Case notes: S01E04 – David Elliott


In 2012, David Elliott filed a lawsuit arguing that the word Google has become so generic, it should be stripped of its trademarked status. Would David or the goliath Google win? Below you will find the data, audio credits, further reading, and a transcript of the podcast.

Audio credits

Scott Holmes – Infrastructure
Kai Engel – Calls and Echoes
Lee Rosevere – Healing (looped)
Scott Holmes – Epic Cinematic

References, sources, and more

June 2003: BBC News
May 2012: CNET
June 2012: Wirtz Law press releases and court documents
March 2017: Ninth Circuit Hearing (video)
May 2017: Ninth Circuit opinion
May 2017: Silicon Beat
June 2017: Ninth Circuit amended opinion
May 2017: Technology & Marketing Blog
August 2017: Petition for writ of certiorari (the “appeal of the appeal”)
August 2017: BusinessWire
October 2017: Supreme Court response to writ of certiorari

Google’s trademark DOs and DON’Ts
US Trademark law
Weird trademarks

Geoffrey Nunberg (Google’s expert linguist)
James T. Berger (Elliott’s expert survey creator)
Patrick Farrell (Elliott’s expert linguist)
Allen Metcalf (Elliott’s expert linguist)

Adam Kilgarriff (2015) Corpus linguistics in trademark cases (article)
Roger Shuy (2002) Linguistic Battles in Trademark Disputes (book)


Case S01E04 – David Elliott

We’ll start this episode a little differently than usual. I’m going to read a list of fifteen words. There are over thirty in the Case Notes transcript in case you’d like to challenge yourself further. Some of these words are, or were brand names – in the UK or beyond. Some, to my knowledge, have never been brand names. See if you can identify the all the ones that are, or were brands. Ready? Here we go.


How did you do?

Okay, so, I lied – an excellent habit for a podcast on language and crime. Actually, they all are, or were, once, trademarked brand names. Some of them are now ghost-marks – that is, dead trademarks. Surprised by any of them? More to the point, why are we playing a strange corporate word-association-game? Here comes a very long answer to that question – namely, the rest of the podcast…

It’s March 2012. We’re in Phoenix, Arizona. Spring has brought with it a few cloudier days and it’s still a little cool – only around the low twenties (that’s around 70 degrees Fahrenheit).

David Elliott, using his friend Christopher Gillespie’s GoDaddy account, has just finished registering 763 domain names. This is nothing unusual. People register thousands of websites every day. However, these domain names are a little different. They all contain one particular word – Google. Elliott has registered, for instance, and and and and… you get the idea. He’s registered a lot of sites with the word Google in the name.

Somewhat predictably, Google responds, and fast. It files a complaint that the registrations have been carried out in bad faith and that Gillespie and Elliott have no legitimate interest in them.

A few weeks pass by, and an almost uninterrupted run of brilliant sunshine rockets the temperatures up as high as the low 40s – that’s close to 110, for those working in Fahrenheit. As May 2012 arrives, so too, does bad news, and it’s addressed to David Elliott. Google’s complaint has been upheld. Elliott must hand the domain names over.

What is Elliott’s response? This one man – very soon two since Gillespie will shortly join the new action, but right now just this one guy, David – decides to take on the goliath that is Google. His ambitious master plan? Strip Google of two of their trademarks. These are trademarks 2884502 (or for ease, the ‘502 mark), and 2806075 (or again, for ease, the ‘075 mark). Why these two? These are the trademarks that protect the brand name, Google. I’m simplifying a lot here, but ‘502 effectively covers computer hardware and software, and ‘075 covers services, indexes, databases, or in short, a global computer information network.

Just for context, I don’t think Elliott or Gillespie are billionaires. I could be wrong but I’m going to assume that they’re both people of ordinary means. By contrast, in 2012, when this all really kicks off, Google is making just shy of $6 million per hour, or around $4 billion per month. It can afford an army of world-leading, sharply dressed super-lawyers.

