LOOSE DECIMATION.

John McIntyre’s latest column demolishes a reader who insists that decimate must be limited to reduce by a tenth”:

There you have the etymological fallacy at its finest. Because CanSpeccy can cite a handful of examples of the limited use of the word, the loose or rhetorical sense of “to subject to severe loss,” for which the OED has examples dating back to 1663, must be ignorant and anyone who uses it should get fifty of the best.
The giveaway is CanSpeccy’s sneer at “misuse on the basis of common usage.” Common usage is what determines language, which is an arbitrary set of sounds or letters that means only what the users agree they mean. That is why in English we can use decimate in a way that Caesar did not,* why we can appropriate words from Scandinavian languages and alter their pronunciation to suit ourselves. Common usage is why we no longer speak Anglo-Saxon or Norman French.

But that asterisk after “Caesar did not” goes to an interesting question: “Not a Latinist myself. Are we sure, really sure, that no classical writer of Latin ever used decimare in a loose or rhetorical sense?” Jan Freeman wrote me to suggest that “this seems like a question for your erudite band of readers,” and I agree, so: anybody know?

Comments

  1. John McIntyre says:

    I’ve corrected it to “decimare” rather than “decimate” for Caesar.

  2. Lewis and Short give 3 meanings, per Perseus:
    1. punish by 3eliminting every tenth man
    2. pay tithes
    3. select (the best) tenth as an offering.
    with a note that the last meaning led to the participle decimatus acquiring the meaning selected, excellent, choice

  3. Lewis and Short is thoroughly outmoded; the Oxford Latin Dictionary only has:
    1 (mil.) To punish every tenth man chosen by lot.
    2 To make an offering of a tithe to a god.
    There is no entry for a “decimatus.”

  4. So it really doesn’t mean giving dimes to panhandlers?

  5. The issue for me has never been that whatever is lost or destroyed must be precisely one tenth of the total, but that 1 in 10 is not, in the grand scheme of things, that much destruction. The leap from that to “total obliteration,” as the term is now often used, seems a little wide. To me this is sort of like taking a term like “pedicure” and giving it a “loose, rhetorical” meaning of “foot amputation.”

  6. By the way, I noticed in the first paragraph of the McIntyre piece that he refers to Finland as a Scandinavian country. Would that be a loose, rhetorical version of Scandinavia to which Finland belongs?

  7. Trond Engen says:

    My century-old Latinsk ordbok doesn’t know decimatus either.
    I’ll note that cimare may simply mean “increase rapidly in numbers”, as attested in cimex.

  8. marie-lucie says:

    decimare, etc:
    I wonder if the extension of meaning from “eliminating 10%” to “eliminating almost every one” (which is how I understand the modern meaning) might not originate from the fact that all the potential eliminees (?) were subject to this random procedure and would therefore all be afraid of it: having escaped it once did not mean that one would always escape it, on the contrary, the soldier having escaped while his friend was killed would feel more rather than less likely to be killed “next time” (even if the risk of death was always inherent in combat, awareness of his personal prowess gave the soldier a hope of surviving, but personal prowess was not a factor in random “decimation”) . Fear of the danger would therefore increase the perception of the likelihood of danger to all vs. the actual percentage of those affected by the procedure.

  9. Trond Engen says:

    My probably lackluster understanding is that Finns accept being lumped with us Scandinavians by outsiders like the English or French but not by fellow Nordics. The result is that ‘Scandinavia’ is a false friend.

  10. Scandinavian could refer to the Scandinavian Peninsula, including Finland and excluding Denmark. It’s not really clear when it’s being used as a geographic term or as a cultural one.

  11. Outmoded as Lewis and Short is, and the OLD is coming out in a second edition, too, …: Lewis and Short cite Livy as the earliest user of decimare; I imagine so does the OLD, unless it cites inscriptional material (I’m away frommy books at the moment). The Latin Vulgate uses the word in the meaning ‘to tithe’.

  12. mollymooly says:

    Finland has been more-or-less extricated from the definition of “Scandinavia”, whereas Ireland has not convinced the world to exclude it from the “British Isles”.

