Lexicalize different?

« previous post | next post »

Ben Zimmer, "The Great Language Land Grab", NYT 3/27/2011:

When tech companies engage in legal squabbles about who gets to use our everyday words, what are ordinary speakers of the language to make of it all?

Microsoft is suing Apple, and Apple is suing Amazon, all over the right to use a simple two-word phrase: “app store.” […]

It’s not the first time the tech industry has claimed commonplace language as its own.

Facebook has been notorious in this regard, filing trademarks on an array of common four-letter words: “like,” “wall,” “poke” and, naturally, “face” and “book.” […]

Microsoft, of course, has long been playing this game by fiercely upholding prosaic brand names like Windows, Office and Word.

There was a flurry of news about the first "app store" lawsuit over the past couple of months, and the strangest contribution, IMHO, came from John Dvorak ("App means Apple", PC Magazine 3/2/2011):

Apple is going to war over the term "App Store," but it doesn't need to waste its time and money. We all know "app" means Apple. […]

The one edge in the debate that Apple has is that app is actually a truncated version of the word Apple and there is no other reason anyone should be using Apple in their store name. When I see App Store, I think Apple Store. By saying it uses App as kind of a pun, meaning both Apple and "application," the company could have a strong case. Although it's obvious to me, I have yet to see if Apple has made this argument.

The idea that "app" might be short for Apple had never occurred to me before, I guess because as Ben points out in his article,

“App” has been used by the computing crowd since at least 1985 as a short form of “application.”

Of course, John Dvorak knows this, since he's been writing columns about computing since about that same time. So maybe he's representative of a group who lexicalized the term differently. Or maybe he just needed a provocative thesis for that week.

Anyhow, the fact that Apple's lawyers don't seem to have advanced this theory, either before or after Mr. Dvorak's proposal, suggests that they don't think it would work. Instead, as I understand it, their theory is simply that the phrase "app store", whatever its etymology or interpretation, belongs to them.

And following up on Ben's "language land grab" idea, it's worth noting that something does seem to have changed over the last few decades. U.S. trademark registration has been available since 1881 — before that, trademarks were dealt with under common law — but over the past 130 years, innovators have failed to make any attempt to register many valuable new words, new combinations of old ones, or new commercial meanings of words or phrases.

For example, the OED's earliest citation for email is from 1984.  I'm pretty sure that the term was in use at Bell Labs when I started there in 1975 — though maybe we just called it "mail", as users of Multics and CTSS apparently did. In any case, at some point between 1965 and 1985, there were various opportunities to assert ownership of "mail" or various derivatives for the puposes of digitally-mediated communication, but this never seems to have occurred to anyone.

Moving back in time, the concept and word supermarket originated in the early 1930s — the OED's first citation is from 1933 — but as far as I know, no one tried to establish ownership of the word.

Alexander Graham Bell patented the telephone in 1876, but his company made no attempt to assert ownership of the word, which had been used for a half-century previous to that in a more general sense, e.g. for "a system of signalling by musical notes, devised by Sudré in 1828", or a "Marine Alarum and Signal Trumpet", or an 1860 invention by Wheatstone "in which musical pipes or free tongues are acted upon by wind. Compressed air or gas is admitted to the pipe by means of a valve acted upon by the magnetized needle of an electro~magnet. The alternation of long and short sounds may be grouped in a similar manner to the long and short lines in the alphabet of a Morse's telegraph."

It might have been plausible for AT&T to argue, as Apple now does about "app store", that use of the word "telephone" by its competitors "will confuse and mislead customers". And this would certainly have been helpful to Theodore Vail in his attempt to consolidate an effective monopoly of U.S. telephony after Bell's patent expired in 1894, under the slogan "One Policy, One System, Universal Service." But as far as I know, this idea never occurred to anyone.

It's easy to think of dozens of similar cases of (luckily?) lost opportunity — just in the area of new types of stores, there's "convenience store" (1960s), "video store" (1980s), "game store" (1980s), etc.  (I'm not talking about traffic in the other direction, where brand names like kleenex or xerox become genericized terms for types of product or process. Rather, the issue here is words or phrases of the general vocabulary, often already in use for varieties of the application in question, that are turned into commercial property.  As Ben's article suggests, this seems to have become increasingly common. The best-known case is probably "windows". Does anyone know what the first one was?