Even if Elliott had an outstanding case, this would be a mountain to climb. Good lawyers can put in endless blocks and delays to increase costs and drag out the battle. Moreover, Google has an extremely vested interest in winning this case. If it lost, the consequences would be seismic. Google would lose control of its brand name. This is a trademark that Google itself estimates is worth $113,669,000,000. Anyone could start using it for their own purposes. All those billions of dollars? They would start tumbling, and fast. So, in its quest to grab a few domain names, could Google have inadvertently lost control of its billion-dollar empire?


Welcome to en clair, an archive of forensic linguistics, literary detection, and language mysteries. You can find case notes about this episode, including credits, acknowledgements, and links to further reading at the blog. The web address is given at the end of this podcast.

Trademarks and tradewars

Before we get stuck back into David versus goliath, to really understand the forces – and theories – at play, it’s useful to know a few important things about trademarks. So, let’s do five facts, just to get started.

Fact one: You can trademark all kinds of surprising things – smells, sounds, colours, images, and so on, but the cases we often hear about and the one this episode is concerned with today, is trademarked words, or phrases, and in particular, trademarked brand names. The principle purpose of brand name trademarks is to tell you the origins, or producer, of the service or product in question. So, these are words that have been given a sort of special, legally protected status. That legal protection means that the trademark holder can control how others use that word or phrase – within reason. Want a quick example? Look up the case of McSleep Inns.

Fact two: you can’t just go out there and trademark any words you like. For instance, in 2004, Donald Trump sought to trademark the phrase, “You’re fired!” and – all terrible, predictable jokes aside – the application was rejected. Common words and phrases are usually a no-go, so you’d be extremely lucky to get a trademark on the word water, or sky, or whatever. I say usually – looking at you, Microsoft and your rather surprising trademark for the word Windows. Also you can’t trademark that which hath been already trademarked. So in 2010, Nicole Polizzi of Jersey Shore tried to trademark her nickname, Snooki. I hope I’m saying that right. The USPTO (United States Patents & Trademarks Office) denied her name trademark status because, as it turns out, there is a British fictional cat with an almost identical name. Go the Brits. This Snooky, spelled with a Y rather than an I, is the eponymous hero of the “Adventures of Snooky” line of books and was awarded trademark of the name years earlier. As a result, the USPTO was concerned that it might be confusing to the public to have two Snookies out there. There can be only one real Snooky.

Fact three: Grammar matters in these kinds of trade mark disputes. There’s an organisation called International Trademark Association (or INTA) that gives guidance to people with trademarks, and helps them try to protect and enforce those trademarks. Their very blunt advice is as follows:

Trademarks and service marks are proper adjectives. Not nouns. Not verbs. A mark should always be used as an adjective qualifying a generic noun that defines the product or service.

They give examples like, Put on your Ray-Ban sunglasses and Send it by FEDEX courier. Notice each time how it’s Ray-Ban sunglasses and FEDEX courier. The specific brand name modifies a generic product. By contrast, they strongly recommend against businesses using their own trademarks in sentences like Put on your Ray-Bans or FEDEX it. In these cases, Ray-Ban has become a noun, and FEDEX has become a verb. Unfortunately, language as an evolving entity, laughs in the face of your rules, and doesn’t give a fig what the INTA thinks. People will verb nouns and noun adjectives all day long if it suits them, and it’s difficult to explain to the ordinary person in the street how saying, “Pass me a Kleenex” or “I hate hoovering” causes any harm to anyone, least of all a corporate giant somewhere out there rolling in billions of dollars. But this tiny, daily erosion, like the rain on cliffs, does eventually have an effect. As the INTA argues, such uses pave the sure and steady road to the ultimate loss of your trademark, as people begin to forget that the brand name is distinct from the product or service itself. This leads us into…

Fact four: you can lose your trademark. There are several reasons why – maybe you obtained it fraudulently. Maybe you got it and never used it. Or, just maybe, your trademarked words have become so widely used by the public, that they have fallen victim to something called genericide. Now, just being a popular brand name alone isn’t good enough, otherwise Coca Cola and Cadbury and so on would all have lost their trademark status. A specific process has to occur, whereby the brand name – Harley Davidson, Sony, Apple – has become a synonym for a specific product or service, to the extent that people use it to refer to that item. For instance, thermos, yo-yo, zipper. These were all brand names, like Hitachi and Microsoft, but in the case of, say, trampoline, that producer has become the word for the product.

