Transliteration Conflicts For Indian- Language Trademarks
Author: LawSphereX
Category: Intellectual Property
Published:
Updated:
TRANSLITERATION CONFLICTS FOR INDIAN-
LANGUAGE TRADEMARKS
Nikita Banerjee, LL.B., Jindal Global Law School
ABSTRACT
Trademark marks in India’s multilingual market frequently pass through more than one script and language. An urgent problem is presented when a mark in an Indian script (e. g. , the Hindi word “अमूल”) exists alongside its Latin script counterpart (“Amul”), resulting in shared pronunciation but differentiated visual look. This study investigates how Indian law of trademarks deals with the issue of confusion arising from such scripttranslated marks. The examination makes use of the provision under the Trade Marks Act, 1999 - Section 11(1) (relative grounds of refusal for confusing similarity) and Section 9(2)(a) (absolute ground against deceptive or confusing marks). It is seen that Indian courts as well as the Trade Marks Registry have always protected against such script confusion, viewing phonetically similar marks in other scripts as being likely to cause confusion among consumers.
In particular, direct translations (words with the same meaning in two different languages) have been litigated in traditional cases, whereas the situation of pure transliteration (same sound, other script) has rarely been directly mentioned as a special category, albeit being implicitly included by the law. From an examination of Indian legislation, test practices, and judicial rulings, this paper explains the principles used to address transliteration disputes and concludes that the existing system necessarily encompasses such disputes under the broader principle of deceptive similarity. The bottom line is that Indian trademark offices must keep rejecting or cancelling phonetically equivalent marks in different scripts to avoid confusing consumers, while not ruling out good faith or honest concurrent use defences in exceptional situations.
Introduction
India’s linguistic richness poses particular trademark challenges1. There are twenty-two official languages, and they are written in eleven scripts. This implies that a brand name may be written in more than one writing system. A trademark can therefore find itself in English (Roman script) and also in local scripts such as Devanagari, Tamil, or Bengali. An example can be the popular dairy firm “Amul,” which can be used in the Latin script as Amul and also written in Devanagari as अमूल. Whereas orthography varies, pronunciation does not. This question must then be asked: how is Indian trademark law to deal with two marks that pronounce the same but appear different because of script? According to the Trade Marks Act, 1999, one of the prime grounds for rejection is likelihood of confusion with prior marks.
Section 11(1) of the Act forbids the registration of a mark if, due to its resemblance to a prior trademark and the similarity or identity of the goods/services, “there exists a likelihood of confusion on the part of the public”. The provisions do not restrict confusion to visual resemblance; phonetic similarity is also a sufficient ground for refusal. In fact, Indian courts have confirmed that the similarity of competing marks has to be considered in respect of their visual, aural and conceptual similarity, having regard to average consumer with imperfect memory. Practically, Indian trademark examiners use phonetic searches to trap sound-alike marks in different scripts.
The Trade Marks Rules, 2017 also mandate that where a mark includes words in a script other than English or Hindi (the two accepted scripts for filing), the application must have a clear transliteration (phonetic transcription in Roman or Devanagari script) and translation of every such word. This would enable the Registry to phonetically index and search marks in all Indian languages. But aside from these procedural approaches, there is a substantive issue: if two marks are phonetically equivalent by transliteration, should they be viewed as effectively the same, or simply as similar marks subject to a standard confusion analysis? Does Indian law treat script-based confusion differently from other similarities? This paper addresses these issues by canvassing Indian case law addressing marks in multiple languages and scripts.
It makes a distinction between translation equivalence (where the marks convey the same meaning in other languages) and transliteration or script equivalence (where
1 Trade Marks Act, No. 47 of 1999, § 11, Acts of Parliament, 1999 (India).
the marks convey the same sound, since one is only the phonic representation of the other in a different script). While Indian courts have often resolved the former (translation cases) on the doctrine of deceptive similarity, the latter (pure transliteration disputes) have not been discussed explicitly as much, being typically dealt with on general principles without distinct doctrinal tags. We will demonstrate that even without express statutory provisions regarding scripts, the current legal principles adequately encompass transliteration disputes, focusing on phonetic similarity and probable confusion. The Legal Framework of Phonetic Similarity and Script Variations Indian trademark law covers both relative grounds and absolute grounds dealing with confusing similarity.