  13. The fact that Lewis and Short is “thoroughly outmoded” has nothing to do with the omission of decimatus (also spelled decumatus) from the OLD. Among the latter’s numerous faults (such as omitting hidden quantities), it arbitrarily cuts off at 200 A.D., thus omitting the last couple of centuries of pagan Latin along with nearly all Christian Latin. L&S attributes decumatus to Symmachus, whose dates are ca345-402, so OLD naturally omits him, along with his friend Macrobius (who makes him and the Vergilian commentator Servius characters in his Saturnalia), Claudian, and other important authors. Though the text of Symmachus’ letters is not on-line, I see no reason to doubt that he calls a young man decumatissimus in one of his letters.
    L&S and OLD agree that the verb decimare is post-Augustan (Frontinus, Tacitus, Suetonius), though L&S helpfully mentions that the practice is attested earlier, in Book II of Livy, writing in the late 1st century B.C. about an event that supposedly took place in 283 B.C. If I could remember where I put my Livy books when I moved them to make room for Martial, I’d tell you what the commentators say about decimation.

  14. marie-lucie says:

    Trond: I’ll note that cimare may simply mean “increase rapidly in numbers”, as attested in cimex.
    Actually, Latin decimare and other forms could be from decem ‘ten’ + verbal suffixes, or possibly from cimare preceded by the negative prefix de-.
    There is French word la cime meaning ‘highest point’ of a tree or mountain (from Lat cima, itself from Greek cyma), and a verb écimer‘to cut off the top branches of’ a tree, presumably from Latin ex-cimare. The TLFI cites a derivative le cimeau meaning ‘large branch stripped of its leaves’ and often set up in the ground as a kind of fake tree in order to attract birds which will perch on the bare branches and be more easily shot by hunters.
    The idea of cutting off the top of a large plant reminds me of the anecdote from ancient times about a king, tyrant or similar ruler sending a messenger to another ruler asking him for advice about how to deal with a rebellion: the second ruler did not bring up the topic with the messenger but took him to a garden and proceeded to cut off the tallest flowers with his sword.
    It is possible that decimare originally referred to lopping off top branches (and by extension, “killing the leaders” of a rebellious group), but that the resemblance with decem ‘ten’ led to a folk etymology whereby army commanders and other persons unfamiliar with rural activities such as tree care or bird hunting understood the meaning as “killing one tenth” of a disobedient group.
    Of course I could be totally wrong about cima vs decem“.

  15. marie-lucie says:

    (Sorry about the extra italics – I forgot to preview).

  16. Interestingly, ScandinavianRail includes Finland.
    “Travel throughout Norway, Sweden, Denmark and Finland in comfort with ScandinavianRail. Our selection of rail passes, train tickets and scenic rail tours offer great value for exploring these exciting destinations to your heart’s content!”

  17. Eurail Scandinavia Pass: Experience Nordic culture with unlimited rail travel in Denmark, Finland, Norway and Sweden.

  18. dearieme says:

    ‘Ireland has not convinced the world to exclude it from the “British Isles”.’ I have been known to refer to Ireland as “the biggest of the Hebrides”.
    Anyway, Ireland is part of the British Isles, as a geographical expression, and, as a geographical expression, the Channel Islands are not. Similarly, as a geographical expression the British Isles are part of Europe and Iceland is not. Nor Cyprus.
    As for where Europe ends to the east, Lord knows; Asia is just an extension of Europe, really.

  19. Trond Engen says:

    marie-lucie: It is possible that decimare originally referred to lopping off top branches
    I like that!
    (Here I think I’m being funny and it turns out I was all serious. Maybe if I start adding smileys it’ll be easier to figure myself out.)

  20. marie-lucie says:

    The Channel Islands cannot be part of the British Isles, since they are much closer to France (just West of the Cotentin peninsula). They were part of Normandy in the Middle Ages, but when Normandy ceased being an English possession and reverted to the French crown (when the kings of England stopped being also dukes of Normandy), the treaty exempted those little islands, which remained British possessions. At least in two of them (Jersey and Guernsey), Norman French dialects are still spoken by part of the population. (Our fellow Hatter Geraint Jennings has told us about them sometimes).

  21. marie-lucie says:

    MHendry: decimatus (also spelled decumatus)
    I seem to recall seeing the form decumatus somewhere, without thinking about the possible etymology. But if Latin cima ‘top, summit’ is an adaptation from Greek cyma (sorry for using the Latin letters),cuma is also plausible as an alternate Latin form for the borrowing. If so, then a hypothetical *de-cumare “to lop” would be more likely than a hypothetical *decem-are as the source for the attested decimare.