Now, I’m simplifying here a bit. Sometimes the brand name and the product are invented together. This is particularly true of new toys. That thing never existed before, someone invents it, and they call it… a trampoline. In such a case, the brand name, and the name for the object are one and the same, and, at least in the early days, there’s only one source for this product. These are known as single-source cases and they’re especially legally difficult. However, in other cases there are multiple competing makers of an object, or it already had a non-proprietary name to begin with, and someone just came along and marketed their version so much better than everyone else. So, escalators were originally called moving staircases, and sellotape is actually sticky tape, and a biro is a ballpoint pen, and a zimmer frame is a walking frame and so on. But with effective marketing, a proprietary eponym can ruthlessly supplant these alternatives so completely that the original, common name is pushed down to second place, or even pushed out of mainstream use altogether. If you like this sort of fun, try saying, “Hey, let’s take the moving staircase” or “My head hurts, do you have any acetylsalicylic acid?” from now on and see what reactions you get. And finally…

Fact five: when there is a dispute over whether a trademark has suffered genericide or not, the main method used in court to determine its status, is the primary significance test. Sometimes it’s called the principle significance test. What is this test? Well, it’s a major cornerstone in trademark law, and at its simplest, it’s a test to see how the public thinks of a word. When they hear, for instance, Hoover, what do they think of? What is the first, or primary, significance in their mind for that word? Do they think of a brand that makes lots of different appliances? Or do they think of a specific object – in this case, a vacuum cleaner, as is probably the case in the UK? According to the United States Code 15, §1064, subsection 3, and I’m kinda paraphrasing here, if the majority of the public who use the term are unaware that it is also a trademark or a brand name – in other words, if it has lost its primary significance as a brand – then the word is deemed generic and the trademark can be lost. A quick sidepoint here: there are specialists who administer questionnaires with the sole purpose of finding out what the primary significance of a brand is for particular demographics, and those specialists can be called up in court to present evidence to help the judge decide on the outcome of a trademark dispute. Another quick sidepoint: corpus linguistics has an enormous amount to offer this field in assisting both brands and courts with understanding how their trademarks are used on a daily basis, but rather than getting stuck into that, I’ll save some examples for a future episode.

Okay, so to wrap up this section, for the purposes of the rest of this case, you need to hang onto just two details: Firstly, the advice is that you shouldn’t use, or allow your trademark to be used, as anything other than an adjective, because verbing or nouning is a quick route to genericide as people forget that it’s a brand. Secondly, if the majority of people, when they hear your brand name, think of a product, rather than a producer, then your trademark has suffered from genericide and can be stripped by the court. Who cares if your lose your trademark? Well, you, the trademark holder, does, because it means your profits will almost certainly tank as rival businesses start using your trademark to sell their own goods, and as people stop coming to you specifically for their jeeps and jetskis and rollerblades.

So now that we have the theory covered, let’s get back to David Elliott who wants his 763 domain names back, and is going about this by trying to strip Google of its brand name trademarks. What were Elliott’s arguments?

2013: David versus Goliath

So, as I said, Elliott’s first move is to try to cancel the two trademarks – ‘075 and ‘502 – on the grounds of generic usage. Quite how persuasive Elliott’s initial arguments were is difficult to determine. It’s really hard to find much documentation on this first step in the lawsuit, but whatever the case, Google takes the challenge seriously. They promptly wheel out the big guns and fire back with a countersuit. In other words, they don’t just defend themselves, they take up litigation in return against Elliott. In this countersuit, Google alleges, amongst other things, trademark dilution, cybersquatting, unfair competition, and false advertising.

