Trademark

A trademark is a word, name, symbol, device, or combination of these that a business uses, or intends to use, to distinguish its goods or services from those of others. It belongs to the family of intellectual property rights and is enforced through law. Its core function is to identify the commercial source of a product, which allows buyers to recognize quality and reputation without testing each item. In this sense a trademark also reduces the information-asymmetry between sellers and buyers, which explains why it is treated as an economic institution as well as a legal one.

The modern trademark system grew out of older practices such as medieval craft marks, hallmarks on precious metals, and anti-forgery rules. Statutory registration emerged in the nineteenth century, with the first comprehensive system in france and the first registration act in the united-kingdom soon after, followed by federal law in the united-states. Today rights arise either from use in commerce or from registration, and they are coordinated internationally through treaties administered by the world-trade-organization and the World Intellectual Property Organization.

This entry first defines the trademark and its kinds, then explains the requirements for protection and how rights are acquired and lost. It traces the historical development, the international registration systems, and the scope of protection and enforcement. It closes with digital-age issues, the economic debate over trademark's social effects, the relationship to other forms of intellectual property, and open controversies.

Definition and Legal Function

In the United States, the central statute is the Lanham Act, which was signed into law by President Harry S. Truman on July 5, 1946 and is codified in Title 15 of the U.S. Code. The statute reads a trademark broadly: it covers any word, name, symbol or device, or a combination of them, that a trader uses to separate its goods from those of others. Registration may also rest on a mark that is not yet in use, provided the applicant has a bona fide intent to use it in commerce, so a mark can be protected at the filing stage before any sales occur. Other legal systems use different wording, but most define the category in similarly functional terms.

The main purpose of a trademark is to tell buyers where a product comes from. Instead of reading the fine print on a can of cola, consumers can look for the Coca-Cola trademark. The mark is therefore separate from the product itself and from any patent or copyright that may protect an invention or creative work embodied in that product. A patent protects the technical solution and a copyright protects the expression; a trademark protects the identifier that links goods or services to a particular enterprise. Some U.S. jury instructions describe the mark as a limited property right in a particular word, phrase or symbol. It identifies the source of goods.

The sign must also actually function as an identifier in the market. A designation that the public never encounters does not perform this role. For example, a trademark fails to serve its source-identifying function when the public has never seen it, for instance when registered for an Internet domain name. Protection therefore depends on the mark being used in a way that creates an association between the sign and its owner in consumers' minds, not merely on the sign's formal existence.

Trademarks are generally treated as property that can be transferred, licensed and financed. Like other assets, a trademark can be sold, transferred, licensed, used as collateral for money financing, etc. The property character is qualified, however. Under the Lanham Act, a transfer must include the goodwill of the business in which the mark is used, and a trademark is only assignable along with the good will of the business in which the mark is used. A licence is likewise constrained. Where the use of a trademark is licensed (for example, to a franchisee) without adequate quality control or supervision by the trademark owner, that trademark will be at risk of losing its distinctiveness. Lenders taking a security interest face the same logic: keeping the associated goodwill attached maintain[s] value of the collateral.

The breadth of the definition allows non-visual signs to qualify. Literal readings of the statute are not restrictive, because human beings might use as a "symbol" or "device" almost anything at all that is capable of carrying meaning. U.S. authorities have accepted a bottle shape, a broadcast sound sequence and even a scent as marks. Acceptance is not automatic, though. What matters is whether the sign can distinguish one source from others, not whether it is unusual in appearance; the ability of a trademark to distinguish the source of the goods it marks, not the uniqueness of its color, shape, fragrance, word, or sign, entitles it to protection. Where two parties claim the same sign, priority generally follows use: exclusive rights to a trademark are awarded to the first to use it in commerce.

Types of Marks

Trademarks are commonly classified by the kind of market activity they identify and by the kind of sign they consist of. A trademark in the narrower sense distinguishes goods, while a service mark distinguishes services such as banking, transport or hotel accommodation. Many jurisdictions apply the same registration rules, examination criteria and infringement standards to both, so the distinction is mainly one of subject matter rather than of legal regime.

Certification and collective marks

A certification mark is owned by an organisation that does not itself sell the certified goods. Its owner licenses authorised users to display the mark, which signals that their goods or services meet standards set by the certifier. Those standards can concern origin, material, mode of manufacture, quality, accuracy or other characteristics. A collective mark is also owned by an organisation, but its function is to signal membership: members use it to show that their goods or services come from an enterprise that meets the association's admission standards. The hotel brand BEST WESTERN, used by its member hotels, is a frequently cited example. Some systems, such as Ireland's, allow any of the ordinary mark types to be registered in either of these collective or certification forms if the additional requirements are met.

Trade dress and shape marks

Trade dress refers to the overall visual appearance of a product, its packaging, or a service. Retail environments can form part of it, as with the layout of Apple stores. In the United States, the Lanham Act protects trade dress when it serves as a source identifier. Shape marks are a specific form of trade dress consisting of a recognisable three-dimensional form applied to a product or its packaging, such as the Coca-Cola bottle. Colour can also function as a mark. The United States Supreme Court accepted in Qualitex Co. v. Jacobson Products Co. (1995) that a colour may qualify when it has acquired source-identifying significance, and a distinctive blue box is associated with Tiffany & Co.

Non-traditional marks

Non-traditional marks cover signs that are not words or conventional logos. Sound marks, scent marks, motion marks, multimedia marks combining image and sound, and hologram marks are the most common categories. Irish law, for example, defines a motion mark as one extending to movement or a change in the position of the mark's elements, a multimedia mark as a combination of image and sound, and a hologram mark as one consisting of holographic elements.

National practice differs most sharply on how such marks must be represented. Historically, many systems required a graphic representation, which suits logos but fits sounds and smells poorly. The United States Trademark Act now states that no drawing is required for a mark consisting only of a sound, a scent, or other wholly non-visual matter, and a scent mark for plumeria-scented sewing thread was registered there in 1990. European practice has been stricter: a written description of a scent, even with a sample, was historically not enough for registration. EU implementing rules have since specified that sound marks may be represented by an audio file or accurate musical notation, and motion marks by a video file or a series of sequential still images. Australian law expressly lists shape, colour, sound and scent as registrable, and requires a recorded representation of a sound to be available to the public. In India, sound marks must still be graphically represented, typically by musical notation; the Yahoo yodel, registered in 2008, was the first sound mark registered there, and a Kingfisher jingle was registered as a sound mark with effect from January 2025. The United Kingdom has replaced its graphic-representation requirement with a clear and precise standard, while Malaysia continues to treat graphical representation as a central condition.

Registrability and Distinctiveness

Distinctiveness spectrum

Trademark law ranks signs by how strongly they identify a single source. At the top are arbitrary marks, which use ordinary words in unrelated contexts, such as APPLE for computers, and fanciful marks, which are invented words with no dictionary meaning, such as KODAK. Suggestive marks sit in the middle. They hint at a quality or characteristic of the goods but require some imagination to connect the two, as with COPPERTONE for sun-protection products. The U.S. Court of Appeals for the Second Circuit set out this ordering in Abercrombie & Fitch Co. v. Hunting World, Inc. (1976), which placed generic terms at the bottom, below descriptive ones. Arbitrary, fanciful and suggestive marks are regarded as inherently distinctive and can be registered without proof of market recognition. The same scale underlies the approach of most national systems that follow the U.S. framework, although the exact labels vary.

Descriptive marks and secondary meaning

Descriptive marks directly convey an ingredient, quality, function, purpose, or geographic origin of the goods or services, so a term such as "quick-dry" for towels describes the product rather than naming its maker. Because competitors need such words to describe their own offerings, most systems refuse registration to descriptive marks unless the applicant shows acquired distinctiveness, or secondary meaning, meaning that the public associates the term with one undertaking. Evidence typically includes sales volume, advertising expenditure, the length and exclusivity of use, and consumer surveys. Under Section 2(f) of the united-states Lanham Act, five years of substantially exclusive and continuous use may serve as prima facie evidence of this association. The european-union trade mark system and the British Trade Marks Act 1994 contain comparable provisions. Even after registration, the owner generally cannot prevent others from using the descriptive word in its ordinary sense.

Generic terms, which name a category of goods or services, cannot be protected at all, regardless of how much secondary meaning the applicant can show. This category is the starting point for the later problem of genericide, which is covered elsewhere in this entry.

Absolute grounds for refusal

Absolute grounds are assessed on the sign itself, without reference to rights held by others. They include lack of distinctiveness, descriptiveness and genericness, as described above. They also cover deceptive marks, which mislead the public about the nature, quality, or geographic origin of the goods. Signs contrary to public order or morality are refused in most systems, and the Paris Convention allows members to refuse registration of such marks as well as official state emblems and hallmarks. The Lanham Act bars marks that are immoral, deceptive, or scandalous, that disparage persons or institutions, that falsely suggest a connection with them, or that consist of flags or national symbols. European law also excludes shapes that result from the nature of the goods, are needed to achieve a technical result, or give substantial value to the goods, and it refuses applications filed in bad faith.