Section 11(1) (a relative ground) prevents registration of a mark if its identity or similarity with an earlier mark, along with the similarity of goods/services, is likely to confuse the public. This corresponds to the international standard that no two confusingly similar marks can exist for related goods, or else consumers may be confused concerning origin. Notably, the standard is not only direct confusion but also likelihood of association, and it means even a suggestion of connection can be sufficient. The comparison of marks under Section 11 is holistic2. Courts use the classic test (from Pianotist Co. ) and Amritdhara Pharmacy v. Satya Deo Gupta (1963) guidance of the Indian Supreme Court based on an average consumer’s overall impression from the marks.
The Supreme Court in Amritdhara mentioned that an innocent buyer with average intelligence and imperfect memory does not analyse marks into parts but applies general sound, appearance and meaning to judge them. Phonetic similarity is therefore an important consideration. A mark which is pronounced like a previous mark may be held confusingly similar, even if it is not spelled or scripted the same. Statutory protection of phonetic similarity is also found in provisions relating to infringement, specifically, Section 29(9) of the Act explains that a trademark can be infringed by word use orally, and sound similarity can constitute infringement as much as visual similarity. While Section 29(9) addresses use, not registration, it emphasizes the recognition by the legislature that phonetic identity is capable of causing confusion in its own right regardless of graphic
2 Amritdhara Pharmacy v. Satya Deo Gupta, AIR 1963 SC 449 (India).
appearance. The same principle applies to examiners and courts when analysing conflicts at the registration level. Absolute Grounds - Falsifying Marks Section 9(2)(a) states that a mark shall not be registered if “it is of such nature as to deceive the public or cause confusion”. In contrast to Section 11, which deals with conflicts against prior marks, Section 9(2)(a) deals with misleading marks per se. Indian courts have in some cases used Section 9(2)(a) together with Section 11 where a subsequent mark in another script/language is clearly adopted to confuse the consumers. An example is a case between the English mark “ROHIT” and a subsequent Hindi mark “रोिहत” for the same products, where the Delhi High Court held the Hindi mark to be phonetically and visually equivalent to the English mark and likely to cause confusion.
The court directed cancellation of the Hindi-script mark, impliedly acknowledging that simply a change of script did not immunize the junior user against a holding of deceptive similarity. Need of Transliteration in Applications Procedurally, Rule 28 of the Trade Marks Rules, 2017 prescribes that any application for a trademark having words in any script other than Roman or Devanagari should be submitted along with a transliteration in either of these scripts (i. e. an English or Hindi phonetic version) and a translation if necessary. For instance, a Bengali name applicant would have to give its pronunciation in Hindi script or Latin, and state the meaning in English/Hindi.
The rule serves a dual purpose: (1) to enable the Trademark Office database to register the mark’s sound for searching, and (2) to inform examiners and the public about what the mark signifies if it is a dictionary word. The Draft Trade Marks Manual also directs examiners to make sure that transliterations are included and to require them if not available. Once the transliteration is on record, examiners perform phonetic similarity searches in every script. The Office’s computerized system contains a phonetic search mode, which enables retrieval of previous marks that “sound alike” although differently spelled. Therefore, if an individual applies for registration of “अमूल” (Amul in Devanagari) for a dairy product in Class 29, the examiner would automatically notice that “Amul” in English is already registered for similar goods, and raise an objection under Section 11(1).