Not to be outdone, Elliot rebuts. In February 2013, his legal team files Document 86, which lays out a slew of reasons why he continues to think that Google should lose its trademarks. He opens with a claim that Google has failed to successfully defend its trademark in this case already because its only evidence is…

its market share, a flawed survey, the opinion of a lexicologist who admits that google when used as a verb bears no trademark significance, and incomplete dictionary evidence from sources it has intimidated into submission.

I’m not sure what’s more gripping here – the subtle shade that seems to be thrown at Google’s linguist or the hints at lexicological terrorism conspiracy theories…

Whatever the case, Elliott now clearly formulates a key argument that runs through all of his legal challenges, which is this: he claims that Google is being used most often as a verb – that is when we say things like, I googled her email address or Could you google that for me? And he says, hey hey, verbs can’t be trademarks, right? So if Google is mostly occurring as a verb, and verbs can’t be trademarked… then tada! The trademark should be lost! We’ll get back to the logic of this position in a bit. For now, we’ll just look at how Elliott proves his claim.

To be fair to him, he does go through quite a range of potential sources. These include how Google is used on television shows, in films, news, magazines, novels, cartoons, and so on. He also points out that it appears in thousands of domain names offered by the GoDaddy auction site where Google is incorporated into the name as a verb (you know, like the domain names he registered himself). He argues that when the American Dialect Society chose their word of the decade, they selected the verb-variant of Google. And he claims that his own search engine, when asked to define google, returns a lower-case verb-only definition. Amusingly, I googled define google on, well, Google, and sure enough, the top hit was indeed its own dictionary definition entry for the verb. I’ve included a screenshot in the Case Notes but you could just run the same query yourself. Even more amusingly, in Google’s own Rules for Proper Usage they give a list of Google Trademark Dos and Don’ts, and their list of Dos contains the following:

  • Use the trademark only as an adjective, never as a noun or verb and never in the plural or possessive form.

  • Use the generic term for the product following the trademark, for example: GOOGLE search engine, Google search, GOOGLE web search.

I’ve put a link to this page in the Case Notes too.

Elliott also uses three sets of survey results. In the first two sets, the questionnaire is done by his legal team. I can only speculate here, but I guess they thought, “Eh, it’s just a questionnaire. How hard can this be?” They interview 1,033 members of the “relevant public” (who relevant is, I do not know). Anyway, these members of the relevant public were asked to complete the phrase, “I most often use the word ‘google’ to mean:”

52.2% chose “to search something on the Internet” – that is, the verb-variant.

Only 28.7% said “the name of a specific search engine” – that is, something like an adjectival or noun variant.

There are some more convolutions to these two surveys and the third one that’s administered by a business that specialises in conducting questionnaires, but I’ll return to this morass at its proper place in the story.

So what else does Elliott cite to back up his argument? He throws in more examples about how people say they “googled on X”, for instance, googled on Wikipedia, googled on IMDb, googled on pinterest, and so on. His document doesn’t really spell this out clearly, and more is the shame, but here he seems to be making the implicit argument that these people mean that they used the inbuilt search functions on each of these non-Google platforms and simply refer to the mechanism or act of searching as googling.

There are some other arguments besides but for brevity, I’ll skip onto the next big one. Elliott argues that Google has failed to show that the primary significance of the word Google is as a trademark. So remember, primary significance is all about what you think of first and foremost when you see a word, like hula-hoop or Velcro. Do you think specific item? Or do you think, “Oh! That’s a brand name! They also make or do other things besides.”