  1. Peter Corbett said,

    March 27, 2011 @ 8:33 am

    If I recall correctly, Dvorak has a bit of a reputation. Hmm, let's see what Wikipedia, the fount of all true knowledge, has to say:

    "Dvorak's pithy style often attracts critics[who?] who point out his frequent, and occasionally egregious, errors of prediction[vague]."

    Yes – this is what I recall.

    I think this is consistent with "maybe he just needed a provocative thesis for that week."

  2. Pflaumbaum said,

    March 27, 2011 @ 9:30 am

    There's a remarkable case going on at the moment over a London restaurant selling an ice-cream flavour called Baby Gaga, which is made from breast-milk. Apparently Lady Gaga has threatened legal action over the name – presumably not just the word gaga but the rhythm and assonance of the whole phrase.


  3. Spell Me Jeff said,

    March 27, 2011 @ 9:45 am

    Le Car might be a candidate: "The North American Renault 5 debuted in 1976 as the Le Car." I imagine the inclusion of "le" would have been key to deflecting any challenges, so maybe this is a less pure example than one might hope for.

  4. Spell Me Jeff said,

    March 27, 2011 @ 10:18 am

    Another possibility, with an interesting history. The Kodak Brownie (1900) was allegedly named after characters in a popular comic strip drawn by Palmer Cox, The Brownies, little sprites named after bona fide creatures in Celtic folklore. Kodak itself attests that this was the source of the name. The animated history is peopled with Palmer's Brownie characters.

    However, the 1900 patent application reveals that, while the assignee was in fact Kodak, the inventor was a fellow named Frank Brownell. Coincidence?

  5. Vicki said,

    March 27, 2011 @ 10:56 am

    "It's short for Apple" could lead to another problem for Apple's lawyers, if they tried to enforce it. Apple Computer doesn't own that trademark: it's licensed from the Beatles' Apple Corps. And then the computer company had to renegotiate that when they started making mp3 players and selling recorded music, because the original trademark licensing agreement was for computers and specifically did not include music. (They don't own "Macintosh" either: maybe they're overreaching now because they're worried someone else will if they don't?)

  6. David said,

    March 27, 2011 @ 11:00 am

    @ Peter Corbett Thank you for saying exactly what I was thinking as I read Liberman's article! What bothers me is that he gets a salary from this kind of stuff!

  7. mollymooly said,

    March 27, 2011 @ 12:30 pm

    "Applet" mkight have triggered stronger "Apple" resonances than "app" does; but Sun unsportingly got there before Apple.

  8. UK lawyer said,

    March 27, 2011 @ 12:37 pm

    To answer Mark's question, I don't know but Coca Cola could be a candidate: if I came up with a beverage that is based on coca wine and flavoured with the kola nut, I had better be careful about the name.

    [(myl) It's true that coca and kola were existing words, but both were a bit off the beaten path. Maybe a better analogy would be trademarking "lemon-lime" or "orange soda" or something like that.]

  9. Lazygal said,

    March 27, 2011 @ 12:50 pm

    In re: "apps". Talk to any restaurant – they've had apps since well before the computer age.

  10. John Cowan said,

    March 27, 2011 @ 1:28 pm

    Vicki: Apple definitely does own the trademark in connection with computer hardware and software, because those are different categories from music, and trademarks in different categories don't infringe one another. Apple Computer needed a license from Apple Records only when they began to sell music.

    On "Windows":

    Every proposed trademark is classified as fanciful, arbitrary, suggestive, or descriptive. The distinctions between the first three groups are merely technical; it's the last group that's important here. Descriptive marks merely describe the trademarked goods, and as such are inherently not trademarkable, as shown by the hypothetical examples of "Fast" modems, "104-key" keyboards, and "Tubeless" computer monitors — unless (and this is crucial) they have acquired in the minds of consumers the "secondary meaning" (outside the U.S., the "acquired distinctiveness") of indicating the source of the goods.

    Most computer operating systems provide windows nowadays, so "Windows" is a descriptive mark for operating systems, and it is only protectable because Microsoft has gone to a lot of trouble to insure that when people see the word "Windows" in connection with computers, they think of Microsoft. Such trademarks are always shakier than the other types.

