The forgotten letter

« previous post | next post »

Probably one of the very worst things about the English writing system (and it has a huge long list of bad things about it) is that it very clearly employs 27 letters in the spelling of words but there is a huge and long-standing conspiracy to market it as having only 26. Insane, but that's what English has done.

The 27 letters are ASCII octal 101 to 132 (the upper-case letters), 141 to 172 (the lower-case letters), and 047 (the forgotten letter).

The punctuation marks are an entirely separate disaster from all of this: the ASCII characters with octal numbers 041, 042, 050, 051, 054, 055, 056, 072, 073, and 077 in some cases have two distinct typographical shapes in print depending on context, one of them has a shape identical to one of the letters, and so on, a terrible mess. There are other characters that appear in written English, but they are unimportant, mostly abbreviations or special spellings for certain words (& for and, % for percent, $ for dollars, and so on), and there are a few other marks and symbols. But the forgotten letter, octal 047 (which causes such a mess for other reasons that it has to have two Unicode numbers, 0027 and 2019), is not a punctuation mark — though it is typographically identical to one of the punctuation marks); nor is it a word abbreviation, or any kind of a textual decoration. It is a letter that you need simply for spelling quite elementary word forms.

One of the worst things about the forgotten letter is that it never stands for a sound in native English words. Indeed, it could be argued that it never appears as a letter within the plain form of any lexeme, and never occurs initially in any word in modern English. But it does appear as the first letter of the two-letter genitive singular suffix of regular nouns; as the second letter of the two-letter genitive plural suffix; as the middle of the three letters that spell the suffix identifying the negative form of auxiliary verbs; as the first letter in the written clitic forms of am, are, had, has, have, is, will, and would; and it has miscellaneous other uses. But though obligatory where it occurs, it never corresponds to any sound in native words. (It sometimes corresponds to the glottal stop in foreign loanwords.)

So don't be too harsh in blaming people who can't spell English. It is tough for them — they will get blamed for their failure, and they will attract the mock ire of Lynne Truss; but it does not mean they have low intelligence. It certainly is not their fault that the alphabet is such a mess that the very number of letters has traditionally been misreported, and is still misreported today.

Share:



72 Comments »

  1. Acilius said,

    September 27, 2010 @ 8:17 am

    'Tis a pity.

    [My estimate of how long it's been since 'tis could really have been said to be a word in English is in the region of 150 years, plus or minus quite a bit. I don't think it could really be counted as a normal word of the contemporary language: just about everyone who uses it is either consciously and deliberately using it as an archaism, or transcribing a Tis!, Tisn't!, Tis!, Tisn't! battle between children. —GKP]

  2. m said,

    September 27, 2010 @ 8:26 am

    Now I know my ABCs next time add apostrophe…

  3. Carl said,

    September 27, 2010 @ 8:44 am

    Pity Hawaiian. The okina looks like ‘ but it's actually a whole different character.

  4. Dierk said,

    September 27, 2010 @ 8:45 am

    Ancient Greek spiritus asper and spiritus leni come to mind.

  5. Alon Lischinsky said,

    September 27, 2010 @ 8:56 am

    Indeed, it could be argued that it never appears as a letter within the plain form of any lexeme, and never occurs initially in any word in modern English

    That's just 'cause you're only thinking of standard modern English. I counted 'bout half a dozen off the top of my head, but I'm just holding out 'til I get somewhere where I can make a decent corpus check. 'Nuff said.

    (I realise I'm just rehashing Acilius' argument, but then none of the forms I mentioned are actually obsolete. And some are likely to have very clear collocational patterns distinguishing them from the full orthographic forms. I'd bet a dollar against a dime for "'nuff", at least.)

  6. Alon Lischinsky said,

    September 27, 2010 @ 9:02 am

    Addendum: I've just discovered that COCA/COHA/BNC have the nasty habit of normalising text so that word-initial apostrophes get linked to the previous word, suggesting that Mark Davies' team shared GKP's belief that they "never [occur] initially in any word in modern English". So one has to look for tude rather than 'tude and be wary of possibly homographs.

  7. Robert Furber said,

    September 27, 2010 @ 9:19 am

    @Alon Lischinsky
    Some people consider 'til to be a mispelling because till was the original word, and until is a more recent form of it.

    [I think 'til is indeed a mistake, albeit a mistake made by people who are carefully trying to avoid a mistake. —GKP]

  8. Moacir said,

    September 27, 2010 @ 9:21 am

    Sadly, this pleasant piece can't count as an apostrophe to the apostrophe.

