language does not merely fail to identify — it identifies two subjects with identical accuracy.
It is this species of ambiguity that Section 96 of the Indian Evidence Act, 1872 — now re-enacted as Section 99 of the Bharatiya Sakshya Adhiniyam, 2023 — is designed to address. The section reads:
"When the facts are such that the language used might have been meant to apply to any one, and could not have been meant to apply to more than one, of several persons or things, evidence may be given of facts which show which of those persons or things it was intended to apply to."
Problem I: Rakesh and Dinesh — The Wagon-R Agreement
Reading the Document on Its Face
When one reads the agreement between Rakesh and Dinesh, it is a picture of clarity. Rakesh agrees to sell "a brand new Wagon-R car of LXI model which is red in colour." Any person reading this document would see no ambiguity at all — it describes a Wagon-R, it specifies the model as LXI, and it specifies the colour as red. So far, the document is perfectly plain. No intelligent reader, on perusal of this document alone, would detect any uncertainty. This conclusively establishes that the ambiguity here is not patent — it is not visible on the face of the document — and therefore Section 93 of the Evidence Act (Section 96 of the BSA) has no application. That section, which deals with ambiguity on the face of a document, excludes oral evidence. But it simply does not apply here.
Where the Latent Ambiguity Surfaces
The ambiguity surfaces the moment one looks at the external facts: Rakesh does not have one car that matches the full description. He has two cars —
A Wagon-R LXI of white colour — which matches the model (LXI) but not the colour (it is white, not red).
A Wagon-R VXI of red colour — which matches the colour (red) but not the model (it is VXI, not LXI).
This is the critical insight. The description in the agreement partly fits the first car (model is correct) and partly fits the second car (colour is correct), but the whole of it does not correctly apply to either. This is precisely the situation contemplated by Section 97 of the Indian Evidence Act (Section 100 of the BSA), which provides:
"When the language used applies partly to one set of existing facts, and partly to another set of existing facts, but the whole of it does not apply correctly to either, evidence may be given to show to which of the two it was meant to apply."
The illustration to Section 97 is directly on point: A agrees to sell to B "my land at X in the occupation of Y." A has land at X, but not in the occupation of Y; and he has land in the occupation of Y, but it is not at X. Evidence may be given of facts showing which he meant to sell. The Wagon-R problem is a modern automotive parallel of this agricultural illustration. The description "Wagon-R LXI, red in colour" is the equivalent of "land at X in the occupation of Y" — a composite description that fits two subjects partially but neither completely.
The Applicable Principle: Falsa Demonstratio Non Nocet
This section is built upon the ancient maxim falsa demonstratio non nocet — a false description does not vitiate the document. The law recognises that parties often describe the subject matter of their agreements with a degree of inaccuracy — they may get the model right and the colour wrong, or the colour right and the model wrong — and it would produce a harsh and unjust result if such an error were to render the entire agreement void. The court does not treat the misdescription as a fatal defect. Instead, it permits extrinsic evidence to identify the car that was truly intended, and the false element of the description is treated as surplusage — rejected, while the remaining description stands as the anchor of identification. The Madras High Court applied this maxim in Karuppa Goundan alias Thoppala Goundan v. Periathambi Goundan (1907 ILR 30 Mad 397), holding that where sufficient description of the subject matter is given, a false description added thereto may be rejected, and the valid description used to identify the subject.
What Evidence May Be Given
In the dispute between Rakesh and Dinesh, evidence of the following kinds would be admissible:
Surrounding circumstances — did the parties inspect a particular car before executing the agreement? Was either car pointed out during negotiations?
Correspondence and conduct — were any letters, messages, or communications exchanged in which either the LXI model or the red colour was specifically emphasised?
Subsequent conduct — how did the parties themselves understand the agreement after it was signed? The Supreme Court in Raj Kumar Rajinder Singh v. State of Himachal Pradesh (AIR 1990 SC 1833) held that subsequent conduct of the parties furnishes evidence to clear the blurred area and to ascertain the true intention.
Trade and commercial understanding — if the model designation (LXI vs. VXI) is an essential term in the context of the transaction, that would be a strong pointer to the LXI model being the intended car, notwithstanding the colour discrepancy.