    When Lindows Inc. started to sell a Linux-based operating system to compete with Windows, Microsoft sued, and they had to change their company name and product name to Linspire — but it was Microsoft who paid Lindows $$$$$ in the settlement, as well as giving them certaint rights to use Microsoft patents, now unfortunately expired.

  11. Rod Johnson said,

    March 27, 2011 @ 5:35 pm

    @David: "Thank you for saying exactly what I was thinking as I read Liberman's article! What bothers me is that he gets a salary from this kind of stuff!"

    Watch those antecedents, my friend (on the assumption that you don't think Mark gets a salary for this).

  12. Stuart Brown said,

    March 27, 2011 @ 6:08 pm

    You use of the OED reminds me of a very particular piece of language grab: that the word Oxford is trademarked by the OUP, and they aggressively protect it. I used to live there and presumably I, and the other 150,000 residents of the city, were infringing their property rights every time we gave our address.

  13. Dan K said,

    March 27, 2011 @ 6:49 pm

    Am I the only one who's surprised to hear that "app" goes back to 1985? I've been using and writing "applications" since the early '80's, but I don't recall ever hearing it back then, or really at all prior to apps for cell phones. Where I come from (east coast usa), that kind of usage would've gotten you branded as a poseur (i.e., someone pretending to be knowledgeable about computers) pretty quickly. Maybe a regional thing?

    [(myl) But you probably recall all the talk about the next "killer app" that was common by 1991 or so? Thus this 1991 story notes that "At this year's Comdex, Dr. Andrew S. Grove, the president of the Intel Corporation, will try to make the case that the next 'killer app,' as such programs are called in industry jargon, will be those that permit small and large groups of employees to work together more effectively." The OED cites a 1998 PC Week article about how "Until we see those first two or three killer apps‥few corporate customers will see much reason to begin to commit to OS/2". My impression is that in the late 1980s and early 1990s, the app usage behind the "killer app" cliché was common among business and marketing types, but not so much among developers.

    Ben Zimmer's Word Routes column ("How We Got an 'App' for that", 3/24/2011) cites a 1985 Info World article about Framework to the effect that ""At first look, the user sees only one new menu ('apps' for applications) at the top of the screen", which offers a screenshot showing the Framework apps menu itself. So at least some applications software was called "apps" in consumer-facing material in 1985.]

  14. Licia said,

    March 27, 2011 @ 6:53 pm

    For those like me who don’t know much about US trademarks, the Wikipedia entry on Trademark distinctiveness is quite informative, particularly the section on Arbitrary marks as it makes specific reference to dictionary meaning and it helps clarify all the references to dictionary entries in Microsoft’s motion: “Such marks consist of words or images which have some dictionary meaning before being adopted as trademarks, but which are used in connection with products or services unrelated to that dictionary meaning”.

    Incidentally, in the App Store discussions the name Windows is always brought up. Does any of you know if arbitrariness was one of the criteria that made Windows sufficiently distinctive at the time, despite it being a generic word? Possibly, when Microsoft submitted their claim for Windows, the IT meaning of window as “a rectangular area on a computer screen in which programs and content appear” was not yet documented by dictionaries, and this would have classified it as a semantic neologism, i.e. "arbitrary", so I am wondering if this is what made it distinctive enough to become a trademark?

    BTW, Word and Office are not registered trademarks, only Windows and the Office logo are, cf Microsoft Trademarks list.

    [(myl) I don't know about dictionary entries, but window(s) was certainly a generic term for "“a rectangular area on a computer screen in which programs and content appear" for some time before Microsoft Windows came out in 1985. I'm not sure whether the term was used for the windows in various 1960s precursors such as Engelbart's system at SRI or Ivan Sutherland's Sketchpad, but it certainly was used in reference to the systems developed at Xerox PARC in the early 1970s. And similarly in the Blit terminal that Rob Pike and Bart Locanthi built at at Bell Labs in the early 1980s.

    Supposedly, when Apple sued Microsoft for copying the look and feel of Apple's GUI, Bill Gates responded “No, Steve, I think it’s more like we both have a rich neighbor named Xerox, and you broke in to steal the TV set, and you found out I’d been there first, and you said, ‘Hey, that’s no fair! I wanted to steal the TV set!’”]