Here, Elliott takes issue with the test that Google uses to determine that the brand name is still the primary signifier. In their evidence, they wheeled out something known as a Teflon survey. What is a Teflon survey? You did one at the start of this episode. In simple terms, give people a list of words and ask them to identify those that are brands and those that are not. Elliott argues here though that this is skewed towards products – that is, nouns or objects, and his whole argument is based on Google’s usage as a verb – something the Teflon survey simply doesn’t address.

One final bit from Elliott’s suit. In section 7, he takes aim at Google’s linguist. To give you this bit straight, I’m just going to quote from the document, because honestly, I can’t make this better than it already is:

Section IV(A(7)), p12.

Defendant’s Linguist Is Conflicted, His Confusions Are of Law and Without Factual Support, and He Admits That Use of Google as a Verb Cannot Have Trademark Significance.

Defendant’s expert linguist, Geoffrey Nunberg, was first contacted regarding his services by Plaintiff’s counsel in December of 2012.

Geoffrey Nunberg, to give you some context, is an American linguist and researcher. At present he’s an adjunct professor at the UC Berkeley School of Information and in the past he’s taught at Stanford. Nunberg would have in his late sixties at the time of the case. Back to the quote.

Plaintiffs counsel sent Nunberg the complaint asking if he would be an expert in the case. In addition, Plaintiffs’ counsel left Nunberg a voicemail explaining Plaintiffs’ theories in the case. In response, Nunberg emailed that he thought the case was “interesting” and that he would be “happy” to discuss being Plaintiff’s expert. On December 5, 2012, Nunberg and Plaintiff’s counsel had a telephone conversation about the case and Plaintiffs’ theories and strategies thereof. Following the conversation, Plaintiff felt that Nunberg was “on board” with being Plaintiffs’ expert witness. The next day, Nunberg sent Plaintiffs’ counsel several expert witness reports he had prepared in prior cases to ensure that he had not previously expressed any conflicting opinions that could be used to impeach the opinion he would give in this case (at that time contemplating an opinion in favor of Plaintiffs). After reviewing these reports, on January 30, 2013, Plaintiffs’ counsel told Nunberg that Plaintiffs were interested in retaining Nunberg as an expert witness. On March 4, 2013, however, Nunberg told Plaintiffs’ counsel “I was looking forward to working on [your case], but some health issues make it difficult for me to make commitments over the intermediate term, so I’ll have to pass.”

Despite all this, Nunberg has now reversed his opinion and provided an opinion for Defendant. He is clearly a hired gun who will say anything he is paid to say. Several of Nunberg’s opinions are conclusions on the ultimate issues of this case: such as whether widespread use of a trademark as a verb affects the validity of the mark. Clearly Nunberg’s opinions on this are improper. In addition, some of Nunberg’s opinions contradict existing law. Nunberg opines that a trademark can be “representative” of a genus without being generic. However the very definition of a generic word is one which represents the genus, rather than the product of one specific producer. Nunberg’s contrary opinion must be disregarded. Finally, Nunberg opines that a verb can never function as a trademark. As will be discussed in the following section (and in Plaintiffs’ motion for summary judgment), the logical conclusion to be drawn from this opinion is that verb usage is necessary non-trademark (or generic) usage. Thus, if the majority usage of a word is as a verb, then the word is generic.

Often legal documents are dry and dull and tedious but just sometimes, they’re better than a good book. Also my sense of humour is warped.

But you know what Elliott and his team don’t use? They don’t use a corpus linguist…

Moving on, in short, as you can see, Elliott’s overriding argument is that Google as a verb is the dominant usage, that trademark guidelines already dictate that verbs are de facto not trademarks anyway, and that as a result, Google should lose its trademark status. And that their linguist is a hired gun, of course.

So, David has fired a whole arsenal at goliath. How will Google respond?

2014: Goliath versus David

We move forward about two and a half years. It’s now October 2014, the court has published its findings, and, to cut a long story short… Google successfully defends itself. Why? Well, the Court takes apart the two prongs of Elliott’s suit quickly and efficiently, as follows.

