26.10 Section 44 - non-Roman characters

7.1 Terminology 

The English alphabet is based on Roman (also known as Latin) characters. 

Many scripts use non-Roman characters, such as Arabic, Chinese (Han) characters (or Hanzi), Devanagari, Cyrillic, Greek, Sanskrit, etc. 

For any non-Roman characters in a trade mark, the applicant must provide: 

• A transliteration into Roman letters, i.e. expressing the pronunciation. Languages that principally use a non-Roman script may have recognised systems of transliteration into Roman characters (known as romanisation). For example, pinyin is the official phonetic alphabet and romanisation for Mandarin Chinese. 

• A translation into English. If the non-Roman element has no meaning (such as for an invented word that is used only as a brand name), this should be stated. 

For example: 

• 苹果 is a Chinese word written in Chinese (Han) characters. 

• The transliteration into Roman characters using pinyin (without tone markers) is PING GUO. 

• The translation into English is apple. 

Refer to Part 10.4 for information about the preferred format for a translation and transliteration endorsement. 

7.2 Eligibility to examine non-Roman character trade marks 

Trade marks that contain: 

• non-Roman characters (e.g. 苹果) or 

• a transliteration of non-Roman characters (e.g. PING GUO) 

can only be examined by a specialist non-Roman character examiner (“NRC examiner”). 

If you are not an NRC examiner, and you retrieve an application for a trade mark that is or might contain the above elements, please email ML-TM Non Roman Character Examiners to check whether the application should be handed on to an NRC examiner. 

7.3 General considerations

7.3.1 Searching 

Please see Guidelines for RIO Search for Trade Marks for detailed information about searching non-Roman characters in RIO. 

Examiners should search transliterations of non-Roman characters in the RIO word search fields to find both: 

• similar non-Roman characters; and 

• trade marks that may not contain any non-Roman characters, but may convey the same sound or meaning to the target audience (e.g. an English rendering of an Arabic word). See below for information about comparing Chinese characters and pinyin. Similar considerations apply to comparisons between transliterations and non-Roman characters in other scripts. 

Transliterations are found in two fields: 

• “Endorsements”: This contains the transliteration provided by the applicant. Use “Other information” in RIO to search this field. 

• “Transliteration indexing constituents”: For Chinese (Han) characters in any language (e.g. Chinese, Japanese kanji, Korean hanja), the transliteration in pinyin is separately indexed in this field. This field is included in the normal RIO “Word” search. “Exact word (transliteration only)” searches this field only. 

In the Transliteration indexing constituents of recent applications (filed after Feb 2023) containing Chinese (Han) characters, the simplified Chinese version of each character is also shown alongside the pinyin. Regardless of what version of the character is used in the trade mark, the indexing constituents will display the simplified Chinese version, if available. 

For example, the trade mark 

馬

would be indexed as: 

Transliteration indexing constituents

马MA

Online translation tools can be used to transcribe traditional Chinese into simplified (and vice versa). Chinese/English dictionaries also often display both variations. See for example: 

• http://www.yellowbridge.com/chinese/chinese-dictionary.php 

• https://www.mdbg.net/chinese/dictionary 

7.3.2 Target audience 

When a trade mark includes non-Roman characters, examiners should consider that the ordinary consumer of the claimed goods/services will include: 

• consumers who can read and understand the characters; and 

• consumers who cannot read and understand the characters. 

In Southcorp Brands Pty Ltd v Australia Rush Rich Winery Pty Ltd [2019] FCA 720, [37] to [38], Beach J noted (emphasis added): 

“In terms of establishing the ordinary signification of a trade mark consisting of a foreign word, what is important is the meaning conveyed by the foreign word to those who will be concerned with the relevant goods (Cantarella Bros Pty Ltd v Modena Trading Pty Ltd [2014] HCA 48; (2014) 254 CLR 337 at [48] per French CJ, Hayne, Crennan and Kiefel JJ).

Accordingly, when assessing questions of trade mark infringement or misleading or deceptive conduct concerning Chinese language including Chinese characters, emphasis is to be placed on the meaning and pronunciation of Chinese characters used in the relevant trade marks.  More particularly, it is necessary to consider the appearance and sound as well as the meaning of Chinese characters when assessing allegations of trade mark infringement.  Further, misleading or deceptive conduct can be established by the use of Chinese characters even though the class of potential customers who might be misled is essentially confined to those who speak and/or read only Chinese, and even though no misrepresentation has been communicated to anyone who could not read those characters (Osgaig Pty Ltd v Ajisen (Melbourne) Pty Ltd (2004) 213 ALR 153 at [112] per Weinberg J).”