What would not be admissible is oral evidence that is offered not to identify which car was meant, but to add a new term or to substitute a different car entirely for the ones Rakesh possesses. The evidence must serve the purpose of identification, not substitution.
Problem II: A's Two White Horses — A Classic Equivocation
The Document and the Ambiguity
This problem is drawn directly from Illustration (a) to Section 96 of the Indian Evidence Act (Section 99 of the BSA). A agrees to sell to B "my white horse" for one thousand rupees. The document is lucid and apparently unambiguous. A reader of this agreement would detect no problem. However, the latent ambiguity emerges upon reference to the external facts: A has two white horses. The description "my white horse" applies with complete and equal accuracy to both horses. This is equivocation in its purest form.
The statute is direct and unequivocal on the admissibility of evidence here: evidence may be given of facts which show which of them was meant. There is no hesitation in the law on this point.
Why This Is Permitted While Patent Ambiguity Is Not
The distinction between this case and the patent ambiguity in Section 93 (Section 96 BSA) — where A agrees to sell "a horse for Rs. 1,000 or Rs. 1,500" — deserves close attention, for it reveals the inner logic of the whole scheme. In the patent ambiguity case, the document is defective because the parties themselves were not agreed on the price; allowing oral evidence to supply the missing agreement would be constructing a contract, not interpreting one. In the white horse case, the parties were perfectly agreed — they agreed upon one white horse. The document fails only in identification, not in agreement. The agreement is complete; what extrinsic evidence does is merely point to the specific subject matter of an already-concluded agreement. This distinction was expressly recognised by the Andhra Pradesh High Court in Pradeep Kumar v. Mahaveer Pershad (AIR 2003 AP 107), which observed that Sections 93 and 94 deal with patent ambiguity while Sections 95, 96 and 97 deal with latent ambiguity — and that the distinction carries crucial consequences for the admissibility of extrinsic evidence.
What Extrinsic Evidence Is Admissible
In the two white horses problem, the following kinds of extrinsic evidence would be admissible to identify which horse was meant:
Evidence that B had seen, inspected, or expressed a preference for a particular horse before the agreement was made.
Evidence that one of the horses was the one Rakesh typically referred to as "my white horse" in dealings with others.
Evidence of declarations of intention made by A at or near the time of the agreement, indicating which horse he had in mind.
On the admissibility of declarations of intention in equivocation cases, the courts have been relatively liberal. In Price v. Page (4 Ves Jr 679), a bequest was made to "Price, son of — Price," leaving the father's first name blank. Both the father and the son bore the surname Price. The bequest could apply to either. Parole evidence of the testator's declarations that he intended the bequest for the son was admitted — not to add a term to the document, but to resolve which of two equally-described persons was meant. Similarly, in Re Jeffrey 1 Ch 275, where a testatrix left property to "my brother B, his wife, and their daughter," and B had four daughters, oral evidence was admitted to show that the testatrix was particularly close to one daughter and intended her to be the beneficiary.
The Supreme Court's observations in Chunchun Jha v. Ebadat Ali (AIR 1954 SC 345) remain authoritative: where there is ambiguity in the language employed, it is permissible to look to the surrounding circumstances to determine what was intended. The restriction is that such evidence must resolve the ambiguity — not override a clear term.
The Connecting Thread
Both problems illustrate the same fundamental truth about the law of evidence as it operates on written documents. The Evidence Act — whether in its original form or as the BSA, 2023 — does not aim to make documents into instruments of injustice by mechanically enforcing language that, when tested against reality, fits two subjects equally. Where the parties have agreed, and the document faithfully records that agreement, but an extrinsic circumstance — the existence of two white horses, or two cars that each partially match the description — creates a puzzle about the specific subject matter, the law sensibly opens the door to extrinsic evidence. The admission of that evidence does not contradict or override the document. It does what good interpretation always does: it uses the surrounding reality to illuminate what the written words were always meant to say.
Get weekly legal insights
Case-law digests, exam tips & curated study guides — straight to your inbox.
No spam. Unsubscribe anytime.