Firstly, they deal with this supposed overwhelming verb-usage. As I’m sure you are now very clear, Elliott’s argument rests on the fact that verb-google is dominant, and also non-trademarkable, ergo it’s generic. Helpfully, the court identifies two usages here – indiscriminate, and discriminate verb usage. The discriminate version, they suggest, is when users say they’re google, and they mean with google. The indiscriminate version is where people say they’re googling, and they mean, they’re just using a search function, perhaps on Wikipedia or Tindr or whatever. I’ll come back to this indiscriminate usage in a bit.

Google themselves don’t contest the fact that their brand name is used as a verb. Instead, they stick to the central tenet that no matter how people use the word, when they hear and think of Google, they associate it with a specific brand. Remember, trademarks are about identifying the source of a product or service, and if you hear Google and still think of, say, the Google logo, the corporation, the creator of a search engine, and so on, then it still has primary significance as a brand. It doesn’t matter whether you also use it as a verb too, nor even whether you use it as a verb almost all the time. So much for grammar.

The second point that the Court addresses is the admissibility of Google’s expert linguist. Remember, Elliott wasn’t so keen on Geoff Nunberg and described him as a “hired gun”? Well, the court has something to say on that front too.

Plaintiffs attack Dr. Nunberg as a “hired gun who will say anything he is paid to say” because he allegedly “reversed his opinion.” (Doc. 86 at 12.) While inconsistencies may be an indicator of reliability, see Daubert, 509 U.S. at 590 n.9, Plaintiffs do not substantiate their allegation that Dr. Nunberg reversed his opinion. The fact that Dr. Nunberg first expressed interest in being retained by Plaintiffs before being subsequently retained by Defendant does not necessarily mean Dr. Nunberg gave inconsistent professional opinions. To the contrary, the only evidence in the record is Dr. Nunberg’s testimony that Plaintiffs never retained, paid, or shared any confidential or work product information with him, that he never shared any of Plaintiffs’ information with Defendant, and that while he may have shared ideas with Plaintiffs, the only expert opinion he rendered was the one contained in his report.

But there’s more. The Court document also takes issue with the three surveys carried out by Elliott’s counsel. So remember, those first two surveys asked people to complete the sentence: “I most often use the word google to mean…” The first survey offered the answers,

  • “to search something on the internet”
  • “the name of a specific search engine”; and
  • “the internet (in general)”

The second survey asks the same question with the following answers:

  • “to search something on the internet” (again);
  • “the name of a company”; and
  • “the internet (in general)” (again)

Notice how neither offers the respondent an answer along the lines of, “to search for information using the Google search engine”? I could dedicate an entire podcast series to how writing and administering a good questionnaire takes an awful lot of thought and care. The questions can be leading or slippery. The range of answer choices can be manipulative. The way the questionnaire is framed and given can change how people respond. Literally just smiling at the person or offering them a drink beforehand or explaining the purpose of your study in one way or another can change the outcomes. There are so many variables to account for and control. In fact, even Elliott’s expert survey designer, Mr James Berger, came in for some criticism and the evidence gathered from his survey was ruled only partially admissible. Document 116 has this damning little aside in it as follows:

Mr. Berger stated that his survey tested neither the primary significance of the term Google to consumers nor whether the term was generic with respect to search engine hardware and software that are the subject of the ‘502 and ‘075 Marks.

One rather wonders quite what it was testing then.

Whatever the case, this wasn’t the only issue with the evidence Elliott put forwards. When it comes to the dictionary usages, Elliott’s team were unable to provide a single example of an entry that didn’t also include the trademark somewhere in the definitions.

They were unable to cite a single instance in which a major media outlet referred to another search engine as Google.

And they couldn’t produce convincing evidence for their assertion that the verb-usage of google is more frequent than non-verb usage. Google even goes as far as to suggest that there is simply no way to provide evidence of this kind. As a corpus linguist, I have so many good responses to this. If you do want to know whether or not a particular usage is more dominant in a particular cross-section of society, speak to a corpus linguist people, come on.