  9. Alon Lischinsky said,

    September 27, 2010 @ 9:41 am

    @Robert Furber: I am aware of the fact, but (a) it doesn't invalidate my premise, as it doesn't apply to any other word I mentioned; (b) prescriptivism aside, the form is in active use, as attested by 1,274 instances in COCA (although it pales in comparison with till's 11,156 or until's 176,613).

  10. Pflaumbaum said,

    September 27, 2010 @ 9:44 am

    @ Prof. Pullum -

    "just about everyone who uses it is either consciously and deliberately using it as an archaism, or transcribing a Tis!, Tisn't!, Tis!, Tisn't! battle between children."

    - Is it not still used in Ireland, so?

  11. Telofy said,

    September 27, 2010 @ 9:46 am

    Amn’t I lucky to be able to quote Robert Bringhurst whenever I feel the need to contrast the blandness of dumb quotes (and 0x27 is just a single dumb quote) with the elegance of quotation marks or the apostrophe?

    dumb quotes These are refugees from the typewriter keyboard. … Yet the dumb quotes are still there, taking space on the font. They have no typographic function. See also double prime, quotation marks[,] and prime.”—Robert Bringhurst, The Elements of Typographic Style

    For me as a programmer, dumb quotes, of course, have the advantage that I can use quotes inside my strings without having to escape them.

    My Linux allows me (by default) to just press AltGr+B and AltGr+N, respectively, for the quotation marks and AltGr+shift+N for the apostrophe or right single quotation mark.

    See also: Wikipedia: Apostrophe, Computing and Wikipedia: Quotation mark, Typing quotation marks on a computer keyboard

  12. Xmun said,

    September 27, 2010 @ 10:04 am

    Re: "a Tis!, Tisn't!, Tis!, Tisn't! battle between children"

    I have heard an eye-witness description of exactly the same battle between the Cambridge scholars F. R. and Q. D. Leavis, and though it was of course oral, I'm sure they would both have written the repeated word as 'Tis.

    I've just found this line of verse in Wendy Cope's poem "Mr Strugnell", which is included in her volume Making Cocoa for Kingsley Amis (Faber, 1986):

    "He had a funny turn in 'sixty-three"

    (meaning, of course, 1963).

    [I grant you, there are people who write 'sixty-three for 1963, or the 'sixties for the 1960s. Go figure. —GKP]

  13. Josh said,

    September 27, 2010 @ 10:15 am

    Wouldn't this logic make the . . . ellipsis—a word?

  14. Matt McIrvin said,

    September 27, 2010 @ 10:22 am

    Many touch-screen devices with virtual keyboards even relegate this letter to a "symbols and punctuation" mode. Auto-complete will supply it in some cases, but it often can't.

  15. Theodore said,

    September 27, 2010 @ 10:50 am

    Filed by Geoffrey K. Pullum:

    it never stands for a sound

    Maybe not a phoneme, but it certainly has prosodic impact.

    Dierk said,

    Ancient Greek spiritus asper and spiritus leni come to mind.

    Didn't the Russian yer before 1918 reforms also fit that category?

    Pflaumbaum said,

    Is ['Tis] not still used in Ireland, so?

    I think that's a question for Frank McCourt :)

  16. Nicholas Lawrence said,

    September 27, 2010 @ 10:51 am

    28, I think. That 'I' is clearly an additional forgotten letter. There is no reason for regarding it as an upper-case i, because it would be mad to use an upper-case letter here, wouldn't it? Don't know the real history, but I've seen it alleged that it was a scribal variant of i, because with spaces each side of it a dot over it was unnecessary.

  17. RT said,

    September 27, 2010 @ 10:52 am

    The Netherlands is a hotbed of apostrophe errors. In Dutch orthography, some vowels in closed syllables are pronounced differently than vowels in open syllables, so {'s} means pronounce this vowel as if it were in an open syllable, but actually there's an [s] after it (e.g., {de paraplu's} umbrella-pl, not umbrella-genitive). Add to that a (somewhat rare) genetive s in Dutch which I think behaves orthographically just like the plural s… Needles too say, Dutch people misplace the apostrophe nearly constantly, and who can blame them.