  22. The issue for me has never been that whatever is lost or destroyed must be precisely one tenth of the total, but that 1 in 10 is not, in the grand scheme of things, that much destruction. The leap from that to “total obliteration,” as the term is now often used, seems a little wide. To me this is sort of like taking a term like “pedicure” and giving it a “loose, rhetorical” meaning of “foot amputation.”
    While my basic point would be that it’s irrelevant how great a semantic leap a word has made, it’s used however it’s used and that’s its meaning, like it or not (and this leap is considerably less than that bead made from ‘prayer’ to ‘little round object’), if it will make you feel better I will point out that killing one in ten of a group of people is in fact “that much destruction” if you happen to be in the group or have a loved one in the group, and it’s pretty easy to see how “they’re decimating the group” would tend to imply very bad things happening. The exact percentage is really pretty irrelevant (we humans are notoriously bad at statistics).

  23. Funnily enough, I usually think of ‘decimate’ as meaning ‘reduce to one tenth of its original size’. Perhaps not accurate historically, but it’s a kind of intuitive ‘etymology’ that fits the current meaning.

  24. Or, put another way, there’s no reason other than classical tradition that decimate has to mean “kill one in 10″ rather than “leave (give or take) one in 10 alive”.

  25. marie-lucie says:

    Since the Classical Latin word refers to deliberately killing a percentage of one’s own troops in order to keep the rest of them in line, “kill one in ten” makes more sense than “leave only one in ten alive” which would very seriously deplete the army. Such major depletion could occur as a result of a crushing defeat in combat, but a competent commander could hardly wish to bring about this result deliberately.

  26. That’s very true, but since army hasn’t kept up that practice, there’s been no need for a need for a word of that particular definition (outside of the classics) for fifteen hundred years or so, so it’s not a disaster if the meaning slips a little. And anyone reading the disciplinary reports of the roman legions is probably smart enough to keep the right definition in mind.

  27. marie-lucie says:

    Probably because of professional habits, as a historical linguist I tend to be more interested in etymology (going back to word origins) than in later, attested meaning shifts. I am not arguing about what the meaning of the word “should be” in current English, but I am not always sure whether a commenter is talking about the ancient or the modern meaning.

  28. Is the general thrust of the question and this discussion to show that the original argument is not actually an “etymological fallacy at its finest” by producing an etymological fallacy that is finer yet? That is, are we trying to show that it is perfectly okay now to use “decimate” loosely because it was used loosely even in ancient times?

  29. I suspect it’s a clever satire of descriptivists by a closet prescriptivist: “Just find one person who ever used the word this way before, and that proves that it must be OK!”

  30. There’s a difference between “punish by eliminating every tenth man” (L&S) and “punish every tenth man chosen by lot” (OLD). In L&S, it is the _unit_ that is being punished, not the individuals who happen to be singled out. In OLD, it is every tenth individual who is being punished, and the rest implicitly are not. L&S probably captures the original meaning better.

  31. @Greg Lee: No, he’s just being a conscientious writer. He wrote, “That is why in English we can use decimate in a way that Caesar did not…”, and then paused and said, wait, that’s a rhetorically effective phrasing, but is it correct? Did the ancients actually not use the loose sense of decimate?

  32. Britain and British as the names for a political entity are newfangled expressions invented by the Romans, of course, whereas they had long been used in one form or another in a geographical sense to refer to Albion, or its people, or the whole archipelago and inhabitants including the territory of what is now the Irish republic. Nonetheless those sensitive to the pains of history often try to avoid the term ‘British Isles’ when Irish citizens are within earshot, preferring euphemisms like ‘these islands’. We could do with inventing a new geographical term.
    In law there is a separate political term ‘British Islands’ which refers to the UK plus Man and the Channel Islands.

  33. Athel Cornish-Bowden says:

    Culturally, Iceland, the Faroes, and maybe even the Shetlands, are Scandinavian: does anyone include them in “Scandinavia”.

  34. Culturally, Iceland, the Faroes, and maybe even the Shetlands, are Scandinavian: does anyone include them in “Scandinavia”.
    Not the Shetlands, but yes, certainly Iceland & I think the Faroes are considered and consider themselves Scandinavian (but not Nordic countries). I think I’ve got that right. Maybe Trond will confirm.
    About the British Isles, I use the term “Irish Isles” for the whole lot as an affirmative, retaliatory action. It hasn’t yet caught on, but people seem to know damn well which bits of land I’m talking about. As for Jersey & Gurnsey etc., I’m planning to start calling them the Manchian Islands, because I prefer the name. Surely it’s time for the Isle of Man to become the Isle of Person.
    Another good name for the Isles Formerly Known as British is “the Western Isles”, and yes, I know that’s currently being used by scotch-whisky distillers but it’s a jolly good, romantic-sounding name. The idea of East Anglia, Eastbourne and East Grinstead being part of the Western Isles would be typically esoteric and confusing.