Relative grounds for refusal

Relative grounds concern conflicts with rights already held by third parties. The most common is a likelihood of confusion with an earlier mark, where the marks are identical or similar and the goods or services are identical or similar. The Lanham Act refuses registration on this basis when a mark resembles a registered or previously used mark. In the European Union, the EU Intellectual Property Office does not examine earlier rights on its own initiative; the holder of an earlier right must raise the objection in opposition proceedings. Other relative grounds include conflicts with well-known marks, earlier names, and personal or copyright rights.

Functionality as a bar to protection

Functional features cannot be monopolized through trademark law because doing so would restrict competition in the underlying technology or design. In Inwood Laboratories v. Ives Laboratories (1982), the U.S. Supreme Court held that a product feature is functional if it is essential to the use or purpose of the article or affects its cost or quality. In TrafFix Devices v. Marketing Displays (2001), the Court confirmed that a feature that is functional in this sense cannot be protected, and that the existence of an alternative design does not by itself make a functional feature protectable. The Lanham Act codifies this bar, and the Qualitex v. Jacobson Products decision (1995) confirmed that color can function as a mark while remaining subject to the same doctrine. European case law under the Court of Justice of the European Union, such as Philips v. Remington (2002), applies the technical-function exclusion to shapes and treats aesthetic and technical functions differently. Functional features that are genuinely novel are instead the domain of patent protection, which has a limited term and different requirements.

Acquisition of Rights: Use and Registration

Trademark systems differ on what creates a right in the first place. In common-law jurisdictions, actual use of a sign in the marketplace can generate protectable rights. Systems grounded in use persisted in Britain and Canada until each converted to a different framework. Many civil-law jurisdictions, by contrast, rely mainly on the register, so a filing and registration usually determine who holds the enforceable right, and earlier unregistered use gives limited protection. Many systems therefore combine both models, with unregistered use protected in limited ways and registration providing the core right.

Effects of Registration

Registration generally strengthens an owner's position. Under older British practice, a name entered on the register was treated as presumptively the proprietor, and an interested party had to rebut that presumption; the presumption would stand before the Patent Office until an interested party challenged it. In the United States, a federal registration gives public notice of the owner's claim, so a later user cannot defend infringement by saying it was unaware of the mark. Priority is also a major consequence. Under the Lanham Act, an application constitutes constructive use of the mark, which must be resolved through registration or another disposition before a later application for the same mark may be registered. The exact scope of these effects depends on national rules, and registration rarely removes the need to show likelihood of confusion in enforcement.

Intent-to-Use Applications

Requiring use before filing was the older norm. Until 1989, a U.S. applicant had to use its mark in interstate commerce before applying to register it, and a use-based application filed before such use was void from the outset. Congress had considered intent-to-use proposals several times, and none of these attempts succeeded before the Trademark Law Revision Act of 1988. The Act, signed on 16 November 1988, took effect a year later. It created a dual system: marks already in commercial use could still be registered on that basis, while applicants could now seek registration of a mark before its use, provided they held a genuine plan to use it. Registration was then conditional on later proof of use, and the applicant's filing date could serve as the constructive use date.

The United Kingdom had reached a similar result earlier. The Trade Marks Act 1938 let applicants register on the basis of a bona fide intention rather than existing use, but it policed that intention after the fact. A registration could be removed if it was registered without any bona fide intention on the part of the applicant that it should be used in relation to those goods, combined with the absence of genuine use over the relevant period. The 1938 Act also allowed applicants who planned to form a company and transfer the mark to it to register without immediate use, a feature that the law review literature treats as an early form of intent-based filing (the Act specified that registration would not be refused for lack of use or intent to use if the applicant intended to form a corporation or assign the mark to an entity that would use it).

Notation Symbols

The ™ and ® symbols carry different legal meanings. The ™ symbol signals a common law trademark claim, and parties may use terms such as "TM" regardless of whether the mark is registered. The ® symbol is reserved for registered marks: it may only be used after the USPTO has issued a Certificate of Registration. It should also be confined to the goods or services listed in the registration (the federal registration symbol should be used only on or in connection with the goods or services that are listed in the federal registration). Misuse carries real costs. Improper use of ® may provide a basis for the examiner to deny registration or for a third party to oppose it, and in the marketplace may support charges of fraud, false advertising, or unclean hands. Deliberate misuse is treated more seriously: a knowing and willful use of the symbol intended to mislead consumers or others in the trade may amount to fraud. Failing to display ® on a registered mark has the opposite effect; in the United States, a rights holder who omits the symbol generally cannot recover profits or damages from an infringer unless that infringer had actual notice.

Early Origins

Marks that show who made an object predate modern trademark law by millennia. One survey of the subject places the earliest marking of pottery in prehistory, more than seven thousand years ago, when makers used marks to show where a vessel came from. Archaeological findings indicate that our prehistoric ancestors began applying special marks to their pottery to distinguish their origin. Marking became a legal duty only later, with the rise of medieval craft guilds, whose marks were used to show that goods met the guild's standards. In these systems the marks were often compulsory, and their purpose was to let buyers identify the party responsible for defective goods.

The clearest early examples of compulsory marking came from food and metalwork. In 1266 England required bakers to stamp their loaves, so that officials could check weight and price against the rules, and bakers who broke the rules could be banished from the trade for life. Weapons were regulated in a similar way: by the 14th century, English manufacturers were required to put a personal mark on blades, and misusing another maker's mark could be pursued in court. A Parma law of 1282 went further, forbidding a guild craftsman from using a mark identical or similar to another member's, with a fixed penalty of ten pounds per offence. Medieval marking therefore served as a tool of regulatory control, not a right in the sense later given to a brand owner.

English hallmarking of gold and silver shows the same logic in its most developed form. A statute of 1300, passed under Edward I, required that gold and silver be tested before sale. Wardens of the Goldsmiths' craft visited workshops to assay the metal, and the leopard's head became the mark of approved silver. The word "hallmark" is usually traced to the practice of bringing wares to Goldsmiths' Hall, where they were assayed and stamped. Later additions made the system more accountable: a maker's mark was added in 1363, and a list of makers and their marks was kept at the Hall, so that a maker of substandard metal could be identified and prosecuted. The date letter, introduced about a century after the original statute, let officials trace which assayer had tested a piece. Sources differ on the precise date of the Goldsmiths' royal charter, with both 1321 and 1327 cited, but they agree on the core arrangement of statutory testing and official marks.

The enforcement behind these systems was criminal in character. Hallmarking was justified as protection for the public against fraud and for honest traders against unfair competition. Counterfeiting a hallmark became a felony in 1757, punishable by death, and the Goldsmiths' Company held a right of search that let it enter workshops and markets to seize substandard goods. Early marking rules thus rested on the criminal law of forgery, counterfeiting and fraud, and it was the consumer and the market that the rules protected, not a proprietary interest in a name.

The private right that became modern trademark law developed later, through the civil law of unfair competition. Guild marks gradually changed their character as goods began to travel long distances, and by the 17th century the English cloth trade distinguished the manufacturer's required mark from the clothier's trademark, which carried goodwill belonging to the merchant who financed production. The decisive step came through the action of passing off. Passing off was first recognised in England during the reign of Elizabeth I, and for centuries it was treated as a form of the tort of deceit, requiring proof of a false statement and an intention to injure. Over time the focus moved from the defendant's intent to the effect on consumers, so that intent to deceive ceased to be required. Trademark infringement in turn evolved out of unfair competition, which had grown from the common-law action for deceit and related equitable claims, and in the United States trademark law has its roots in state unfair competition laws that prohibited passing off.

Nineteenth-Century Codification

The first comprehensive modern trademark statute was France's law of 23 June 1857 on marks of manufacture and trade. It made registration a matter of deposit: the owner lodged a representation of the sign with a designated public registry, and that filing, rather than a judicial finding of prior use, fixed the claim. The French model gave continental systems a template based on administrative recordal.

Britain followed a different path, rooted in criminal law. The Merchandise Marks Act of 1862 made it an offence to forge or imitate a mark with intent to defraud, so the state prosecuted misuse instead of leaving enforcement to private suits. The Trade Marks Registration Act of 1875 then created a formal register, which opened on 1 January 1876. It was registered on 1 January 1876 under a new piece of legislation, the Trade Marks Registration Act (1875). Under this act no-one was allowed to take legal action to prevent infringement of a mark unless it was registered. The first entry, the red triangle of the Bass brewery, is widely cited as UK Trade Mark No. 1. In 1876, the Bass red triangle became the first officially registered trade mark in the United Kingdom, securing its place as UK Trade Mark No. 1. A commonly repeated account adds that a Bass employee queued overnight at the registrar's office to be first in line, although that detail rests on anecdote.