  18. Telofy said,

    September 27, 2010 @ 10:56 am

    By the way: “Most digital fonts now include, among other things, a prefabricated ellipsis [I use AltGr+. to get …] (a row of three baseline dots). Many typographers nevertheless prefer to make their own. Some prefer to set the three dots flush … with a normal word space before or after. Others prefer . . . to add thin spaces between the dots. Thick spaces (M/3) are prescribed by the Chicago Manual of Style, but these are another Victorian eccentricity. In most contexts, the Chicago ellipsis is much too wide.”—Robert Bringhurst, The Elements of Typographic Style, § 5.2.7 Use ellipses that fit the font. (emphasis in original)

    Most of the time, I prefer the one with thin spaces. In LaTeX I can just call \ldots{}. Everywhere else I use the “prefabricated ellipsis” (U+2026) instead, because thin spaces (U+2009), unfortunately, allow line breaks. I’d like to use U+202F, of course, but its width is very inconsistent between typefaces.

  19. Theodore said,

    September 27, 2010 @ 11:16 am

    Nicholas Lawrence said,

    'I' is clearly an additional forgotten letter.

    I think it it is a logograph like &, %, @.

    Telofy said,

    By the way: “Most digital fonts now include, among other things, a prefabricated ellipsis

    Some word processors automatically substitute this when three consecutive periods (dots/full stops) are typed. Too bad the software doesn't have the additional intelligence to correct the "commas of ellipsis" when they are typed.

  20. Eli said,

    September 27, 2010 @ 11:31 am

    Maybe people would pay more attention to ' if we started writing it ʔ ? There may not be a lot of (or any) underlying forms in English that include it, but anyone who's heard a bad Cockney accent knows there are surface forms that include it.

  21. Yuval Marton said,

    September 27, 2010 @ 11:41 am

    Re: "But though obligatory where it occurs"…

    if u txted more im sure ud notice its not as obligatory ;-)

    If the apostrophe does deserve promotion to a letter status, I would like to ask the Revolutionary Language Log Letter Committee to design an uppercase form for it, too.

  22. John Cowan said,

    September 27, 2010 @ 11:44 am

    Octal?! Zounds and likewise gadzooks! How archaic.

    But to clarify: U+0027 is the neutral single quote or apostrophe; U+2019 is the closing single quote (in Anglo-American typography, anyhow) or typographical apostrophe, and U+02BC is the actual letter, which is not used in English, but is used in IPA for ejectives and in various other orthographies (but not English nor Hawai'ian) for glottal stops and sometimes other things. Hawai'ian uses U+02BB, which is the letter that looks like an opening single quote (which is U+0218).

    English apostrophes are typographical marks, not letters, precisely because they represent something other than consonants, vowels, tones, syllables, or morphemes. They mark historical elisions, invariably so in certain words, idiosyncratically in others.

  23. Henning Makholm said,

    September 27, 2010 @ 11:47 am

    @Eli: So you want to use a different glyph than the one GKP is talking about, to denote a different function than the one GKP is talking about? In which sense are you then even speaking about the same thing?

    (At least, I assume you're speaking about the Cockney glottal stop, and I don't think there are glottal stops involved in any of the places standard English orthography uses the apostrophe).

  24. Rob Shearer said,

    September 27, 2010 @ 12:00 pm

    I'd never heard this distinction between letters and punctuation before, but doesn't it also imply that the hyphen is sometimes used as a letter? "Ex-wife" and "all-inclusive" are surely words in their own rights and not phrases; most versions of English include at least a few words that must be spelled using the hyphen.

  25. Kylopod said,

    September 27, 2010 @ 12:03 pm

    >and never occurs initially in any word in modern English

    That's so '90s.

  26. mollymooly said,

    September 27, 2010 @ 12:17 pm

    'tis, 'twasn't, 'twouldnt've, etc are still alive and well in Ireland. In quoted speech you won't see them unless the rusticity of the speaker is being emphasised; just as only in an uneducated person's speech is he represented by 'e even though everybody usually drops the H .

    Alongside 'il for till, in America 'round exists as a hypercorrection of round, in the more-or-less mistaken belief that it is short for around. ('Round Midnight, "shot heard 'round the world", etc.) In the UK, where preposition/adverb round is less rare, this mistake is far less common.

  27. J. W. Brewer said,

    September 27, 2010 @ 12:29 pm

    It's perhaps hard to find words that obligatorily versus merely most of the time are spelled with a hyphen in the middle, but that would seem to be an instance of the same phenomenon. And in AmE (as opposed to BrE) you can't properly spell "Mrs." without the final "." (A significant example because it can't be dismissed as an abbreviation insofar as there's no way to "spell it out in full," sort of the way it's hard to call "ain't" a contraction because you can't cleanly uncontract it.)