Cross-Language Trademark Disputes in Indian Case Law To have a sense of how script-based (transliteration) disputes are handled, it is worthwhile first to consider cases of similarity based on translation where the marks involved communicate the same message in another language. Indian courts possess a settled line of authority holding that mere translation of a well-known or established mark into a different language cannot avoid liability (or refusal of registration). The reason is that consumers who use different languages may still be deceived if the second mark suggests the same notion or idea as the first. Most of these rules extend a fortiori to absolute transliterations where phonetic effect is identical. The “Fruit Salt”3 Case, Eno v. Vishnu Chemical Co. is one of the first Indian cases on translation.
The plaintiff sold an effervescent antacid under the English name “Fruit Salt,” a name which by long usage had come to signify exclusively the plaintiff’s product. The defendant started selling a similar product under the Marathi name “Falaxar”, a Marathi portmanteau of fal (fruit) and kshar (salt). The High Court ruled that “Fruit Salt” had become distinctive of the plaintiff’s product, and that no one might use those words “in any language” for similar goods. That is to say, an established mark’s goodwill extends to its vernacular translations. The court resorted to the principle that if the sole distinction between two marks is language but meaning is the same, then the subsequent mark cannot be allowed, as confusion is unavoidable. In Hindustan Lever Ltd. v. T. G. Balaji Chettiar4, the Madras High Court had a similar case of soap brands.
Hindustan Lever (HLL) had been selling its iconic “Sunlight” laundry soap (with a sun device) all over India, including in non-English advertisements. The defendant Chettiar attempted to register “Surian” (सू+रयन्), the Tamil term for “sun,” as a trademark for soaps, with a logo incorporating a sun. HLL argued that “Surian” was no more than a translation of “Sun” and that Tamil-speaking consumers would probably equate Surian-named soaps with HLL’s Sunlight soap. The court concurred: while one mark was written in Tamil script and the other was in English, each referred to the same item (the sun) and each made use of a sun emblem. Observing HLL’s extensive earlier use of Sun/Sunlight, the court considered a “close resemblance” between the marks’ salient concept, the sun, and held that “Surian” for soap would mislead customers to believe it was HLL’s Sunlight product.
This case emphasizes that
3 Eno v. Vishnu Chemical Co., AIR 1972 Cal. 306 (India).
4 Hindustan Lever Ltd. v. T.G. Balaji Chettiar, AIR 1995 Mad. 272 (India).
in determining similarity, courts consider not only literal spelling but the concept conveyed, especially when a mark is a word-for-word translation. The Delhi High Court ruling in Bhatia Plastics v. Peacock Industries Ltd. 5 stretched the translation principle from direct dictionary equivalents to include synonyms in languages. The plaintiff marketed plastic items under the trade mark “MAYUR” (a Hindi/Sanskrit word for “peacock”) and also used the Punjabi word “Mayuri” and the English term “Peacock” for its products. The defendant started using “Peacock” for similar plastic goods. Although Mayur and Peacock look and sound completely different, the court held that one cannot adopt a mark in another language that conveys the same idea as an existing mark. Thus, Peacock was held deceptively similar to Mayur, and the defendant was restrained from using it.
It was immaterial that “Mayur” is a Sanskrit-derived word that is not ordinarily used in general usage; what was significant was the same idea of a peacock connecting the plaintiff’s wares and the defendant’s trademark. Where two marks indicate the same thing in different languages, Indian courts have always held them to be likely to confuse consumers when applied to similar products. The hidden public policy is to preclude a trader from performing indirectly (by translation) what they are prohibited from performing directly (using the same mark). Permitted translation or synonymous marks would allow unscrupulous copying aimed at consumers who speak a single one of the languages. In fact, courts have pointed out that in a multilingual community people are often bilingual, and even those who are not might be confused if a local-language mark is an equivalent mark in another language.