The United States attempted federal regulation earlier, in 1870, but its approach proved fragile. The statute was attached to a larger revision of copyright and patent law, and it allowed registration of marks used in interstate and foreign commerce. The federal protection proved short-lived, however, when in 1879 the Supreme Court in the Trade-Mark Cases held that the statute had been unconstitutionally grounded on the patent and copyright clause of the Constitution. The Court's reasoning meant that Congress had no authority to regulate marks under that clause, and the Court also rejected the commerce power as a basis without an explicit interstate limit. The Court quickly dismissed talk of using the copyright clause of the Constitution as the basis for the laws, and found that without an explicit limitation to interstate commerce, the law could not be premised on the commerce clause.

Congress responded with narrower legislation. The 1881 act rested on commerce-clause powers, but in reaction to the Trade-Mark Cases, registration under the act was limited to marks used in commerce with foreign nations and the Indian tribes. A broader federal revision followed in 1905, and the modern federal registration system developed from that point.

These statutes emerged alongside the expansion of packaged, branded consumer goods made in factories and sold far from the place of production. As manufacturers sold to distant buyers who could not inspect the maker, a name, device or label became a way to carry reputation across markets. Legislation in France, Britain and the United States therefore dealt with the same underlying problem: the protection of identifying symbols on goods whose quality buyers could not easily verify on the spot.

Twentieth-Century Reform and the Lanham Act

Britain: the 1905 Act and the 1938 consolidation

The British Trade Marks Act of 1905 dropped the older requirement that an infringing mark be identical to the registered one or that the defendant intend to deceive. In its place it adopted the broader test of likelihood of confusion, which later became the central question in infringement disputes in many systems. The 1905 Act was amended over the following decades, and in 1938 Parliament consolidated it with the Trade Marks Act of 1919 and the 1937 amending act into a single statute, which took effect on 27 July 1938 and remained in force until the United Kingdom replaced it in 1994.

The 1938 Act reorganized registration around examination. Applications were assessed by the Registrar before acceptance, and accepted applications were published so that third parties could oppose them. The register was divided into Part A, which required a mark to be inherently distinctive, and Part B, which required only that the mark be capable of distinguishing goods. The Act also introduced several concepts that were new to British law. Associated trade marks, which were registered as a group, could be assigned or transmitted only as a whole, so that closely similar marks could not be split between owners. Defensive registration allowed the proprietor of a well-known invented word to register that word for goods it did not intend to sell, so that the mark could not be used by others on related goods in a way that suggested a connection with the proprietor.

Influence on other systems

The 1938 model was widely imitated. Its examination-based registration, publication-and-opposition procedure, and treatment of associated marks shaped trademark statutes in several other countries, and Japan's postwar trademark legislation is generally described as drawing on the same examination-based approach. Its influence was thus not limited to the Commonwealth, and it also served as a point of reference for debates in the United States about how federal registration should work.

The Lanham Act of 1946

Before 1946 the United States had no comprehensive federal trademark statute. Earlier federal attempts, including the 1870 law struck down in 1879, had given little protection to trademark users, and most trademark law was left to state statutes and common law. A reform movement that began in the 1920s was championed in the 1930s by Representative Fritz G. Lanham of Texas. The resulting Trademark Act of 1946 was signed by President Harry S. Truman on 5 July 1946, and it is now codified at 15 U.S.C. § 1051 et seq.

The Act had several main features. It established a national registration system administered by the Patent and Trademark Office, and it gave registered owners access to federal civil remedies. Its Senate report stated that the aim was to place trademark matters in one statute and to simplify registration. The Act also extended protection beyond registration: it protected unregistered marks used in commerce through what is now section 43(a), which covers false designation of origin. The Department of Justice opposed the bill on the ground that trademarks could operate as monopolies favoring large businesses, and that debate over the monopoly character of trademark rights has continued ever since. The Act was amended several times, most extensively by the Trademark Law Revision Act of 1988.

The spread of service mark registration

Service marks were treated as a distinct category in both countries during the late twentieth century. In the United States, the 1946 Act made service marks registrable on the same terms as trademarks, but they were kept on a separate register. The 1988 Trademark Law Revision Act removed that separate register, and the same statute created the intent-to-use application, which took effect in November 1989. In the United Kingdom, the Trade Marks (Amendment) Act of 1984 introduced registration for service marks, and parliamentary debate on the 1980s reform had already stressed that the growth of service industries made statutory protection for service marks increasingly important. Registration for services thus became a standard feature of national systems in industrialized economies during the 1980s and 1990s, and it was carried into the 1994 British statute that replaced the 1938 Act.

International Harmonization

International cooperation on trademarks rests mainly on the Paris Convention. The Paris Convention for the Protection of Industrial Property is a multi-national treaty, established in 1883, and administered by WIPO. Its aim was to reduce differences in how industrial property was protected, and two principles from it still underlie most cross-border trademark arrangements. The first is national treatment: a member state must extend to foreigners the protection it gives its own nationals. The second is the right of priority. An applicant who files in one member state has six months to seek protection in the others, and those later filings keep the date of the first application. Subsequent applications are regarded as filed on the same day as the first, so they outrank applications for the same mark filed by others during that six-month period.

The Convention leaves the substance of registration to each member. Its Article 6 provides that the conditions for the filing and registration of trademarks shall be determined in each country of the Union by its domestic legislation. A member may not refuse a foreign national's application, or invalidate a registration, merely because the mark was not registered in the applicant's home country.

The TRIPS Agreement of 1994 extended these principles to the World Trade Organization. Adopted in 1994 as part of the Uruguay Round of trade negotiations, TRIPS is the most far-reaching international intellectual property agreement in terms of both scope and membership. Its trademark rules occupy Articles 15 to 21. Any sign, or any combination of signs, capable of distinguishing the goods or services of one undertaking from those of other undertakings, shall be capable of constituting a trademark. Where a sign is not inherently distinctive, members may condition registration on distinctiveness acquired through use. TRIPS also requires that service marks must be protected in the same way as trademarks used for goods. Enforcement obligations are included as well: the agreement establishes minimum standards for civil actions for infringement and, at least for copyright piracy and trademark counterfeiting, for criminal actions and border measures. The agreement works as a floor rather than a ceiling, since it allows Members to provide more extensive protection of intellectual property if they so wish. Obligations also phase in unevenly: developing countries will have a longer period to phase them in. Disputes over compliance are handled through the WTO's Dispute Settlement Understanding.

Harmonization has clear limits. TRIPS does not prescribe how trademark rules must be built into domestic law. Members are left free to determine the appropriate method of implementing the provisions of the Agreement within their own legal system and practice. The result is not a global trademark. Rather than creating a single worldwide right, TRIPS ensures that WTO members provide minimum standards of trademark protection to each other's citizens. Each state still decides which signs it will protect under its own law, and the same mark may be owned by different parties in different countries. Unifying procedures, filing rules and minimum standards therefore does not make trademark law supranational in substance.

Regional systems go further in a limited way. The European Union trade mark (EUTM) is the most developed example. The EU trade mark system creates a unified trade mark registration system in Europe, whereby one registration provides protection in all member states of the EU. The right is administered by the EUIPO in Alicante, Spain, and an EU trade mark registration lasts for 10 years but can be renewed indefinitely in blocks of 10 years. Because the right is unitary, its weakness is also shared. An objection against an EU trade mark application in any member state can defeat the entire application. A failed application can be converted into national filings in the states where no conflict exists, keeping the original filing date. National trademark systems, and the separate Benelux system, continue to operate alongside the EU route.

The Madrid System

The Madrid System for the International Registration of Marks is governed by the Madrid Agreement, concluded in 1891, and the Protocol relating to that Agreement, concluded in 1989. The 1891 treaty was signed by France, Switzerland, Tunisia, Spain, and Belgium under the framework of the 1883 Paris Convention. The Protocol, adopted in Madrid on June 27, 1989, entered into force on December 1, 1995. The system is administered by the International Bureau of the World Intellectual Property Organization in Geneva, which is the main reason it is often described as a WIPO system.

The Protocol was largely a response to the original agreement's rigidity. Under the 1891 text, applications had to be completed in French, and an applicant first needed a fully registered mark at home. The Protocol relaxed both conditions. It allowed filings in English as well as French, and it permitted an international application to rest on a national/regional application, in addition to an existing registration. These changes made the system acceptable to a wider membership, and the United States, which had stayed out for more than a century, joined in 2003.

In practice, an applicant with a home-office mark files a single international application through that office, which forwards it to WIPO. The filing is made in one language, pays one set of fees, and names the countries where protection is wanted. Each designated country then requires its own fee. The WIPO basic fee is CHF 653 for a black-and-white mark and CHF 903 for a colour mark, plus a fee for each designated state. The procedure therefore replaces a separate filing, translation, and fee schedule in every target market with one coordinated file.

The system is a filing treaty, not a substantive harmonization of trademark law. WIPO checks only formal matters such as fees, the identification of goods and services, and the match with the basic mark. Each designated country examines the application under its own law and can refuse it. Provisional refusals must generally be issued within a fixed period; in practice this is commonly within 12 or 18 months, and some offices may later raise opposition-based refusals. Consequently, the same sign may be accepted in one designated state and rejected in another on the basis of local criteria for distinctiveness, prior rights, or public order.