  28. blahedo said,

    September 27, 2010 @ 12:43 pm

    As I was reading the original post, my first thought was he was talking about spaces (and thus noting the fact that some compound words are written with a space in them), but then I rejected that; then I thought of both the hyphen and the apostrophe and wasn't sure which was meant until fairly late in the post.

    It's now quite clear that he means the apostrophe, of course, but I'm trying to formulate a reason why that might be considered a rightful English letter but hyphens are not. Possibly also forward-slashes might fit this category, though the case isn't as strong. If an apostrophe is a "letter" because some standard words can't be spelled without it, wouldn't hyphens be too? Geoff has a hyphenated word right in his first sentence, indeed before he even hit one with an apostrophe (this is admittedly a coincidence convenient for my argument).

  29. Dan T. said,

    September 27, 2010 @ 12:46 pm

    Some "geeks" like myself tend to have "gut feelings" in favor of using plain-ASCII straight quotes instead of the curly variety due to knowledge of the many sticky issues of character encoding that the non-straight variety cause; for instance, any document that is supposedly encoded in ISO-8859-1 can't legitimately include curly quotes because they are not part of that encoding, but many pieces of software will merrily embed control characters that represent those characters in proprietary encodings anyway. Sometimes you end up with question marks, superscripted numerals, accented letters, or ligatures where quotes and apostrophes belong due to mis-rendering of such characters. ASCII is safer, even if typographically bland.

  30. Abi said,

    September 27, 2010 @ 1:07 pm

    My estimate of how long it's been since 'tis could really have been said to be a word in English is in the region of 150 years, plus or minus quite a bit. I don't think it could really be counted as a normal word of the contemporary language: just about everyone who uses it is either consciously and deliberately using it as an archaism

    It's making a minor comeback in the world of instant messaging, or at least appeared to be in my mid to late teens, when 'tis was convenient shorthand/textspeak. Probably me and my now-former schoolmates have been superseded by younger and cooler kids having some even shorter version, but that's what we wrote then.

  31. Kai Samuelsen said,

    September 27, 2010 @ 1:37 pm

    What, precisely, is the reason to consider ' a letter, but no other punctuation marks? This is the real weakness of this argument.

  32. Ellen said,

    September 27, 2010 @ 1:41 pm

    But, Abi, 'tis, if you use the apostrophe, has the exact same number of characters as it's, and even the exact same ones, just in a different order. So how does 'tis, offer any advantage?

    (Using italics instead of quotes so as to avoid the apostrophe next to quote marks.)

  33. groki said,

    September 27, 2010 @ 1:58 pm

    [I think 'til is indeed a mistake, albeit a mistake made by people who are carefully trying to avoid a mistake. —GKP]

    'Til, that is, enough of us make the mistake: then it's just usage. ;)

    @Ellen: when 'tis replaces [yes,] it is, maybe?

  34. Ellen said,

    September 27, 2010 @ 1:59 pm

    Ah, okay, thanks.

  35. L said,

    September 27, 2010 @ 2:06 pm

    I get the argument of why the apostrophe is a letter, rather than a punctuation mark. And I think I understand why most other punctuation marks don't qualify as letters. But it does seem that, as blahedo, argues above, the hyphen (not to be confused with dashes, of course) must be considered a letter for the same reasons.

    Also, given the logic, it seems wrong to say that there are 27 letters (or 28 if we bring in the hyphen). Aren't there really 53 (or 54)?

  36. L said,

    September 27, 2010 @ 2:07 pm

    Embarrassing extra comma above, please ignore.

  37. Spell Me Jeff said,

    September 27, 2010 @ 2:22 pm

    @Kai

    Punctuation delimits sentences and other word groups into syntactic units. It has no meaning itself. It purpose is to make plain the meaning of the content it interacts with.

    ', when used as a missing letter, actually substitutes for content. In this sense it can be said to have meaning.

  38. Abi said,

    September 27, 2010 @ 3:43 pm

    Ellen: Of course it isn't any shorter, nevertheless it's what we wrote, and we (then) called it either shorthand or textspeak. Lots of the textspeak we used wasn't really shorter, just trendier (much of it is shorter, of course).