  15. Ilana said,

    March 27, 2011 @ 8:21 pm

    Then cheeseburger could be a candidate? At least Wikipedia says that it is trademarked and to put cheese on a hamburger and call the result cheeseburger seems so trivial to me that I find it hard to believe that cheeseburgers weren't called cheeseburgers before 1935. (But I can find only sparly references to cheeseburgers earlier, maybe they were indeed not so common.)

    If trademarks outside the Anglosphere are allowed, then
    Teekanne and Teehaus come to my mind, meaning 'tea pot' respective 'tea house'. In German these are generic names as in English. But what's in a name? Tea bags! ;-)
    So it's maybe not exactly what are you looking for and I am missing the point …

    But I can also remember, there is a Slovenian company Petrol which sells … uhm… petrol at its gas stations.

  16. Ellen K. said,

    March 27, 2011 @ 9:10 pm

    The "cheeseburger" trademark claim seems to be inaccurate. It doesn't even make sense in the context of the webpage… since the webpage otherwise treats it as a generic food name.

    According to this webpage, someone applied for a trademark for "cheeseburger", but wasn't successful: http://www.waymarking.com/waymarks/WM25C1_Cheeseburger_trademarked_in_Colorado

  17. maidhc said,

    March 27, 2011 @ 10:47 pm

    A trademark is granted for a particular business type, for which the USPTO has a big list. Having a trademark for one particular business type doesn't affect businesses in other areas.

    If Microsoft has a trademark on Windows as an operating system name, that doesn't affect the use of "window" as a generic term in other software contexts.

    Of courses companies who can afford to hire lots of lawyers can always argue that other trademarks dilute the value of theirs, so there's never a fixed answer. It can always end up in court.

  18. baylink said,

    March 27, 2011 @ 11:22 pm

    The Killer App was originally desktop publishing, which originated in 1985 when Apple brought out the Laserwriter to complement the Macintosh, and aldus released Pagemaker.

  19. Ben Zimmer said,

    March 27, 2011 @ 11:47 pm

    Very interesting discussion so far. Just a quick note: Mark mentioned the early popularization of email — as Dennis Baron noted here, CompuServe tried to trademark email but abandoned the attempt in 1984.

  20. phspaelti said,

    March 28, 2011 @ 2:51 am

    Well, at least we all know who trademarked "Breakfast experiment".

    [(myl) As I understand it, a trademark actually has to be registered in an official way to be valid — unlike copyright, which is considered to arise from the mere fact of creation. So there is no valid trademark in this case. But if there were, it would be a sort of Open Access one, with the widest possible use by others actively encouraged…]

  21. Graeme said,

    March 28, 2011 @ 6:30 am

    @Stuart Brown and others. A trade mark is just a way of registering what the common law fumblingly protects from 'passing off'. Its purpose is to protect attributions of a product's source. It isn't akin to copyright. Eg it lapses from non-use, and it's not infringed by social usages that are insufficiently commercial or which don't risk assumptions about origin.

  22. J.W. Brewer said,

    March 28, 2011 @ 7:57 am

    Maybe it hasn't happened as much in the computer end of high-tech, but in other innovative areas (perhaps esp. the chemically-based) it is common for those who have created a new product which will be patent-protected (such that the inventor and/or its licensees will initially have the market to themselves) to create both a trademarked name (which will hopefully be protectible in perpetuity) and a generic name which competitors will be able to use to describe their own version of the product once it goes off-patent. So, e.g., NutraSweet brand aspartame or Xanax brand alprazolam. (Note that the generic names sound both "scientific" and sort of commercially unappealing.) Originally the two names are de facto synonymous (i.e. when A is the only brand of B that's on the market, A and B have effectively the same referent from a consumer point of view), but this will change once competitors are free to enter.

  23. Rob P. said,

    March 28, 2011 @ 9:46 am

    A couple of questions above relate to whether a foreign language equivalent word could be registered (Le Car, being an example). In general, US trademark law treats foreign language words as if they were in English, both for registration and for infringement purposes. "Under the doctrine of foreign equivalents, foreign words from common languages are translated into English to determine genericness, descriptiveness, as well as similarity of connotation in order to ascertain confusing similarity with English word marks." But this is not an ironclad rule, "When it is unlikely that an American buyer will translate the foreign mark and will take it as it is, then the doctrine of foreign equivalents will not be applied."