Similarly, in Southcorp Brands Pty Limited v Gotop Australia Pty Ltd [2024] ATMO 205, [17] to [18], the hearing officer considered the perspectives of both Chinese-Mandarin speaking wine buyers in Australia, as well as non-Chinese speaking consumers, when considering deceptive similarity:

“I am in agreement with the Opponent that the relevant class of consumers is the general population of alcohol drinkers, which consists of most of the adult Australian population and that a relevant subset of that class is the Chinese-speaking adult Australian population. I agree, too, with the broad proposition put forward by the Opponent that consumers look for key indicators on the labels when selecting alcoholic beverages. Likewise, I am mindful of the possible notional fair use of the Trade Mark by the Applicant, which might include use on an identically stylized bottle and label to that sold by the Opponent.

Since the Trade Mark contains a Chinese character it is reasonable to assume that an intended class of consumers would be Chinese-Mandarin speakers. The Opponent insinuates that those consumers would regard the Chinese character element of the Trade Mark as descriptive. The writer’s own enquiries suggest that the Chinese character 坊 transliterates to “square”. There is no evidence on the matter to assist me, but it appears that the character has no descriptive sense or obvious connotation when used in relation to the Applicant’s Goods – certainly none to indicate that the character 坊, alone, would suggest “wine shop”.

That said, my own knowledge of Chinese characters is that they more often or not have more than one meaning and that the relevant meaning is contextually determined. I find that Chinese speakers seeing that character in use in relation to the Applicant’s Goods may infer a meaning of “wine shop” or similar. But whether those consumers drew that inference, or treated the character as contextually meaningless, the numbers 389 would exist as the memorable part of the Trade Mark. Further, it can reasonably be inferred that those same consumers would have no understanding of the word “bin” and disregard it. In short, they would perceive 389 as the dominant and memorable part of both marks.

To the wider class of consumers - the adult Australian population who do not read Chinese -the character 坊 would have no meaning and it would more likely than not be treated as ornamental, or perhaps a translation of the number 389 into a different script. Further, most of those consumers would have seen the word “bin” used by the Opponent and others on labels. Whether they understood its meaning in connection with wine (described already) or had no precise understanding of its meaning, they are likely to treat it as descriptive/generic and principally recall the number 389.”

See also Guangzhou Baiyunshan Pharmaceutical Holdings P/L v Multi Access Ltd [2022] ATMO 204, [23] to [25]. 

7.4 Guidelines for comparison – Chinese (Han) characters 

The usual tests for comparing word trade marks are applied when deciding whether trade marks containing non-Roman characters or transliterations of non-Roman characters are deceptively similar to other trade marks, and whether there is a ground for rejection under section 44 of the Act. 

Outlined below are some common considerations relating to trade marks containing Chinese (Han) characters. Chinese (Han) characters are used for Chinese languages/dialects, as well as Japanese (kanji), Korean (hanja), and historically, Vietnamese. 

7.4.1 Character vs Character 

Traditional vs Simplified 

There are two standard contemporary Chinese (Han) character scripts: traditional and simplified. 

For example: 

Traditional: 馬

Simplified: 马

Those who can read Chinese would understand the above to be the exact same word, but written in different writing systems. This is similar to a word having different regional spellings, e.g. color and colour, organise and organize. 

Since they sound the same and have the same meaning, a ground for rejection under section 44 would be appropriate when comparing simplified and traditional versions of the same character/s. 

Other variations 

In addition to standard simplified and traditional Chinese, there are also other variations based on period, language and region. The usual rules for comparison apply, as informed by research. 

For example, 徴 vs 徵. Research shows that these are Japanese and Chinese (traditional) versions of the same character. They look almost identical except for a single stroke difference in the middle. Consumers are likely to gloss over the minor difference or assume that they are variations of the same character (even if they do not know for certain), such that they convey the same sound and meaning. These characters are deceptively similar. 

Identical characters in a different typeface / style 

Chinese (Han) characters can be rendered in different typefaces or styles. Some styles of calligraphy in particular can look very different to typed print. 

For example, the below are different renderings of 馬: 

馬

This is similar to how English words look different when rendered in different styles, such as copperplate calligraphy vs Times New Roman type. 

In this situation, although there are some visual differences, consumers who can read these characters would identify them as the same word. They therefore sound the same and have the same meaning. 

A ground for rejection under section 44 would be appropriate when comparing the same character/s in different typefaces or styles. 

Reading order 

Many languages using Chinese (Han) characters were traditionally written downwards in vertical columns, and from right to left. Nowadays, depending on the language, region and context, it is common to find Chinese (Han) text in different reading directions, typically: 

• horizontally, left to right

• in vertical columns, proceeding right to left 

• horizontally, right to left (this is typically reserved for traditional style headings or signage). 