    (It's also only fair to note that apostrophe use varied considerably.)

  39. Matthew Kehrt said,

    September 27, 2010 @ 3:54 pm

    @Yuval Marton Everyone knows the capital form of ' is ". After all, it's on the same key on the keyboard!

  40. Kai Samuelsen said,

    September 27, 2010 @ 3:59 pm

    @Spell Me Jeff

    But the apostrophe as used in the way you outline also has "no meaning itself" – its meaning is established by context, just as all other punctuation.

  41. ignoramus said,

    September 27, 2010 @ 4:13 pm

    There I be a thinking " ' " be an excuse to drop a sound as 'tis the poets that luved to tell readers that they knew 'ow it be spel' but needed to get the rhythm to be smooth'.
    I could be advised but being just a clod 'opper letting the elite know, we can spel like the dictionary when we want but we rit wot we say.
    'tis the way we wish teh sound to our peers ain't that so, 'eatin ' , 'untin', physhin', 'hootin', all but the 'houtin'.

  42. Sili said,

    September 27, 2010 @ 4:13 pm

    I don't think ['tis] could really be counted as a normal word of the contemporary language

    'strewth!

  43. Richard Sabey said,

    September 27, 2010 @ 4:51 pm

    @Kai Samuelsen: There is no real reason. A while ago, GKP spun the line that some adverbs are prepositions. Now he's spinning the line that the apostrophe (in the English writing system) is a letter. So I take this blog entry in the humorous spirit that several other commenters already have.

    @blahedo: Yes, you're right; the hyphen is just as much a letter as the apostrophe is. It's needed to spell words such as twenty-one, great-grandson, Sarah-Jane, etc., and to distinguish pre-date from predate. Slash, too. "and/or" is obligatorily spelt with a "/".

    @GKP You missed a trick in not declaring upper-case letters to be different letters from lower-case ones. That adds 26 to the total. "Polish" is obligatorily spelt with a P; "polish" may be spelt with a p, which proves that it is spelt differently and that P is a different letter from p.

    You call "&" a special spelling for the word "and", and as such "&" is a logograph. However, in e.g. AT&T, the "&" is an essential part of the spelling. ATandT is not a correct spelling of AT&T. So this "&" is a letter.

    You write that "One of the worst things about the forgotten letter is that it never stands for a sound in native English words." Not the way I speak it! It stands for the KIT vowel in words such as "church's" and "class's".

  44. Dan T. said,

    September 27, 2010 @ 5:15 pm

    The period is needed to spell "Mrs." or "Ms."

  45. Juergen Lorenz said,

    September 27, 2010 @ 5:35 pm

    It's those missing 26 letters that have made my life difficult signing in to some web sites. I know my password, don't I? Have written it hundreds of times. AH, THE CAPS LOCK WAS ON. Add 26 letters.

  46. Dougal Stanton said,

    September 27, 2010 @ 5:58 pm

    @Rob Shearer

    While it's true there may be chickens living in co-ops I'm sure they'd much rather they lived in coops :-)

  47. John said,

    September 27, 2010 @ 7:30 pm

    And here I thought we were going to bring back ð and Þ.

  48. Chris Travers said,

    September 27, 2010 @ 7:54 pm

    I suppose it never stands for a sound in the official grapholect of English. However there are plenty of times it is used to represent a glottal stop in fiction, theater, etc.

  49. Xmun said,

    September 27, 2010 @ 8:17 pm

    Talking of forgotten letters, why is "fo'c'sle" so spelt? Why not "fo'c's'le"? After all it's a shortening of "forecastle". The t is silent in speech but it's there in the spelling, and its omission ought to be noted just like the omission of the other letters.

  50. Hawke said,

    September 27, 2010 @ 8:24 pm

    Meh. "&" is just a ligature of "Et" (Latin for and). ' is just a typewriter apostrophe, the bastard offspring of single-quote and a true apostrophe.

  51. chris said,

    September 27, 2010 @ 9:30 pm

    There is one difference between the apostrophe and the hyphen or slash, namely the tendency for people to ignore its status as a letter is a lot more annoying!

    I immediately thought of the apostrophe upon reading the first sentence of this post, because just last night, in order to comply with a requirement imposed on me by my bank, I was registering my credit card for the "Verified by Visa" service. This registration process requires you to enter a "Personal Message" consisting of about 15-20 characters. But even though it says you can use any letters or numbers, you can't just write any words you want, because the apostrophe is not among the letters you can use. Neither is the hyphen or the slash, but the likelihood of you wanting to use them is vastly less. Eliminating the apostrophe, on the other hand, means that your "personal" message is possibly going to end up rather formal and impersonal. Or poorly spelled.