The most significant structural weakness is dependency on the basic mark. For the first five years after the international registration date, the bundle of designations rests on the home registration or application. If the basic mark is refused, withdrawn or cancelled during that window, all designations can fall together. This is known as central attack. Every country in the international registration can fall with it, even ones where nobody objected. After the five-year period, the registration becomes independent of the basic mark. The Protocol also provides a remedy: a failed international registration may be converted into national applications in the designated states, which preserves the original filing date for each.

Central attack is generally regarded as uncommon but consequential. Practitioners observe that successful central attacks tend to hit businesses whose home registration was already shaky or contested before they filed internationally. Some commentators have reported a rise in such cancellations, from just 200 total or partial cancellations in 1996 to 7,000 cancellations in 2017 alone, although the figures are drawn from secondary sources. The usual mitigation is to file internationally only after the basic mark has passed examination and appears robust, so that the most vulnerable period is spent on a secure foundation.

Scope of Protection and Infringement

Likelihood of confusion

Infringement of a registered or common-law mark generally turns on whether an accused sign is likely to confuse consumers about the source of goods or services, or about their sponsorship, approval, or affiliation with the mark owner. Most systems apply a multi-factor assessment rather than a single bright-line rule. In the United States, federal circuits developed their own lists: the Second Circuit's Polaroid factors (1961) and the Ninth Circuit's Sleekcraft factors (1979) both weigh the strength of the mark, the similarity of the marks, the proximity of the goods, evidence of actual confusion, marketing channels, the purchaser's degree of care, the defendant's intent, and the likelihood of expansion. The Court of Justice of the European Union takes a "global appreciation" approach, requiring the court to consider all relevant factors together in the eyes of the average consumer (Sabel v. Puma, 1997).

Dilution: blurring and tarnishment

Dilution protects famous marks against harm that does not depend on confusion. The U.S. Federal Trademark Dilution Act of 1995, later revised by the Trademark Dilution Revision Act of 2006, recognizes two main forms of injury. Blurring weakens the distinctiveness of a mark by associating it with unrelated goods, so that a single famous name gradually stops calling one origin to mind. Tarnishment occurs when a mark is linked to shoddy, unwholesome, or sexually explicit contexts that damage its reputation. The U.S. standard requires that a mark be widely recognized by the general consuming public. The European Union and the United Kingdom also protect marks with a reputation against unfair advantage or detriment taken without due cause, but the threshold and the scope of the standard differ across jurisdictions. Several other systems, including many that protect only confusingly similar marks, do not extend this protection generally.

Related goods and services

Protection is not limited to identical goods. Courts assess whether the goods or services are related enough that consumers might assume a common origin, and whether a senior owner could reasonably expand into the defendant's market. Similarity of the marks is judged by appearance, sound, meaning, and overall commercial impression. The EU's "interdependence" principle, articulated by the Court of Justice in Canon (1998), holds that the greater the similarity between the goods or services, the less similarity between the marks is needed to find a likelihood of confusion, and the reverse also applies. Courts also consider the channels through which goods reach buyers, since overlapping distribution increases the chance of confusion, and the degree of consumer care, since sophisticated purchasers of expensive products are less likely to be misled than impulse buyers of inexpensive ones.

Limits: fair use and parody

Several doctrines restrict trademark rights. Descriptive fair use, codified in Section 33(b)(4) of the Lanham Act, permits a person to use a term in good faith, other than as a mark, to describe its own goods or services. Nominative fair use allows a party to refer to a competitor's trademarked product where no other reasonable way exists to identify it. The Ninth Circuit's test in New Kids on the Block v. News America (1992) asks whether the product is readily identifiable without the mark, whether only the reasonably necessary part of the mark is used, and whether the user suggests sponsorship. European law provides analogous limitations for indications of intended purpose, such as accessories or spare parts. Parody can defeat confusion claims when it is clearly recognizable as a joke, but it does not automatically avoid liability: in Jack Daniel's Properties v. VIP Products (2023), the U.S. Supreme Court held that a parodic use that serves as a source identifier for the defendant's own goods is not shielded by the Rogers test, although the parody remains relevant to the confusion analysis.

Remedies and Enforcement

Trademark enforcement takes place within national legal systems, but international minimum standards now determine which remedies must be available. Under the World Trade Organization's TRIPS Agreement, tribunals must be able to grant three broad kinds of relief: orders that halt infringing conduct, compensation for harm, and measures that remove infringing goods from commerce or destroy them. Judicial authorities must be able to award three types of remedies Interim relief is also required, so authorities must have the authority to order certain remedies, including provisional or interim remedies.

Injunctions are central to most trademark disputes. A permanent injunction generally follows a finding of infringement, while a preliminary injunction usually requires the claimant to show a likely success on the merits and a risk of harm that money cannot repair; these tests differ between countries. Monetary relief may be measured by the owner's losses, by the infringer's gains, or, in some statutes, by a fixed sum set by the court. Where appropriate, the infringing goods, and the materials and implements used mainly to produce them, can be seized, forfeited and destroyed.

Criminal enforcement is narrower than civil relief. Under TRIPS, members must provide criminal procedures and penalties at least in cases of wilful trademark counterfeiting on a commercial scale, and the sanctions must be severe enough to deter, potentially including imprisonment or fines. Other forms of infringement, including patent infringement, are outside the treaty's mandatory criminal requirement, so there is no obligation to apply criminal procedures and remedies to them. Many national systems go beyond this minimum.

Border measures allow rights holders to act before counterfeit goods reach the market. Rights holders can enlist customs services, which may intercept suspect shipments at the frontier and block their release. The obligation is strict only for certain categories: the requirements are mandatory only where imports of counterfeit trademark or pirated copyright goods are concerned. Some WTO members apply border measures to goods that infringe other IP rights and to infringing goods destined for export, but this is only optional under TRIPS. The window is short. Under TRIPS, customs suspension is typically around ten working days, extendable in appropriate cases, after which the holder must start court proceedings or the goods are released. In the European Union, Regulation (EU) No 608/2013 governs this procedure.

Administrative procedures sit alongside courts. Some national legal systems deal with certain IP enforcement cases through administrative procedures, rather than court proceedings. If so, the TRIPS Agreement requires the same principles to be applicable to them to the extent that civil remedies can be ordered. In practice, opposition lets third parties challenge an application during a publication window before registration, while cancellation or invalidity proceedings attack an existing registration. In the United States, cancellation petitions on most grounds may be filed within five years of registration, and the grounds narrow after that period. In the EU, the EU Intellectual Property Office handles both opposition and invalidity actions.

Disputes over domain names have their own streamlined tools. The Uniform Domain-Name Dispute-Resolution Policy, established by ICANN, applies to most generic top-level domains such as .com and .net. A complainant must show that the domain is identical or confusingly similar to a mark it holds rights in, that the registrant has no legitimate interest in the name, and that the name was registered and used in bad faith (The domain name is identical or confusingly similar to a trademark in which the complainant has rights. The respondent has no legitimate rights in the domain name at issue. The respondent registered and is using the domain name in bad faith.). A panel cannot award money; the only available relief is relinquishment of the domain name.

In the United States, the Anticybersquatting Consumer Protection Act offers a court route. Codified at 15 U.S.C. §1125(d), it became operative on 29 November 1999. A plaintiff may elect statutory damages instead of actual damages and profits. The election must occur before the trial court enters final judgment, and the statute sets a range of $1,000 to $100,000 per domain name as the court considers just. Actions directed at the domain itself are narrower, and the remedy is limited to forfeiture, cancellation, or transfer. In "exceptional cases," attorney's fees can also be recovered. The Act does not reach legitimate uses: it does not prohibit a registrant's fair use of a domain name, or any uses protected by the First Amendment.

Duration, Maintenance, and Loss of Rights

Trademarks differ from patents and copyright works in that they have no fixed expiry date. A registered mark can in principle protect a brand indefinitely, but only while the owner meets periodic renewal requirements and, in common-law systems, keeps the mark in use. In the United States, the registration cycle runs in ten-year steps: a declaration of use falls due in the ninth year, and a renewal application must be filed before the tenth year ends, with the same pattern repeating thereafter. Many other jurisdictions, including the European Union and the United Kingdom, use ten-year registration terms that can be renewed.

Rights that arise from use alone last only as long as the use continues. Under the Lanham Act, rights in the United States are acquired and kept through use in the ordinary course of trade, so a mark that is no longer used gradually loses the basis for its protection. Those rights are also tied to the use itself, not to a registration certificate, so a company that stops selling under a name generally cannot rely on earlier use to hold it indefinitely.