The inclusion of dictionary words/phrases can help to determine the likely reading direction for someone who speaks the language. However, there may also be no obvious reading direction, such as for invented words or names. 

Traders may use different versions of their brands with different reading directions. For example, in respect of Class 43: restaurant services, a trader may use: 

• horizontally left to right for plain text online: 

老李麵館 

(transliteration: LAO LI MIAN GUAN)

• horizontally right to left for a composite mark, evoking a traditional style: 

(transliteration: LAO LI MIAN GUAN)

As a result, trade marks that contain the same characters but in a different reading direction may be deceptively similar. 

See for example: Guangzhou Baiyunshan Pharmaceutical Holdings P/L v Multi Access Ltd [2022] ATMO 204, where the delegate found that two trade marks composed of the same three characters arranged horizontally left to right (王老吉), and right to left (吉老王), were deceptively similar. 

Note: The transliteration and translation endorsement provided by the applicant may provide an indication of the intended reading direction, but is not determinative. Consumers will not have the benefit of the endorsement when encountering a trade mark in the marketplace. Applicants may also give the transliteration/translation according to the English reading direction, even if that is not how it would be read natively. 

When in doubt as to the likely reading direction or impact on a comparison, examiners should consult a subject matter expert. 

7.4.2 Character vs Pinyin (Transliteration) 

Pinyin is a standardised system of transliteration/romanisation used in Chinese, and represents the pronunciation of Chinese (Han) characters in Mandarin Chinese. Pinyin is a part of the language. It is included in modern Chinese dictionaries as a pronunciation guide and for alphabetical searching, and is commonly used to type in Chinese. 

For example: The pinyin for 馬 or 马 is MǍ. 

The diacritic (or “tone marker”) indicates the tone. Tone is sometimes also represented by number, e.g. MA4. Tone markers are often omitted e.g. in signage or branding. 

Most characters have one standard pinyin transliteration in Mandarin Chinese, but some may be pronounced differently depending on context, e.g. when the character is used for different meanings or to form different words. For example, 会 meaning “to be able to” is transliterated as HUI, but in 会计  meaning “accountant”, the same character is transliterated as KUAI. 

A character will be indexed with all standard pinyin transliterations, though some pronunciations may not apply to the particular trade mark. For example: 

Single Chinese character vs Pinyin transliteration 

For example: 正 vs ZHENG. 

ZHENG is the transliteration of 正. 

Each character is pronounced in Mandarin Chinese as one syllable and many characters sound the same. 

As a result, one syllable of pinyin can refer to a multitude of different characters (more so when tone markers are not used), which may differ widely in look and meaning. For example, ZHENG could refer to: 

正

证

郑

整

政

争

挣

蒸

and many more. 

For consumers who can read Chinese, there is not a real and tangible risk that they would assume, based on a single syllable of pinyin alone, that it refers to any specific Chinese (Han) character. 

Therefore, when comparing a single Chinese (Han) character to its pinyin equivalent, a ground for rejection under section 44 typically does not arise unless there are also other similarities. This is because even though the character and pinyin may sound similar, the meaning of the pinyin is highly ambiguous, and they differ visually. 

For transliterations of Chinese (Han) characters other than pinyin, the above rationale may apply where single character transliterations are also very ambiguous. E.g. In Cantonese and Korean, each Chinese/hanja character is also pronounced as a single syllable and many characters sound the same. 

However, where a single character has a polysyllabic transliteration, confusion tends to be more likely, so a section 44 ground for rejection is more likely to be appropriate. For example, in Japanese, the single kanji character 冬 can be read as FUYU. 

When in doubt, examiners should consult a subject matter expert. 

Multiple Chinese characters vs transliteration 

Example 1: 

ZHENG WU

vs 

ZHENG WU is a transliteration of 正誤 using pinyin. 

Once multiple characters are represented in each trade mark, the risk of confusion increases. Consumers are more likely to reasonably assume that they are the same word/s, referring to the same trader. 

A ground for rejection under section 44 would be appropriate when comparing two or more Chinese (Han) characters and their transliteration (including in pinyin). 

See for example Southcorp Brands Pty Ltd v Australia Rush Rich Winery Pty Ltd [2019] FCA 720, where 奔富 (and other trade marks containing 奔富 plus descriptive words in Chinese) were found to be deceptively similar or substantially identical to trade marks for both the same characters, 奔富, and their transliteration, BEN FU. 