  35. In case it’s useful, here’s the results from Perseus using their ‘search by lemma’ function.
    Searching for the lemma ‘decimo’ gives 11 results.
    The first 3 can be discounted as they are forms of ‘decimus’ (tenth); the remaining 8 seem to me to be truly from ‘decimare’.
    The last 3 of these 8 come from the New Testament, and appear to be used in the sense of ‘tithe’.
    So we are left with 5: 3 from Suetonius and 2 from Tacitus. A quick glance at these suggest that they all allow the sense ‘to select by lot every tenth man for punishment’ – but I would turn to an expert on these authors to tell if they were being mathematically precise!
    Of course, because Perseus doesn’t cover all authors from all periods, this isn’t the full picture. Not to mention the curse/beauty of our limited evidence for Latin being that we can never be sure that a word wasn’t used with a particular sense anyway…

  36. First, regarding Lewis and Short, that dictionary is most certainly not “thoroughly outmoded.” It, along with the Oxford, remains one of the two standards. (Since dead languages don’t change, a century-old dictionary of a dead language is not at a disadvantage.) The Oxford is certainly more recent, but it fails in one area. It does not include post-classical sources, so it doesn’t cover the whole history of Latin like L&S does.

  37. Trond Engen says:

    AJP Queen de Drag: Maybe Trond will confirm.
    As I understand it, the Icelandic view isn’t very different from the Finnish. Because of the linguistic and genealogical affinities, they may be more content than the Finns with being Scandinavians internationally, but there are some who object. But they are definitely Nordic. The Faroese are even closer to mainland Scandinavia. Traditionally they are quite vocal about feeling closest to Iceland, then Norway, and Denmark only third, but almost all actual contact is with Denmark. And not only politically or economically, since everybody has family ties with Denmark.
    I’m always wrong about Iceland, though.

  38. Is the general thrust of the question and this discussion to show that the original argument is not actually an “etymological fallacy at its finest” by producing an etymological fallacy that is finer yet?
    No, it’s just curiosity.
    First, regarding Lewis and Short, that dictionary is most certainly not “thoroughly outmoded.” … (Since dead languages don’t change, a century-old dictionary of a dead language is not at a disadvantage.)
    That’s a very strange statement. So you think nothing has been learned about either Latin or lexicography since 1879? I’m sure you’re fond of your battered old copy, but come on, that’s just silly. If L&S weren’t “at a disadvantage,” why would Oxford have expended so much time and effort on producing another dictionary?

  39. I like the idea of the Irish Isles. And as it would be possible to argue that there are more Irish on the island of Great Britain than there are British on the island of Ireland it is not too illogical, either.
    But my vote goes to the Western Isles, at least it does if we are allowed to call the two largest isles Western Western Island and Eastern Western Island.

  40. That’s two of us, Picky.

  41. As far as I know, “decimare” derived from “decem”. Another “similar” word is hecatomb, derived from “hekaton” (one hundred).

  42. if it will make you feel better I will point out that killing one in ten of a group of people is in fact “that much destruction” if you happen to be in the group or have a loved one in the group
    There was no need to make me out to be a empathy-deficient sociopath. I’m not dismissing anyone’s death. I do feel that if large numbers of my loved ones were killed, I would prefer the chosen verb to express the extent of that carnage without giving any classicist who might be reading about the carnage reason to think it might have been far less than it was.
    Also, I am increasingly inclined to think that there is nothing particularly rhetorical/figurative about the evolution of the word “decimate.” I think it just sounds a lot like “devastate” and people got confused.

  43. Having said this, décimer is now also used very figuratively in French, where it does not sound at all like dévaster or any other comparable verb, so my theory does have some problems.

  44. Also, I am increasingly inclined to think that there is nothing particularly rhetorical/figurative about the evolution of the word “decimate.” I think it just sounds a lot like “devastate” and people got confused.
    Even if this theory is right, it has no affect on the rightness or wrongness of using decimate in the sense, well, pretty much everyone does. The majority of linguistic change owes to people either being confused or lazy; it’s the rare change that’s conscientious. The fact remains, decimate denotes mass destruction even in written Standard English.

  45. I suspect it’s a clever satire of descriptivists by a closet prescriptivist: “Just find one person who ever used the word this way before, and that proves that it must be OK!”
    Your quote by your supposed clever satirist displays a total misunderstanding of descriptive linguistics. I hope, even if you find such drivel clever, that your own view is more informed.