Abandonment and the presumption of nonuse

Under the Lanham Act, a mark is abandoned when its use has stopped and the owner does not intend to resume it, and that intent may be inferred from the surrounding circumstances. Three consecutive years of nonuse count as prima facie evidence of abandonment. Once that period is shown, the burden shifts to the owner to prove either use during the gap or a concrete intent to resume use within a reasonably foreseeable time. Courts have held that a plausible excuse for nonuse, such as lengthy litigation, does not by itself stop the statutory period from running, and that only a real intention to resume in the foreseeable future rebuts the presumption. A brief interruption is generally insufficient to establish nonuse, and even minimal sales of a covered product can defeat a nonuse claim. Other jurisdictions set their own periods and evidentiary rules.

Abandonment is usually not automatic. It typically has to be established through a failure to meet a USPTO maintenance deadline or through a cancellation proceeding. A successful challenge removes the registration and the presumption of validity that accompanies it, and abandonment can also defeat common-law rights.

Genericide

Genericide is the process by which a mark loses protection because the public uses it as the name of the product category rather than as an indicator of a single source. The loss is driven by the mark's own success: a name that becomes the ordinary word for a product type is no longer able to identify one maker. Courts have treated this as a question of consumer perception, asking what buyers understand the word to mean. In the classic case concerning aspirin, Judge Learned Hand framed the inquiry around what buyers understood the term to refer to, and the name was found generic in the United States in 1921, after which competitors could sell the drug under that name, although Bayer still holds the mark in several other countries.

The escalator illustrates a similar path. The term began as a brand name for moving staircases, but by 1950 the Trademark Trial and Appeal Board had cancelled the registration, which was held by Haughton Elevator Company in the account of one legal analysis, after the public had come to treat the word as the name of the technology. Other accounts name Otis Elevator as the original brand owner, so the record on which company held the mark is not uniform. Other former brands, including thermos, cellophane and yo-yo, followed a similar route.

Genericide is not inevitable. In a 2017 Ninth Circuit decision concerning GOOGLE, the court held that using a trademark as a verb did not by itself make the mark generic, because the public still recognized the term as pointing to one search-engine provider. Owners commonly reduce the risk by using the mark as an adjective attached to a generic noun, by reminding the public through advertising and disclaimers that the term is a brand, and by objecting to generic usage in the press and in competitors' materials.

Maintenance duties

Rights remain enforceable only if the owner keeps them active. This involves filing the required declarations and renewals on schedule, using the mark in commerce in the form registered, and monitoring the market for infringing and generic uses. A registration that lapses through a missed deadline, or a mark that is left unpoliced while competitors use it generically, can be lost even though the owner never intended to give it up.

Trademarks in the Digital Economy

Online commerce has moved many trademark disputes from shop counters to search engine results, marketplace listings and account names. The main areas of litigation are keyword advertising, marketplace counterfeits and the identity of accounts and domain names.

Keyword advertising and metadata

Domain names raise the oldest version of the problem: a registrant takes a name that matches someone else's mark, which is treated in the remedies section. Keyword advertising raises a newer question. In Google France v Louis Vuitton (2010), the Court of Justice of the European Union found that search engine operators such as Google do not themselves infringe trademark rights if they allow advertisers to use a competitor's trademark as a keyword. The advertiser, not the search provider, is the party that "used the brand name in trade, in relation to goods and services", and liability for the advertiser depends on whether the advertisement lets an average internet user tell whether the goods come from the mark owner or from a third party. Decisive is mainly whether the text of the ad is sufficiently transparent, and the court left that factual assessment to national courts.

National courts have not always followed this approach. A French court in 2005 ordered Google to stop displaying competitors' advertisements against searches for a hotel chain's mark, with daily penalties for delay. In the case of Geico, a car insurance company, a US court ruled that Google's adword sales were consistent with other comparative advertising practices in the states, which shows how the treatment of comparative advertising differs between systems. Similar questions arise when a mark appears in hidden page metadata or tags, where courts ask whether the use is a use in trade that could mislead consumers.

Online marketplaces and counterfeit listings

Marketplaces face liability for listings they host. In L'Oréal v eBay (2011), the CJEU held that online operators do not "use" a trade mark by merely allowing its customers to market goods with third party trade marks on its platform. However, online marketplace operators may be responsible for trademark infringements carried out by users on their platform if they played an "active role" in facilitating the infringing conduct, such as by optimising or promoting offers. The ruling also held that a platform may lose the hosting exemption where it had information showing that a sale was unlawful and did not act promptly to remove the item or block the sale. Courts may also order platforms to take measures against repeat infringements. Although the court held that marketplaces are not under a general obligation to police items sold via their sites, the specific notices the operator received created duties to act.

In the United States, Tiffany v eBay followed a different route based on contributory liability. The Second Circuit affirmed in 2010 that "For contributory trademark infringement liability to lie, a service provider must have more than a general knowledge or reason to know that its service is being used to sell counterfeit goods. Some contemporary knowledge of which particular listings are infringing or will infringe in the future is necessary." Because eBay removed listings once Tiffany identified them, the court found no liability for general awareness that counterfeits were circulating. The contrast between the two regimes is central to notice-and-takedown policy: European law ties liability to knowledge of a specific unlawful item and the speed of response, while the US test focuses on specific knowledge and whether the service continues to supply the infringing seller. More recent EU rules on notice-and-action for online intermediaries have made these procedures a formal legal obligation for platforms.

Social media accounts and usernames

Account names and handles raise a different set of questions. A username that copies a brand name can suggest that the account belongs to the brand, which engages the same confusion analysis used for other uses in commerce. Parody accounts, fan pages, criticism and complaint accounts sit close to the boundary. Courts and platform policies generally ask whether the name is used to identify the account holder, whether the account presents itself as affiliated with the brand, and whether the content is commentary rather than a substitute for the brand's own communications. Impersonation of executives or brands in profile details, and the use of logos in avatars, are common areas of complaint, and the answers depend heavily on the facts and the jurisdiction.

Platform brand-protection tools

Many large platforms now offer rights-owner programmes that allow brand owners to register their marks, report listings in bulk, and apply automated filters to new postings. Such tools matter most for cross-border enforcement, because counterfeit goods and accounts can be created in one country and sold or used in another, while the legal rights remain national. These systems complement, but do not replace, court actions and customs measures. Their effectiveness depends on how quickly they remove listings and whether they allow repeat offenders to be identified across platforms.

Economic Functions and Theory

The most influential economic account of trademarks comes from William Landes and Richard Posner. Their 1987 article argued that trademark doctrine, like tort law more broadly, is best explained as an attempt to promote economic efficiency. Their central claim is that a mark's economic function lies in cutting the information costs that buyers face, whether those costs take the form of travel between shops, time spent searching, or effort spent comparing options (the transaction costs incurred in the process of acquiring information). In this framing, a mark is a cheap answer to the information asymmetry between sellers and buyers, since clear and reliable signs are a cost-effective way of addressing the information asymmetry between suppliers and consumers.

The search-cost argument also explains why owners have reason to keep quality stable. A mark is useful to a buyer only if the branded product is recognisably the same from one purchase to the next, so the benefits of trademarks in reducing consumer search costs require that the producer of a trademarked good maintain a consistent quality over time and across consumers. Hence trademark protection encourages expenditures on quality. This signaling role matters most for goods whose qualities are learned only after purchase. Economides argues that, for such experience goods, where the buyer links the mark to the product before buying, there is no other mechanism which would work as efficiently. A consistently high-quality producer is also well placed to capture the price premium that reduced search costs make possible (consistently high quality are best placed to capture the trademark "premium" associated with a trademark's reduction in search costs.).

The value that accumulates in a mark beyond the underlying product is usually called brand equity. Added value attached to the basic product is often broadly termed "brand equity". On the demand side, this added value can comprise consumption benefits such as image and information benefits such as quality reputation. Protection supports investment in that reputational capital. Landes and Posner observed that unchecked free riding would erode the information capital embodied in a mark, and the prospect of such free riding could remove the incentive to build a valuable mark at all (free riding will eventually destroy the information capital embodied in a trademark, and the prospect of free riding may therefore eliminate the incentive to develop a valuable trademark in the first place.). Empirical work on trademark transactions points in the same direction in a different way: Kost and Pearce report that buying firms see sales rise faster than costs, which they read as a markup effect tied to acquired goodwill (sales increase more than costs.).

The economic critique holds that the same protection can create market power that goes beyond what is needed to prevent confusion. Van Caenegem notes that given the significant market power afforded by brands in the contemporary marketplace, the exercise of rights by incumbent trademark owners may have a significant effect on competition. Economides suggests that firms may gain some small market power from first use of the most appropriate symbol and words as trademarks. Such monopoly power is limited by restrictions against use of generic and descriptive terms. The effect on barriers to entry is contested. Economides finds that the effects of trademarks on barriers to entry are ambiguous. The intertemporal effects of perception advertising may create barriers to entry for newcomers. Such barriers will be beneficial to society when they tend to decrease the number of brands toward optimality. Critics go further: Galbraith argued that branding is socially wasteful because it inspires a zero-sum spirit and increases barriers to entry. The concern about scope is also visible inside the theory's own authors, who, although they explained most doctrines through search costs, registered some concern about dilution-based infringement claims. U.S. law reflects the antitrust side of the debate, since the Lanham Act lists violation of antitrust law as a defense to infringement (the mark has been or is being used to violate the antitrust laws of the United States).