  24. möngke said,

    March 28, 2011 @ 11:09 am

    @ Ilana: of course, in Slovene, petrol isn't called petrol, but "bencin". But I have always wondered how an English-speaking foreigner would look at the issue…

  25. Mark Dunan said,

    March 28, 2011 @ 11:17 am

    I'm bothered about the spelling "email" (or "Email") — do we call 6/6/44 "dday", followed the next year by "veday" and "vjday"? Are babies born by "csections"? Are government agents "gmen"?

  26. Jon Weinberg said,

    March 28, 2011 @ 11:22 am

    @myl: A trademark does not need to be registered to be valid. It does need to be used in commerce, which in this context means your using or displaying the phrase "Breakfast Experiment" in the sale or advertising of Language Log services. Perhaps you can assign the task to a functionary at Language Log Plaza.

  27. Rick Bryan said,

    March 28, 2011 @ 12:05 pm

    In 1987 or thereabouts, Symbolics had a product on the market called "Dynamic Windows". Microsoft came along and offered sufficiently interesting money to rename it, so they could concentrate (i.e. anti-dilute) their brand.

    In 1991 or so, at a a pen computing startup, I was involved in proposing to develop a custom software application for an insurance company. They were investigating a more direct way to capture data when a customer was applying for insurance. We were trying to sell our apps to do their apps, and the shared vocabulary for completely different terms was, well, perhaps it's just as well the deal never happened.

  28. Thom said,

    March 28, 2011 @ 2:44 pm

    It seems that Google might (inadvertently) agree with Ben. Using the Google quick search bar, typing in "app sto" will provide a direct link to the Apple website. Also, searching for "app store" on the Google website, the first four entries are for Apple.

  29. Thom said,

    March 28, 2011 @ 2:45 pm

    Opps, I meant Dvorak, not Ben!

  30. MikeA said,

    March 28, 2011 @ 3:41 pm

    @Graeme – Far from being a protection from "passing off", registration of trademarks is (was?) often an early variant of "domain squatting". When a company develops a new product in one country, it often has to register the name as a trademark in other markets it may someday wish to expand to, lest someone else do so, and be in a position to "sell" the trademark to the originator. IIRC, in the 1980s this was a particular problem for video games in Spain, Italy, and parts of Latin America and Asia. Sometime accompanied by counterfeits of the products, which were the only items legally labeled with the disputed name.

    It could take a long time to straighten such things out, and a choice between paying the squatter or paying "expediting fees" to the local judiciary.

  31. K said,

    March 28, 2011 @ 5:05 pm

    It's worth noting that the article myl linked to above, the one that mentions Apple suing over Microsoft's GUI, features another trenchant piece of analysis from Dvorak:

    "I think Windows 3.0 will get a lot of attention; people will check it out, and before long they'll all drift back to raw DOS. Once in a while they'll boot Windows for some specific purpose, but many will put it in the closet with the Commodore 64."
    "– John Dvorak, 1990"

    Quite a track record he's got there.

  32. Nathan Myers said,

    March 29, 2011 @ 12:26 am

    I have distinct recollections of MS being explicitly denied a trademark for Word alone, such that they had to settle for "Microsoft Word"™. So, they tried.

    As a historical note, as I understand it, Apple paid for an explicit license from Xerox for the user-interface elements they lifted. In Bill Gates's analogy, Steve Jobs was steamed because he bought the TV from Xerox, but Bill Gates burgled it before he could take delivery. We thus have to take his remarks as ribbing Steve over being so naïve as to have actually paid for such a license.

  33. Keith M Ellis said,

    March 29, 2011 @ 3:21 am

    It's surprising that that "History of the GUI" article has (it claims) quite a few critical reviewers provided corrections because some of the parts I know best (core OS) are (at least) subtly wrong. For example, Win 3.x and prior were not "truly multitasking", assuming that the author intended to mention the distinction between the cooperative multitasking of those version against the preemptive mulitasking of everything after Win95 (on the non-NT branch, I mean). And the writer gets the evolving stability of NT exactly backward–it was much more stable in its original incarnation and became less stable as time went on, most notably with the increased 16-bit app and multimedia and driver compatibility and other consumer-friendly features that (necessarily, for speed and compatibility) were given more direct access to the hardware.