Transliterated Marks: Phonetic Identity Across Scripts Whereas translation cases concern equal meanings, transliteration cases concern equal sound. Here, the marks do not merely share similar ideas -they are one and the same name with the same pronunciation, only spelled differently. Such marks create the same sound impression on consumers (e. g. , Pepsi and पे1सी are similarly pronounced), and script difference alone is no protection against a conclusion of deceptive similarity under the phonetic test of Indian law. Traditionally, most reported cases involving cross-language marks in India were presented as
5 Bhatia Plastics v. Peacock Indus. Ltd., 2002 SCC OnLine Del 610 (India).
problematic translations, since the imitated mark would frequently possess a dictionary meaning (such as “Sun” or “Peacock”). True transliteration disputes -in which an English coined or arbitrary mark is imitated in an Indian script (or vice versa) have been relatively less common in reported decisions. This could be due to renowned trademark owners preemptively registering their significant marks in various scripts, or such oppositions are settled at the opposition stage without resulting in a reported judgment. However, notable cases and principles can be seen. Hindi and English Identical Names There was a recent Delhi High Court case directly confronting a transliteration dispute. In Anshul Vaish v. Hari Om & Co.
(2025), the petitioner had been using “ROHIT” (a given name, English) as a trademark for PVC pipes since 2000, and the respondent had subsequently registered “रोिहत” (the same name in Hindi script) for similar goods. The court held that the respondent’s mark was nothing other than the Hindi-script rendition of the petitioner’s mark, and was therefore sure to cause confusion. In February 2025, Justice Mini Pushkarna directed the Hindi “रोिहत” mark to be cancelled since it was effectively the same as the English “ROHIT” mark. The court relied on earlier precedents like Bhatia Plastics (Mayur/Peacock) and the Indian Express case of the Bombay High Court. In Indian Express Ltd. v. Shivhare6 (Bom.
HC 2015), a title of an English newspaper, Indian Express, was replicated in Hindi script by another; use was enjoined by the court, observing that a well-known English mark should not be taken over in Hindi in a way that leads to confusion for readers. These cases demonstrate that Indian courts consider an English mark and its Hindi transliteration to be fundamentally the same mark for purposes of confusion analysis. A significant early case involving phonetic identity between scripts is Hitachi Ltd. v. Ajay Kumar Agarwal (Del Delhi HC 1995). The famous Japanese electronics firm Hitachi tried to restrain a defendant from using “HITAISHI” in Devanagari script for electronic products. Although there were slight differences in spelling, the phonetic pronunciation of “Hitachi” and “Hitaishi” in Hindi was almost similar.
The High Court (Justice Anil Dev Singh) ruled that phonetic similarity was most likely to mislead consumers, and making alterations in the script or introducing minute variations would not prevent confusion. The court emphasized that since
6 Indian Express Ltd. v. Shivhare, 2015 SCC OnLine Bom 5135 (India).
India is a multilingual nation, trademarks must be protected from phonetic “piracy” across scripts. In essence, an imitator cannot escape liability by modifying the script (spelling or script) while maintaining the sound of a famous mark. In essence, modification of script was considered a minor variation that did not affect the mark’s uniqueness from the consumer’s point of view. At the registry level, the practice has been to reject marks that are clear transliterations of current marks. For instance, the Trade Marks Registry will most probably reject an application for “कोका-कोला” (Hindi for “Coca-Cola”) when Coca-Cola (in English) is already registered for drinks under Section 11(1) (and maybe Section 9(2)(a)).
The rationale would be that the two marks sound the same, and people perceiving the Hindi name (or hearing it used) would think of some association with the renowned Coca-Cola. The Registry’s own facilities -like phonetic search algorithms -tacitly recognize that the same or similar sounds, irrespective of script, constitute a basis for refusal. Script-Based Confusion under Sections 9 and 11 The case laws examined above illustrate that Indian courts do not shy away from holding deceptive similarity when marks vary by way of only language or script. The statutes (Sections
11 and 9) employ loose language such as “identical or similar” marks and probable confusion,
without necessarily defining language or script leaving it open to flexible interpretation to encompass cross-script disputes. The “novelty” of scripting is divorcing script-conversion as a separate issue is largely theoretical, in practice, it has been addressed on the general confusion analysis. The general language of Sections 11 and 9 has allowed courts to apply transliterated marks within the general confusion paradigm. There is no distinct category or test based merely on transliteration problems; rather, phonetic sameness is merely one type of similarity. In fact, in operation courts quite commonly regard a mark and its transliteration as essentially the same mark for purposes of confusion. For instance, the Delhi High Court in the cases of Rohit and Hitachi dealt with the Hindi-script marks as having the same substance as their English counterparts.
