Founders name companies after themselves constantly, and it is the one naming instinct the Trademark Act treats with suspicion. Section 2(e)(4) says a mark that is primarily merely a surname cannot go on the Principal Register without more.
Why the law singles out surnames
The reasoning is about fairness between people who share a name. If the first Miller to file could lock up MILLER, every other Miller in the country would be shut out of using their own name in business. So a surname starts unregistrable and has to earn its way onto the register, in the same way a descriptive word does.
Note the wording, though. The bar is not "is a surname". It is "primarily merely a surname", and the manual notes that "primarily" was added deliberately when the Act was written.
What the USPTO actually asks
The test is the primary significance of the mark as a whole to the purchasing public. TMEP section 1211.01 lists five inquiries the Trademark Trial and Appeal Board has identified, and they are worth knowing before you file:
- Is the surname rare? A name almost nobody encounters may not register as a surname in the public mind at all.
- Is it the surname of anyone connected with the applicant? Your own name counts against you here, which feels backwards but follows from the logic.
- Does the term have any recognized meaning other than as a surname? This is the most useful one, and the next section covers it.
- Does it have the structure and pronunciation of a surname? Some words simply read as surnames even when unfamiliar.
- Is the lettering distinctive enough to create a separate commercial impression? This only applies to stylized marks. For a plain word mark in standard characters, the manual says this inquiry is unnecessary.
There is no fixed amount of evidence required, and the manual is explicit that each case is decided on its own facts.
The "other meaning" escape route
The third inquiry is where most winnable arguments live. If the word has a recognized meaning beyond being somebody's name, the primary significance to the public may not be surname at all.
Ordinary words that happen to be surnames are the clearest case. So are given names, place names and words with dictionary meanings. The question is always what the public thinks first when they see the word, which is why the analysis depends so heavily on evidence rather than on instinct.
The two ways to register one anyway
Acquired distinctiveness under section 2(f). If you can show the public has come to treat your surname as a brand, it registers on the Principal Register. Five years of substantially exclusive and continuous use is the usual benchmark, though it is not a guarantee, and advertising spend, sales volume and unsolicited press all help.
The Supplemental Register. A surname can go there in the meantime under section 1 or section 44. You do not get the full benefits of the Principal Register, but you get a federal registration, the ® symbol, and a basis for blocking later confusingly similar applications while your distinctiveness builds.
Before you commit to the family name
Two practical things. First, the surname question is separate from availability: even a perfectly registrable surname is no use if another business already holds something similar for related goods, which is what a search is for. Second, a surname refusal is only one of the grounds an examiner can raise, and the others apply to your name too.
If you are weighing a family name against a coined one, the distinctiveness spectrum explains what you give up by choosing a name that the law makes you earn.