    And the author just hand-waves about his theories of Mac disaffection driving the burst of popularity of alternative OSs such as BeOS, which I find entirely unpersuasive.

    It makes me not very trustful about the credibility of the bits that I don't myself have independent and reliable knowledge about. Not that this really matters and I've just gone on an embarassing computer geek-out.

  34. Michael S said,

    March 29, 2011 @ 6:54 am

    Windsurfer was trademarked in the 1970s.

    Baird's first television in the 1920s was called a Televisor – so perhaps he controlled that name, forcing another to be invented.

  35. Bob Violence said,

    March 29, 2011 @ 5:58 pm

    As a historical note, as I understand it, Apple paid for an explicit license from Xerox for the user-interface elements they lifted.

    Xerox sued Apple around the same time they went after Microsoft — both suits were eventually dismissed — so that seems unlikely to me. One (debatable) factor in Apple's defense is that they were invited to take a look at Xerox's stuff, which I think shows remarkable naivete on Xerox's part.

  36. Bob Violence said,

    March 29, 2011 @ 5:58 pm

    As a historical note, as I understand it, Apple paid for an explicit license from Xerox for the user-interface elements they lifted.

    Xerox sued Apple around the same time they went after Microsoft — both suits were eventually dismissed — so that seems unlikely to me. One (debatable) factor in Apple's defense is that they were invited to take a look at Xerox's stuff, which I think shows remarkable naivete on Xerox's part.

  37. Bob Violence said,

    March 30, 2011 @ 3:20 am

    Addendum to the above: the Apple/Xerox ruling, which says Apple and Xerox had a deal to "[implement] the Smalltalk-80 language and system on a hardware system to be developed by [Apple]"; other sources say Apple gave Xerox some (pre-IPO) stock in as part of the deal. Apple ended up creating products (the Lisa and the Apple) that bore similarities to Xerox's products but had no actual Xerox involvement.

  38. PeterW said,

    March 30, 2011 @ 9:27 pm

    Re: Dvorak and Windows 3.0 –

    Although he was wrong, he was more right than it seems in retrospect – throughout the 90's, there were a lot of programs, including almost all computer games, that required you to boot into dos to use the program or to play the game. Games in particular themselves either wouldn't play under windows, or if they could play under windows, they would play so sluggishly that it was preferable to play under dos.

    (And to get the best performance, you typically needed to make a "boot disk" with certain settings and boot from that disk before playing the game).

    These problems continued through windows 95; I think it wasn't until windows '98 that there was no need to boot into dos to play modern games.

  39. ASG said,

    April 19, 2011 @ 4:04 pm

    Late to the party as always, but it doesn't seem that anyone's yet mentioned a pervasive urban legend about TSR, the company that made Dungeons & Dragons. To this day you'll hear gamers claiming that TSR tried to trademark the term "Nazi." Apparently it was all a misunderstanding about an ill-placed (TM) on an illustration caption (http://en.wikipedia.org/wiki/TSR,_Inc.#Criticism), but, be that as it may, TSR is notorious for its aggressive pursuit of what it perceives as its intellectual property. I do remember early editions of the D&D rulebooks containing long list of copyrighted terms in the front matter, including plenty of uncopyrightable words like "elf" and "orc." Maybe someone who's still got all the books can back me up on this.

  40. M. Radhakrishnan said,

    April 25, 2011 @ 3:37 pm

    Of course, Mars has trademarked the letter M…. I don't think you can get more specific (in terms of shortness) and generic (in terms of general use) than one letter!

  41. Razor Tight | Poetry & Contingency said,

    November 9, 2012 @ 8:00 pm

    […] think about how that phrase got there in the first place. It is a combined case of reanalysis and lexicalization, the process of making words in which a concept, often generated by metaphorical processes, gets […]

  42. Mein App Store, dein App Store › Sprachlog › SciLogs - Wissenschaftsblogs said,

    November 21, 2013 @ 4:03 am

    […] Liberman im Language Log: Lexicalize different?, 27. März […]

RSS feed for comments on this post