Examiners also normally object if an application’s transliteration takes the form of a pre-existing mark. It is normally presumed that if a subsequent mark sounds identical to a prior mark (particularly an invented or renowned mark), confusion will arise. Visual distinctions (alternate script or spelling) are accorded little significance in such instances since Indian customers habitually identify through sound and are used to changing scripts. As one court noted, altering the script of a well-known mark is no more useful than changing the font, the ordinary purchaser, hearing the name, would still be confused. The cautious practice of the Registry and judiciary is thus to regard identical pronunciations as potentially confusing unless very compelling reasons exist to the contrary.
Honest concurrent use defences are theoretically available, but in practice it is unusual to encounter an innocent adoption of a phonetically indistinguishable mark in another script. Courts tend to infer that a trader choosing a name identical in sound to an existing brand did so deliberately (absent evidence to the contrary). This ensures that unscrupulous businesses cannot exploit language differences to ride on another’s reputation. Meanwhile, fears that this approach grants over-broad monopolies have not materialized, since India’s linguistic diversity simply requires brand owners to be diligent in vetting names across major languages. Finally, preventing cross-script confusion is shielding consumers and preserving honest competition by excluding a firm from impersonating another by means of transliteration.
Conclusion India’s regime of trademarks, based on consumer protection and equity, has fared very well in the face of a multilingual nation. The situation of co-existence of a Hindi mark with its Latinalphabet counterpart name is not speculative it has arisen in courtrooms and at the Registry, and the uniform result has been to prohibit such coexistence if it would lead to consumer confusion. By provisions such as Section 11(1) of the Trade Marks Act, 1999, which protects against probable confusion with previous marks, Indian law has the necessary mechanism to refuse or invalidate phonetically equivalent marks in different scripts. The case law survey reveals that irrespective of the statute’s lack of explicit reference to “script” for the statute, the courts have in practice expanded the principle of deceptive similarity to cover script-based equivalents.
Initial rulings such as Eno Fruit Salt established the precedent of blocking misuse of translations that would cause deception to monolingual consumers. Later decisions -from Sunlight/Surian to Mayur/Peacock to Wilkinson Sword/Talvar -reaffirmed that the concept or pronunciation of a mark cannot be duplicated in another language without leading to confusion. Most recently, direct transliteration battles like English “Rohit” vs Hindi “रोिहत” or “Hitachi” vs “िहतैशी” affirm that a script change is a distinction without a difference in the mind of consumers. For trademark regulators and practitioners alike, the policy is obvious; when assessing likelihood of confusion, pronunciation matters.
If two marks, when pronounced, would be essentially the same (or very close) and relate to analogous goods or services, they must be considered conflicting -whether one is in English, Hindi, Tamil, or any other writing. The Indian Trade Marks Registry already applies this principle through phonetic search algorithms and by making transliteration of non-English marks compulsory in applications. In the future, care must be exercised to avoid unscrupulous traders misusing script differences to make attempts at registration of misleadingly similar marks. Overall, the handling of transliteration conflicts in Indian trademark law is consistent with the universal objective of avoiding consumer confusion. The lack of a specific doctrinal label has not prejudiced vigorous protection of brand names throughout India’s linguistic mosaic.
The available legal framework militates in favor of the argument that a mark and its vernacularscript equivalent are virtually indistinguishable identifiers. Indian registries and courts, therefore, stand by combining statutory basis with wise reasoning to ensure that identicalsounding marks written in distinctive scripts are considered likely to deceive and thus maintain trademark integrity in India’s multilingual market.