Finally, the outcome is that Elliott has not been able to meet the burden of evidence required, and the court rules thus:

The Court is mindful that “summary judgment is generally disfavored in the trademark arena” due to “the intensely factual nature of trademark disputes.” Rudolph Int’l,, 482 F.3d at 1199 n.3 (quoting KP Permanent Make-Up, Inc., 408 F.3d at 602). However, summary judgment is nevertheless appropriate when there is no genuine issue of material fact. Id. At 1199. Such is the case here.

A little later on, the Court adds…

Plaintiffs, at their peril, neglected their burden of proof under the primary significance test, instead electing to present evidence about whether a majority of the consuming public understood the word google as a verb. Disregarding primary significance resulted in an absolute failure of proof that is fatal to Plaintiffs’ claim for genericide.

2017: David doesn’t give up that easily

Surely, this is the end of the matter? It is not. Of course. Elliott had the game to bring the suit in the first place and he sure as heck isn’t giving up easily. Elliott appeals the judgement and the case goes to the Ninth Circuit. In March 2017 there is a hearing in Courtroom Two of the San Francisco Court of Appeals, and finally, we get an answer – of sorts – to quite what Elliott wanted the sites for anyway. Have you been wondering this? It bugged me for the longest time! What was he even gonna do with all those random domains? So, that mystery at least, is resolved… kind of…

[Excerpt from 17th March 2017 hearing here, 9m 50s to 11m 50s:]

In the end, Judge Richard C. Tallman gives his opinion and in it, he affirms the findings the district court’s summary judgement in favour of Google. Elliott’s appeal is described thus:

The plaintiffs produced thousands of pages of largely irrelevant evidence showing merely that “google” is sometimes used as a verb. The sliver of potentially relevant evidence purporting to show that the public uses the verb “google” to refer to searching the Internet with any search engine (as opposed to Google’s search engine in particular) is too insubstantial to save the plaintiffs’ case.

That’s the key little bit I wanted to come back to. Elliott had in his possession a few slivers of evidence showing that sometimes people use verb-google to mean searching using the search function on any site. The court noted this as indiscriminate verb usage. Had he pursued that angle, he might have actually made some serious headway.

Anyway, a little later, the Court adds that,

the expert linguists [working for Elliott’s Counsel] conceded in their depositions that, despite their opinion that “google” is used in verb form without regard to a specific search engine, the term has not become a generic name for search engines.

Those linguists, incidentally, were Dr Patrick Farrell, Professor Emeritus of the Department of Linguistics at the University of California, Davis, and Dr Allen Metcalf, Professor of English at MacMurray College, in Jacksonville, Illinois.

Surely, that is really, truly, the end of the matter? You know what’s coming. Elliot heads for the Supreme Court with a writ for certiorari. In extremely simplistic terms, he wants to appeal the appeal and for this, he puts forward three more arguments: that the Ninth Circuit didn’t sort out where the whole verb angle of his argument fits into trademark genericisation; that their opinion conflicts with an argument by the Second Circuit about changing the test for genericness from “majority usage” to “majority understanding”; and that the Ninth Circuit supposedly misinterprets or follows a contradictory and unworkable prior decision.

Finally, five months later, on Monday 16th October 2017, there is a hearing to determine whether this writ of certiorari – that is, the appeal of the appeal – will be allowed.

The petition… is denied.

Will Elliott try any further action? I do not know. I get the sense that this guy does not give up easily, so… stay tuned.


This episode of en clair was entirely researched, narrated, and produced by me, Dr Claire Hardaker. However this work wouldn’t exist in its current form without the prior effort of many others. You can find acknowledgements and references for those people at the blog. Also there you can find data, links, articles, pictures, older cases, and more besides.

The address for the blog is And you can follow the podcast on Twitter at _enclair. Or if you like, you can follow me on Twitter DrClaireH.