7.4.3 Examples of comparisons

Example 1: 

Trade Mark 1 (Earlier mark)	Trade Mark 2 (Subject mark)
Class 35: clothing	鴛鴦   Class 35: clothing
Nil	Endorsements: The applicant has advised that the words 鴛鴦 appearing in this trade mark may be transliterated as YUAN YANG, and translated into English as MANDARIN DUCKS (AIX GALERICULATA).
Nil	Transliteration indexing constituents:  YUAN鸳 YANG鸯

A section 44 ground for rejection arises here. YUAN YANG is the pinyin transliteration of 鴛鴦. They are the dominant elements of these trade marks. Consumers aware of the 鴛鴦 trade mark are likely to assume that YUAN YANG refers to the same words (and vice versa), and therefore share the same sound and meaning.  

Example 2: 

Trade Mark 1 (Earlier mark)	Trade Mark 2 (Subject mark)
Class 14: jewellery	Class 14: jewellery
Endorsements: The applicant has advised that the Chinese words 庆丰珠宝 appearing in this trade mark may be transliterated as HING FUNG ZYU BOU, where 庆丰 has no meaning and 珠宝 may be translated into English as JEWELLERY.	Endorsements: The applicant has advised that the Chinese words 慶豐 appearing in this trade mark may be transliterated as QING FENG, and translated into English as CELEBRATION OF HARVEST.
Transliteration indexing constituents:  QING 庆 FENG丰 ZHU珠 BAO宝	Transliteration indexing constituents:  QING 庆 FENG丰

A section 44 ground for rejection arises here. When comparing the trade marks, key considerations include: 

• We can reasonably assume a significant target market for each trade mark would be consumers who can read Chinese. 

• 庆丰 and 慶豐 are simplified and traditional Chinese versions of the same characters. Note the Transliteration indexing constituents, which show all Chinese (Han) characters in simplified form. This can also be confirmed with a dictionary. 

• 庆丰/慶豐 is the dominant element in both trade marks. 

• In Trade Mark 1, the additional word 珠宝 means JEWELLERY, and is directly descriptive of the relevant goods. 

Note: For the purpose of section 44, the translation and transliteration provided by the applicant should be considered, but are not determinative. 

• Translation can differ based on intention and context. For example, names or invented words in Chinese are often formed from characters that have a dictionary meaning, but are used phonetically and/or evocatively only, so applicants may advise that the word/phrase as a whole has no meaning. 

• Transliteration can differ based on dialect/language. For example, Trade Mark 1 provides the Cantonese Chinese transliteration, while Trade Mark 2 provides a different transliteration in Mandarin Chinese (pinyin). What matters here is that, in any given Chinese dialect/language, the trade marks will be read and understood as the same words. 

7.5 Guidelines for comparison – Japanese characters

Japanese utilises three writing systems or alphabets:

• Kanji (漢字) – essentially traditional Chinese (Han) characters. Used for regular Japanese words.

• Hiragana (ひらがな) – the basic phonetic alphabet. Used for grammatical particles and simple words.

• Katakana (カタカナ) – an alternative phonetic alphabet. Used to write foreign loan words and sounds.

   

A Japanese word written in any of these three writing systems would have the same meaning and pronunciation. Although there would usually not be any visual similarities, each of these would generally be considered deceptively similar to each other.

The below example shows the same words written in each of the three writing systems and the equivalent transliteration. They all have the same pronunciation and all mean “Funakoshi Set Meal”.　

The above 4 examples would all be considered to be deceptively similar. 

Trade mark 1 above would also be considered deceptively similar to the Chinese transliteration of these characters (CHUAN YUE DING SHI), as per the above guidelines for Chinese characters.

7.6 Summary

The following table summarises the general approach for comparing trade marks comprised of non-Roman character/s against: 

• the same non-Roman character/s; or 

• the transliteration of the same character/s; or 

• the English translation of the same character/s. 

Please note that this is a general, illustrative guide only. Each case should be considered on its own merits, applying a risk management approach and the usual tests. When in doubt, examiners should consult a subject matter expert. 

7.7 Common descriptors in Chinese 

The following are some descriptive terms which commonly appear in Chinese trade marks. When comparing Chinese trade marks, it may be appropriate to give these descriptors less weight compared to more distinctive elements. However, the usual tests for deceptive similarity apply, and these descriptors cannot be entirely ignored in the comparison. For more guidance on considering non-distinctive material when comparing trade marks, see Part 26.6.7 The descriptiveness of the trade mark.

Four categories of common descriptors are covered:

• Company descriptors

• Trade mark descriptors

• Common geographical names

• Common trade descriptors

* Please note: this is not an exhaustive list of common descriptors in Chinese. This list may be expanded over time.

Company descriptors

Please note these are descriptors only and do not necessarily indicate that the company has legal personality.

Only the Chinese transliterations in the List of Acceptable Overseas Entities (see Ownership Crib Sheet) can always be considered as indicators of legal personality.

Trade mark descriptors

Common geographical names

Common trade descriptors