  52. J.H. said,

    September 28, 2010 @ 5:24 am

    As an apostrophe-related aside, my English teacher last year spelled "plane" (as in airplane) with an apostrophe at the beginning. Same with "'phone", I think. Now that's just a little pretentious…

  53. Lugubert said,

    September 28, 2010 @ 5:45 am

    @ Matthew Kehrt
    "After all, it's on the same key on the keyboard!"

    On your English keyboard. Circumstantial evidence that neither is a Swedish letter, is the 23 cm distance between them on mine.

  54. Ginger Yellow said,

    September 28, 2010 @ 5:53 am

    Many touch-screen devices with virtual keyboards even relegate this letter to a "symbols and punctuation" mode. Auto-complete will supply it in some cases, but it often can't.

    The most baffling instance of this is in (Nokia's) T9 system, where if you type /I'/ followed by the /mno/ key, the suggested auto complete is always the nonsensical /I'o/. Why? /I'm/ must one of the commonest words typed into a phone, it has no conflicting T9 interpretations, and yet it's entirely beyond the wit of Nokia's engineers to design a predictive text system that defaults to it.

  55. SeanH said,

    September 28, 2010 @ 7:27 am

    As I said to my captain immediately before my court martial for mutiny, I sha'n't go to the fo'c's'le.

  56. David said,

    September 28, 2010 @ 9:03 am

    It used to be that Swedish had 29 letters but was marketed as only having 28. The 29th letter was w, treated as a variant form of v (so in dictionaries and encyclopedias words beginning with v and w would be mixed together). Thankfully, from the 2006 publication of the 13th edition of the Swedish Academy Glossary (Svenska Akademiens Ordlista, SAOL) and onwards, it seems like this is no longer the case. An increasing amount of English loanwords beginning with made a separation necessary. (The three extra letters, BTW, are , and .)

    (There is no room for conspiracy accusations concerning the use of the apostrophe, however, as this is employed very rarely, and its use is probably declining. The informal versions of words like "någon" (someone) and "staden" (the town) are now much more likely to be written "nån" and "stan" than "nå'n" or "sta'n", but this would not have been the case 100 years ago.)

  57. L said,

    September 28, 2010 @ 9:35 am

    I always thought the word was pronounced fo'c'sle but spelled forecastle. My dictionary gives "(also fo'c's'le)," which indicates to me that forecastle is the preferred spelling. For what it's worth.

  58. Tom O'Brien said,

    September 28, 2010 @ 10:49 am

    Having an apostrophe in my name (which is really a fada (long) mark that fell off the first Irish word Ó into the former u0020 (word space) that must at one time have separated it from the second word Bríain), I am all too familiar with the troubles generated by this character's second-class status, or rather by the idiocy of software designers who choose to ignore or abuse it. There are many programs and sites I simply can't use without mutilating my name; and in some cases I can't use them at all. I've had credit cards fail to work because of it, and sites have accused me of attempting an SQL injection attack. So let's give it first-class status.

    (While we're at it, let's add the schwa to the alphabet as well. It seems to be the most frequently used vowel in the language, but we have no character for it.)

  59. Army1987 said,

    September 28, 2010 @ 12:04 pm

    English apostrophes are typographical marks, not letters, precisely because they represent something other than consonants, vowels, tones, syllables, or morphemes.
    The same applies to aitches in Italian, but I know no-one who would consider them to be punctuation.

  60. J.W. Brewer said,

    September 28, 2010 @ 12:46 pm

    Possible counterargument I was reminded of by the romanization-of-Chinese thread going on up the block: one of the reasons Wade-Giles never got sufficiently firmly established to resist displacement in the Anglophone world by the Communist-backed hanyu pinyin is that apostrophes are ubiquitous and crucial in Wade-Giles (ok, maybe it's something called something else that looks like an apostophe) but were/are commonly omitted in practice by users not working for scholarly presses, thus leading to massive ambiguity between distinct consonants (wikipedia tells me aspirated/unaspirated rather than voiced/unvoiced). If Anglophones had been more comfortable with their alleged 27th letter (and thus able to generalize it beyond the specific limited set of uses it already had the same way they can hopefully get used to, e.g., Q and X being used in pinyin in ways they aren't used in English), maybe this wouldn't have been so much of a problem?