The debate also concerns method. Supporters of the law-and-economics approach argue that the consumer is not nearly so gullible, and Landes and Posner dismissed the view that brand advertising could mainly bamboozle the public and thereby promote monopoly ("the power of brand advertising to bamboozle the public and thereby promote monopoly."). Critics respond that the theory fails to account for the idea that a mark's persuasive power may be more impactful than merely its informational content. Others argue that the framework cannot fully explain trademark doctrine such as distinctiveness and dilution, and that semiotic analysis adds insight. Formal legal approaches, by contrast, anchor the field in statutory purposes: the Lanham Act is usually read as aiming to protect producers from unfair competition and consumers from deception (first, to protect producers from unfair competition; and second, to protect consumers from deception.). The disagreement is therefore not only about whether trademarks are efficient, but about whether efficiency should set the outer limits of protection or whether the legal purposes of source identification and fair dealing should do so.

Speech, Culture, and Non-Commercial Use

Trademark law gives an owner control over the use of a sign in commerce, while freedom of expression protects commentary, satire, scholarship, and art that often names brands. The conflict is sharpest when a brand name appears in a film title, a song, a novel, or a parody product. Legal systems handle it in different ways. In the United States, the main tool has been a threshold test that filters out certain expressive uses before a confusion analysis begins. The Rogers test aims to balance freedom of expression (protected under the First Amendment) with freedom from consumer confusion (protected by the Lanham Act).

The test comes from the Second Circuit's 1989 decision in Rogers v. Grimaldi. The case concerned the film Ginger and Fred, whose fictional cabaret performers imitated Fred Astaire and the actress Ginger Rogers. The court held that the Lanham Act does not reach a minimally relevant use of a celebrity's name in an artistic title unless the title explicitly claims endorsement or misleads about content. Under the framework that most federal circuits later adopted, use of a mark in an expressive work is only infringing if it has no artistic relevance to the work whatsoever, or has some artistic relevance but explicitly misleads as to the source or the content of the work. Courts later applied the test beyond titles to the content of works. Though the original case involved a dispute over a movie title, courts have applied Rogers to other types of "expressive" works and their content. Because the test is a threshold filter, it has often ended cases early; commentators have noted a recent and growing willingness of courts to dispose of trademark claims against expressive works at the pleading stage.

The limits of the test were redrawn in Jack Daniel's Properties, Inc. v. VIP Products LLC (2023). The defendant sold a squeaky dog toy, the "Bad Spaniels," shaped like a whiskey bottle with parodic labeling. The Ninth Circuit had treated the toy as expressive because it conveyed a humorous message. The Supreme Court unanimously held that when an alleged infringer uses a trademark as a designation of source for the infringer's own goods, the Rogers test does not apply. The Court also held that the statutory non-commercial exclusion from dilution liability does not protect a parody when the parodist uses the mark to identify its own products. Justice Gorsuch's concurrence, joined by two others, questioned the legal basis of the Rogers framework, and the majority left that question open. Some commentators have argued that Jack Daniel's could mean the end of the Rogers test, though the case did not actually overturn Rogers. The practical line that emerged distinguishes a mark used to communicate a message or title from a mark used to tell consumers who makes the product.

Trademarks appear throughout cultural production, including films, music, games, and literature. Expressive works include things like movies, plays, books, and songs. In one video-game dispute, a court applied the Rogers test to the use of a mark within an expressive work and rejected the argument that a mark must be a "cultural icon" before the test applies. The court held that was not the case, and there was no "threshold test" for a trademark to fall under the Rogers test. Courts elsewhere have weighed similar interests, often through parody and fair-use analysis, and outcomes depend heavily on whether the use identifies source, comments on the brand, or simply borrows its recognition for a creative work.

Brand names and symbols also carry cultural and public weight, and some of that weight is protected through separate international rules. Article 6ter of the Paris Convention has since 1883 allowed countries to block the registration or use of their armorial bearings, flags and other State emblems as well as official signs and hallmarks indicating control and warranty as trademarks. In 1958 the protection was extended to armorial bearings, flags, other emblems, abbreviations and names of international intergovernmental organizations. This protection is not a trademark right in itself: Article 6ter does not generate a trademark right, or any other type of IP right, over the signs that are covered by that provision. Its scope is also limited, since the protection is not available for private organizations such as non-governmental organizations (NGOs). National systems add their own rules; for example, Swiss law protects coats of arms and emblems of the Confederation, cantons, and local authorities.

Public-interest exceptions appear in several forms: statutory exclusions for non-commercial and news uses in dilution law, descriptive and nominative fair-use defenses, and the expressive-work filters discussed above. Their boundaries remain contested, particularly where a cultural work and a commercial product share the same name.

Relationship to Other Intellectual Property

Comparison with patents and copyright

Trademarks sit alongside patents and copyright within intellectual property law, but each regime protects a different kind of subject matter for a different purpose. A patent gives an exclusive right over a new, useful and non-obvious invention for a limited term, generally twenty years from filing in most systems, after which the invention passes into the public domain. Copyright protects original works of authorship such as texts, music, software and visual art, and in most countries it lasts for the author's life plus a fixed number of years. A trademark protects a sign that tells consumers where goods or services come from. Because it does not depend on novelty or creativity, a trademark can be protected even when the sign is familiar or simple. Its protection is also not tied to a fixed expiry: registrations can be renewed repeatedly, typically in ten-year periods, as long as the sign stays in use and continues to identify a source. The three regimes therefore differ in what they cover, what qualifies for protection, and how long protection lasts.

Geographical indications

A geographical indication names a product whose quality, reputation or other characteristics are essentially attributable to a particular place. Unlike a trademark, which has an individual owner, a geographical indication is shared by all qualifying producers in the region. The TRIPS Agreement, administered within the WTO, sets the baseline rules in Articles 22 to 24. Article 22 requires members to refuse or invalidate a trademark that contains a geographical indication for goods from elsewhere when its use would mislead the public about origin. Article 23 provides stronger protection for wines and spirits: a geographical indication for these products can be protected against use for goods from other places even where no one is deceived, including where a translation or a qualifier such as "style" or "kind" is added, and trademarks containing such a false indication must be refused or invalidated. Article 23.3 permits homonymous indications to coexist where they can be practically distinguished. Article 24 contains exceptions that preserve certain earlier trademarks and generic terms. Coexistence is therefore governed by conflict rules that weigh the type of goods, the timing of registration and good faith, rather than by a single hierarchy in which one right always prevails.

Overlapping protection and the functionality limit

A single product can be covered by several regimes at once. A pharmaceutical, for example, may be protected by a patent on its compound, a copyright in its label text, a design right in its bottle, and a trademark for its brand name. These rights are cumulative because each protects a different element. Tension arises when a producer tries to use trademark law to protect functional features that a patent once covered. The functionality doctrine addresses this by denying protection to features that are essential to a product's use or purpose, or that affect its cost or quality. In the United States, the Supreme Court has described the doctrine as a means of keeping trademark law from giving a producer control over useful product features that patent law grants only for a limited time. In <i>TrafFix Devices</i> (2001), the Court held that the existence of an expired utility patent is strong evidence that a design is functional, although it does not bar trade dress protection as a matter of law. Secondary meaning cannot normally overcome a finding of functionality. In the European Union, the trade mark regulation excludes shapes that result from the nature of the goods, are necessary to obtain a technical result, or give substantial value to the goods. These limits are intended to prevent trademark rights from becoming perpetual monopolies over functional features.

Trade secrets and design rights

Trade secrets protect confidential commercial information, such as formulas, manufacturing processes and customer data, when that information has value because it is secret and its holder takes reasonable steps to keep it so. Article 39 of TRIPS requires members to protect undisclosed information against unfair commercial use. A trade secret requires no registration and involves no public disclosure, so its protection can last as long as secrecy is maintained. However, it gives no exclusive right against independent discovery or reverse engineering. Brand names and trade secrets often coexist: the name is public and registered, while the underlying recipe or process remains confidential.

Design rights protect the appearance of a product, including its lines, contours, colours, shape and texture. Many systems offer registered designs, unregistered design rights or, in the United States, design patents. Their terms are usually shorter than trademark terms. A product shape can attract both a registered design and trademark protection, but the trademark route requires that the shape be distinctive and non-functional, and the design route requires that the design be new and individual.

Open Questions and Controversies

The expansion of protectable subject matter has produced some of the hardest questions in trademark law. Sounds, smells, colours, shapes and movements can all function as source identifiers, but the classic tests were built around words and figurative logos that could be printed and examined on a register. In the European Union, the Court of Justice ruled in Sieckmann (2002) that a smell described in words and a chemical formula did not meet the graphic-representation requirement then in force. Later EU reforms replaced that formula with a looser standard: a sign must be represented in a way that allows authorities and the public to determine its subject matter clearly and precisely. The underlying difficulty remains. It is hard to show that a scent or a short sequence of notes is distinctive, to fix its scope, and to avoid handing one firm a monopoly over a functional or aesthetic feature. Colour marks raise a related problem, because a shade that makes a product appealing may be wanted by many competitors, which brings the doctrine of monopoly-like control into the analysis.