  61. Xmun said,

    September 28, 2010 @ 1:57 pm

    @ Tom O'Brien: "While we're at it, let's add the schwa to the alphabet as well. It seems to be the most frequently used vowel in the language, but we have no character for it."

    No, the five or six characters we have already — a e i o u y, plus j and maybe a few others in some contexts — are quite enough for the thirty-odd vowels and diphthongs we have in English, including the schwa. Indeed, what is represented as a schwa in IPA is often, if you listen carefully, recognizable as a reduced form of one of the other vowels.

    (I remember hearing an open e in Mr Attlee's pronunciation of the word "problem". Yes, I was a kid at the time.)

  62. richard howland-bolton said,

    September 28, 2010 @ 2:10 pm

    @ Tom O': Same problem with the hyphen

  63. unekdoud said,

    September 28, 2010 @ 3:02 pm

    Some meaning is sometimes assigned to the usage of this forgotten letter. I was told, and it is quite possible, that the frequency of apostrophe usage is related to formal vs informal writing. In this sense we can say it is special, since the frequencies of other letters don't really affect the tone of the text, with the exception of alliteration and similar wordplay.

    (There are 3 apostrophes in the post, but no z.)

  64. John Cowan said,

    September 28, 2010 @ 4:03 pm

    Wuddaya mean, no z? It's right there at the end — well, it won't be the end after I've posted this.

  65. Jason L. said,

    September 28, 2010 @ 11:56 pm

    One could make the case that the apostrophe has the value of a schwi when occurring between sibilants, as in Buzz's. Surely James's would be pronounced differently without the apostrophe.

    On the other hand, it could just be that the genitive and contraction-of-"has"-or-"is" 's is phonologically identical to the plural s — it's just the orthography that's different. Still, the issue arises of why it's necessary to add es for the plural after sibilants to avoid something like fishs, which would map phonetically onto a disallowed /ʃs/ cluster, but it's not necessary to add 'es for the genitive.

  66. Boudica said,

    September 29, 2010 @ 7:57 am

    If we're going to add letters, I vote for symbols to replace 'ch' 'sh' and the two variants of 'th' (thorn and eth?).

  67. Sandra Wilde said,

    September 29, 2010 @ 5:24 pm

    I wonder if adding the schwa to the alphabet would drastically reduce the incidence of misspelling since so many adult invented spellings involve using the wrong letter for a schwa sound.

  68. John F said,

    September 30, 2010 @ 5:17 am

    J.H. That reminds me that I used to get sports news on the BBC's Ceefax service. Ceefax could only show twenty-four lines that were forty columns of characters wide, so text had to be very concise. Even so, because this was the BBC, influenza was always shortened to 'flu rather than flu.

  69. Julie said,

    October 2, 2010 @ 7:44 pm

    This is slightly off-topic, but I couldn't find any e-mail address or contact information to just ask a question/suggest a topic for a post.
    In Canadian English spelling, we typically use the spelling "cheque" when referring to a bank cheque, but "check" when referring to a checkmark, for example, or for the verb "to check" and the like. However, I just realized that we spell this idiom as "raincheck," although it seems to me that the morpheme is more likely the bank cheque morpheme than the verification-type one. I thought it was interesting that we stick to the spelling so closely when it appears on its own, but as soon as you put it in a compound word, the American spelling is the norm.
    Any thoughts?

  70. iching said,

    October 4, 2010 @ 2:14 am

    Regarding Julie's OT comment, Australian English also preserves the cheque/check distinction in writing. Online etymological dictionaries (for what they're worth) relate the origins of the term to 19th c. US baseball (meaning "rain cheque" in Canadian and Aus. usage). I would guess that is why the "raincheck" spelling is used in Canada and Australia.

  71. Mark Mandel said,

    October 5, 2010 @ 4:03 pm

    'Twas brillig, and the slithy toves
    did gyre and gimble in the wabe…

  72. Bob Knowles (@BobK99) said,

    January 16, 2012 @ 1:18 pm

    'One of the worst things about the forgotten letter is that it never stands for a sound in native English words.' – Oh yeah? There're many such cases. /'ðeərə/ , for example. (The quoted statement is obviously true of words that don't have an apostrophe – which, by definition, rules out any useful interpretation of the word 'word') ;)

    b

RSS feed for comments on this post

Leave a Comment