The debate over dilution and anti-free-riding protection is equally unsettled. Dilution doctrine protects famous marks against blurring, which weakens a sign's uniqueness, and against tarnishment, which damages its reputation, even where consumers are not confused about source. Supporters argue that the investment behind a famous brand deserves protection against parasitic use that erodes the value of the sign. Critics respond that such protection can extend into unrelated markets and that harm is difficult to measure. The United States added federal dilution protection in 1995 and tightened it in 2006 so that actual dilution must be shown rather than a likelihood of it. The European Court of Justice went further in L'Oréal v Bellure (2009), accepting that imitating a competitor's packaging to gain an unfair commercial advantage can infringe even without confusion. That ruling brought an anti-free-riding logic into EU law and reopened the question of how far such protection should reach beyond the confusion standard, a question that overlaps with competition law and its concern with market power.

Multi-jurisdictional enforcement exposes the national character of trademark rights. Because rights are territorial, the same sign can belong to unrelated owners in different countries. The name BUDWEISER has been contested for decades between the Czech brewer Budějovický Budvar and the American Anheuser-Busch, and the dispute has been managed largely through coexistence arrangements that divide markets rather than through a single binding outcome. In first-to-file systems, a party may register a foreign brand before its owner does and then seek payment for its transfer, a practice that tests the boundary between legitimate registration and bad faith. Harmonized procedures do not resolve these conflicts, because the question of who owns a sign in a given territory is still answered by local law.

Genericide raises its own open issues. The legal threshold at which a once-distinctive mark becomes the common name for a product is contested, and courts must decide how to weigh dictionary usage, consumer perception, and the owner's own promotional efforts. The example of aspirin shows how a brand name can pass into general use, while verb forms of brand names used by the public complicate the question of whether an owner's enforcement can prevent a mark from losing its source-identifying function.

Online intermediaries present a parallel uncertainty. Marketplaces and search engines host listings, sell advertising keywords, and process large volumes of notices, so the question of when they must act on counterfeit goods or trademark-triggering ads remains contested. In the United States, the Second Circuit held in Tiffany v eBay (2010) that general knowledge that counterfeits were sold on a platform was not enough to impose liability. The Court of Justice of the European Union, in L'Oréal v eBay (2011), held that an operator playing an active role in promoting listings may lose the protection given to hosting providers. The two approaches place the line between neutral infrastructure and active participation differently, and the outcome shapes how e-commerce platforms design their takedown systems.

Finally, the balance between brand owners and competitors is still being drawn case by case. Comparative advertising, nominative use of a rival's name, descriptive use of a term, and the resale of genuine goods all involve the use of another's sign, and courts differ on how much room these practices should have. The unresolved question is whether trademark law should be read chiefly through its signalling function or whether its protection of investment in brand value justifies broader control over the way a sign may be used in the marketplace.

References

1. Lanham Acten.wikipedia.org
2. Lanham Actlaw.cornell.edu
3. Overview of Trademark Lawcyber.harvard.edu
4. 15.2 Definition—Trademark (15 U.S.C. § 1127)www3.ce9.uscourts.gov
5. What Is an Assignment of Trademark? - Revision Legalrevisionlegal.com
6. PERFECTING SECURITY INTERESTS IN IPdorsey.com
7. Casebook Chapter Fiveweb.law.duke.edu
8. Trademark Licensing: How to License Without Losing the Markdaeryunlaw.com
9. Intro to TMs — That's a Trademarkthatsatrademark.com
10. Trademarken.wikipedia.org
11. Categories of Trademark: Certification Marks, Collective Marks, Well known Marks and Non-Conventional Markslawbhoomi.com
12. Types of Trade Marks - IPOIipoi.gov.ie
13. Non-conventional trademarken.wikipedia.org
14. The scent of a trademark: removal of graphic representability requirementdennemeyer.com
15. Non-traditional markswipo.int
16. Graphical representation for registration of non-traditional trademark under Malaysian and UK trademark lawinderscienceonline.com
17. Non Traditional Trademarks in Indian IP regimekhuranaandkhurana.com
18. Non-Conventional Trademarks: Sound marksbananaip.com
19. Check out this article...Types and Examples in the List of Trademarksupcounsel.com
20. INTENT-TO-USE LEGISLATION UNDER THE TRADEMARK LAW REVISION ACT OF 1988:scholarship.shu.edu
21. University of Miami International and Comparative Law Reviewrepository.law.miami.edu
22. All Info - S.1883 - 100th Congress (1987-1988): Trademark Law Revision Act of 1988congress.gov
23. Bona Fide Intention to Use a Trademark - Morse Lawmorse.law
24. TRADEMARK LAW REVISION AcT OF 1988 AND THEjournals.library.wustl.edu
25. Washington University Law Review Washington University Law Review Volume 68openscholarship.wustl.edu
26. cvi Trade Marks Act, 1938. [1 & 2 GEO. 6. CH. 22.] ARRANGEMENT OF SECTIONS.legislation.gov.uk
27. Improper Trademark Symbol Use: Hidden Risksgleamlaw.com
28. Registration Symbol Misuse As Trademark Fraud?duetsblog.com
29. Circle Up - When And How Trademark Symbols Should Be Used: Amundsen Davisamundsendavislaw.com
30. Difference Between TM and R - Trademark Symbols in US Law ®sierraiplaw.com
31. Trademark Examination Guides, Notes and Announcements - Examination Guide NO.3-89ipmall.info
32. On Your Mark—"TM" or "®"? Trademark Notice Guidelines - Ward and Smith, P.A.wardandsmith.com
33. A Brief History of the Trademarkcfc.com
34. Historical Development of Trademark Lawlegal60.com
35. Mark (sign)en.wikipedia.org
36. York Assay Officeen.wikipedia.org
37. Hallmarks: The Oldest Form of Consumer Protectiondoerrvaluations.co.uk
38. Hallmarkingassayoffice.co.uk
39. Goldsmiths' Company Assay Office in Londonassayofficelondon.co.uk
40. History of Hallmarking - Silver Editions UKsilver-editions.co.uk
41. Goldsmiths' Hall: The Birthplace of the Hallmarkskyjems.ca
42. Assay officeen.wikipedia.org
43. Passing Off and Unfair Competition: Conflict and ...scholars.law.unlv.edu
44. 1 Passing Off: An Uncertain Remedy Peter Charleton and Sinéad Reilly1fordhamipinstitute.com
45. The Role of Unfair Competition in the Common Lawscholarship.law.columbia.edu
46. Casebook Chapter Fourweb.law.duke.edu
47. Reverse Confusion and the Justification of Trademark Protection - The George Mason Law Reviewlawreview.gmu.edu
48. The famous red triangle trademark - Brewery Heritagebreweryheritage.com
49. The Bass Red Triangle: Britain’s First Trade Markdiscovereaststaffordshire.com
50. TRADEMARK LAW An Open-Source Casebook Version 8 (2021) Digital Editiontmcasebook.org
51. Federal Trademark Law: From Its Beginningsamericanbar.org
52. Unconstitutional trademark actsen.wikipedia.org
53. Trade-Mark Casesen.wikipedia.org
54. Bass, the UK's oldest trademarkthetrademarkhelpline.com
55. www.legislation.gov.uklegislation.gov.uk
56. Trade Marks Act 1938en.wikipedia.org
57. Trademarkneoencyclopedia.fandom.com
58. Wilson Gunnwilsongunn.com
59. Chapter 12: Trademark Law Primer 12.1 ...law.berkeley.edu
60. Lanham Actencyclopedia.com
61. Check out this article...Lanham Act: Trademarks, False Advertising, and Remediesupcounsel.com
62. Trademark Law Revision Act of 1988, Title I & Satellite Home Viewer Act of 1988, Title II (Public Law 100-667, 102 Stat. 3935), United States of America, WIPO Lexwipo.int
63. Summary of S. 1883 (100th): Trademark Law Revision Act of 1988 - GovTrack.usgovtrack.us
64. Lanham Act as amended by the Trademark Law Revision Act of 1988ipmall.info
65. 15 USC Ch. 22: TRADEMARKSuscode.house.gov
66. Trade Marks Act 1938 (Amendment) Bill [H.L.] (Hansard, 22 December 1982)api.parliament.uk
67. Trade Marks, Patents & Copyright in Paris Conventionadams.africa
68. International Trademark Rights - International Trademark Associationinta.org
69. Appendix P - Paris Conventionuspto.gov
70. What is TRIPS Agreement?signa.so
71. WTOwto.org
72. WTOwto.org
73. Trade Policyuspto.gov
74. WTOwto.org
75. Trademark - International Systems and Territoriality Study Deckremnote.com
76. European Union trade marken.wikipedia.org
77. European Union Intellectual Property Officeen.wikipedia.org
78. "The Trademark Provisions in the TRIPS Agreement" by Christine Farleydigitalcommons.wcl.american.edu
79. Trade-related aspects of IP rightsuspto.gov
80. Trademark Protection in the EU. - Law 4 Small Business (L4SB)l4sb.com
81. EU trade mark registration at EUIPOrodall.ro
82. EU Trademark Registration — EUIPO Complete Guide — tmarkmetrictmarkmetric.com
83. Madrid Agreement Concerning the International Registration of Markswipo.int
84. The Madrid Systemladas.com
85. Madrid Protocolitlaw.fandom.com
86. 89–010 Calendar No. 101 107TH CONGRESS REPORT " ! SENATE 1st Session 107–46congress.gov
87. The Birth of the Madrid Protocolwipo.int
88. International trademark via WIPO, Madrid Systemrodall.ro
89. Madrid Protocol Singapore: Register Your Trademark Overseasexcellencesg.com
90. WIPO Madrid System vs. Direct Filing: How to Protect Your Brand Globally — Skalaskala.io
91. 9 Disadvantages of the Madrid System for International Trademark Registration - Lexologylexology.com
92. Protocol Relating to the Madrid Agreement Concerning the International Registration of Markswipo.int
93. The History of the Madrid Protocol · Smit & Van Wyksvw.co.za
94. International Trademark Registration: Madrid Protocol & WIPO Guide (2026) - IGBSig-bs.com
95. Madrid Protocol Countries and How It Works — tmarkmetrictmarkmetric.com
96. How to Trademark Internationally — tmarkmetrictmarkmetric.com
97. WTOwto.org
98. TRIPS Flexibilities relating to Enforcementipaccessmeds.southcentre.int
99. TRIPS Trade-Related Aspects of Intellectual Property Rightswipo.int
100. Uniform Domain Name Dispute Resolution Policy (UDRP)uk.practicallaw.thomsonreuters.com
101. Cybersquatting, Typosquatting, and Domaining: Ten Years Under the Anti-Cybersquatting Consumer Protection Actjw.com
102. Anticybersquatting Consumer Protection Act Guidethefirstfile.com
103. The Anticybersquatting Consumer Protection Act: Key Information - Internet Library of Law and Court Decisionsinternetlibrary.com
104. Cybersquatting Involving Trademarks and Service Marks & Related Lawsuitsjustia.com
105. Uniform Domain-Name Dispute-Resolution Policyen.wikipedia.org
106. What is the Anticybersquatting Consumer Protection Act?thefashionlaw.com
107. Use it or lose it: A good reason for non-use does not save an abandoned trademarkfieldfisher.com
108. Death, Taxes, And Trademark Abandonment: Trademark Transition Strategieslegal500.com
109. Abandonment and Nonuse of a Trademarkwilsonlegalgroup.com
110. Abandonment and Nonuse of a Trademarkny-trademark-lawyer.com
111. How Long Does a Trademark Last?trademarklawyerfirm.com
112. What Is Trademark Abandonment Due to Non-Use?tmexpress.com
113. Trademark Abandonment & Loss of Trademark Rightsnextrendlegal.com
114. Trademark Abandonment - Carstens Allen Gourleycaglaw.com
115. Trademark Abandonment: Causes, Consequences, and Preventiontrademarklawauthority.com
116. Avoiding “Genericide”: Lessons from a Recent Case - Lexologylexology.com
117. Too Much of a Good Thing: How Genericide Sends Trademarks to the Graveyarduclawreview.org
118. Generic trademarken.wikipedia.org
119. The popular brands that had their trademarks revoked by lawbentley.edu
120. Genericization: 10 Famous Trademark Examplescarsonpatents.com
121. Trademark Genericide: 7 Smart Ways to Avoid Itperspireip.com
122. Genericide: Famous Brands That Lost Their Trademarks — Braslow Legalbraslowlegal.com
123. Case Studies on Genericide: How Some Famous Trademarks Became Everyday Words - Trademark Angeltrademarkangel.com
124. Google v Louis Vuittonen.wikipedia.org
125. Can Your Competitor Use Your Brand To Capture Internet Traffic? The European Rule on Keyword Advertising Takes Shapedorsey.com
126. epi Informationinformation.patentepi.org
127. This article is more than 1 year oldtheregister.com
128. L’Oréal v. eBay: the Court of Justice of the European Union tightens liability of online marketplace operators the CJEU - Lexologylexology.com
129. But was it worth it? The significance of the L'Oréal v eBay trade mark ruling for brand owners and online marketplacesmarketinglaw.osborneclarke.com
130. Tiffany (NJ) Inc. v. eBay Inc.loeb.com
131. EU: increasing liability of online marketplaces for trademark infringementlegalblogs.wolterskluwer.com
132. ECJ Issues L'Oreal v. eBay Ruling: Online Marketplaces Can Be Liable for Counterfeit Goodsfinnegan.com
133. Tiffany (Nj) Inc. v. Ebay, Inc. (Tiffany (Nj) Inc. v. Ebay, Inc., 576 F.Supp.2d 463 (S.D. N.Y. 2008)) - vLex United States - vLexcase-law.vlex.com
134. The Second Circuit's Decision in Tiffany v. eBayfinnegan.com
135. William M. Landes and Richard Posnercyber.harvard.edu
136. Full article: In search of value: trade marks and search costs in the age of the Internettandfonline.com
137. Rethinking Trademarks and Competition: When is a Brand a Barrier to Market Entry? - Bond University Research Portalresearch.bond.edu.au
138. U1 - Trademark Law Economic Perspectivescribd.com
139. Vol. 78 TMR 523 THE ECONOMICS OF TRADEMARKS By Nicholas S. Economides*neconomides.com
140. UNITED STATES OF AMERICA BEFORE THE FEDERAL TRADE COMMISSIONdigitalcommons.law.scu.edu
141. The economics of brands and branding - ScienceDirectsciencedirect.com
142. The Last Breakfast with Aunt Jemima and Its Impact on ...journals.library.columbia.edu
143. Market Power through the Lens of Trademarks Kyle Kost Jeremy Pearceliangjiewu.com
144. (PDF) The Economics of Trademarksacademia.edu
145. The Economic Structure of Intellectual Property Lawinter-droitetaffaires.com
146. The role of trademark rights expansion in the formation and abuse of market power - ScienceDirectsciencedirect.com
147. Search and Persuasion in Trademark Lawrepository.law.umich.edu
148. mckinneylaw.iu.edumckinneylaw.iu.edu
149. (PDF) A Search-Costs Theory of Limiting Doctrines in Trademark Lawresearchgate.net
150. Unanimous Supreme Court Finds Rogers Test Does Not Apply When Trademark Is Used As Source Identifierhaynesboone.com
151. Trademark use Within an Expressive Work Must Only Pass the Rogersnatlawreview.com
152. The Trademark and First Amendment Dance Continues: Application of the Rogers v. Grimaldi Test After Jack Daniel’sfinnegan.com
153. COMMITTEE REPORT USING TRADEMARKS IN EXPRESSIVE WORKS:jenner.com
154. JACK DANIEL'S PROPERTIES, INC. v. VIP PRODUCTS LLC (2023)caselaw.findlaw.com
155. Rogers v. Grimaldien.wikipedia.org
156. ONLINE DATABASE ON PROTECTED EMBLEMS AVAILABLE FROM WIPOtheicod.org
157. Article 6terwipo.int
158. Jack Daniel's Properties, Inc. v. VIP Products LLCen.wikipedia.org
159. Supreme Court Sharply Limits Applicability of Rogers v. Grimaldi Test for Trademark Infringementskadden.com
160. Jack Daniel’s v. VIP Products: Supreme Court Clarifies When Rogers Test Applies in Trademark Casesdwt.com
161. Jack Daniel’s Properties, Inc. v. VIP Products LLC: The Supreme Court Corrects Course on Rogers Testicemiller.com
162. U.S. Supreme Court Vacates Ninth Circuit Opinion in Jack Daniel’s Properties, Inc. v. VIP Products - International Trademark Associationinta.org
163. Flags and emblems - Swiss Federal Institute of Intellectual Propertyige.ch
164. Functionality doctrineen.wikipedia.org
165. TMEP 1202.02(a)(ii): Purpose of Functionality Doctrine, Oct. 2017 Ed. (BitLaw)bitlaw.com
166. Trade Dress Functionality: The Legal Standardrevisionlegal.com
167. Functionality and Trademark Law: Why Some Product Features Can Never Be Protectedgleamlaw.com
168. 72 MODULE IV GEOGRAPHICAL INDICATIONS Introduction Generalwto.org
169. WTOwto.org
170. r:\publish\sct-forum\eng\sct5_3e.doc WIPO E SCT/5/3 ORIGINAL: Englishwipo.int
171. Geographical Indications and Trademarks: Synergies and Conflicts in theorigin-gi.com
172. THE CASE FOR AESTHETIC FUNCTIONALITYbtlj.org