  46. “Britain and British as the names for a political entity are newfangled expressions invented by the Romans”: from the Greek Pretanike (spelling from memory) which referred to the whole (geographical) British Isles. The Roman word applied for most of the time only to (most of) the current England and Wales and (for some of the time but nobody seems to be sure exactly when) to the current Southern Scotland. A book I read recently seemed to be saying that it’s not clear when Hadrian’s Wall was used as the definition of the frontier and when it was a geographically convenient defend-and-control line located behind the frontier.
    But One Thing Is For Sure: it was ages ago.

  47. There was no need to make me out to be a empathy-deficient sociopath
    Good lord, that was the farthest thing from my intention, and I apologize if it came across that way!

  48. Well, dearieme, it was certainly well behind at least one of the frontiers: the Antonine Wall is far, far to the north.
    It was indeed ages ago. Sorry, the “newfangled” was a feeble attempt at a joke.

  49. when the kings of England stopped being also dukes of Normandy
    They still are!
    The Queen is also the Duke of Normandy, it’s in the list of her official titles – for Jersey and Geurnsey. Not Duchess, but Duke. And there isn’t a Duchy of Normandy in France anymore.

  50. Since Cyprus, who was about to be decimated with a 10% levy on deposits, has been mentioned, can I ask what the new proposal for a 30% levy should be called? Trigintamate? Tricesimate? Tricemate?

  51. Oh, I don’t think the Queen is Duke of Normandy, except in the humorous toast. I think we gave the title away to some French people in the 13th century or thereabouts.

  52. but there’s no other head of state in Îles Anglo-Normandes. They are not part of the UK, not part of France and not even part of the European Union.

  53. “As for where Europe ends to the east, Lord knows; Asia is just an extension of Europe, really.”
    dearieme, Asia begins at the Rhine.

  54. Asia begins at the Rhine.
    It’s a matter of perspective. In Vienna it is commonly held that Asia begins at the Rennweg.

  55. Tricemate?
    and how to pronounce this verb?

  56. I think, Sashura, that they are subject to the Queen in Right of Jersey and the Queen in Right of Guernsey; in other words they are not ducal possessions but Crown Possessions. It’s true that she is often referred to on the islands, semi-jokingly, as Duke of Normandy, but I believe the English Crown lost that title in the Treaty of Paris.

  57. dearieme says:

    “Well, dearieme, it was certainly well behind at least one of the frontiers: the Antonine Wall is far, far to the north.
    It’s also a matter of speculation whether it was intended to define the frontier: it seems perfectly plausible, but that ain’t evidence. Historians of Roman Britain are blessed, or perhaps cursed, with a paucity of written evidence, so they argue by analogy with other bits of the Empire, over-interpreted (maybe) fragments from continental records and from archaeology. ‘least so it seems to me.
    Still, it is widely agreed that they discussed – the Romans, not the historians – invading Ireland and decided not to bother even though, inevitably, they had been invited to invade to take sides in an Irish war. History, eh? One bloody thing after another.

  58. The arms of the Isle of Man turn out to be legs. and why do they put their motto in Latin, of all things? Quocunque Jeceris Stabit – “whithersoever you throw it, it will stand” (“whithersoever” must be Manx, probably one of those -hin things; what’s wrong with “wherever”?).

  59. Come, come: where didn’t equate to whither in them Olden Days when mottoes were mooted, as you are whell awhare.
    And the legs, of course, represent the traditional three-legged Manx cat.

  60. David Marjanović says:

    Funnily enough, I usually think of ‘decimate’ as meaning ‘reduce to one tenth of its original size’. Perhaps not accurate historically, but it’s a kind of intuitive ‘etymology’ that fits the current meaning.

    That seems to be what the German usage has in mind. Interestingly, our prescriptivists never mention the word…

    Also, I am increasingly inclined to think that there is nothing particularly rhetorical/figurative about the evolution of the word “decimate.” I think it just sounds a lot like “devastate” and people got confused.

    There’s no such word as “devastate” in German – other than the native verwüsten, stressed on the ü.

    It’s a matter of perspective. In Vienna it is commonly held that Asia begins at the Rennweg.

    That’s the Balkans. Nobody cares about Asia in Vienna.

    what’s wrong with “wherever”?

    There must have been a time when whither, where and whence must have been distinguished as strictly as wohin, wo and woher still are in German. In those days, “wherever you throw it” designated the area where the throwing happened, not or not only the target site.

  61. Three legs and no tail, poor cat.
    Hvor we take something hin is still used in Norwegian too. I can’t remember if it’s split in two also in German.
    Nobody cares about Asia in Vienna.
    Tell that to Suleiman the Magnificent.

  62. Bugger, that was meant to say:
    Hvor we take something hin is used in Norwegian. I can’t remember if it’s split in two also in German.

  63. Trond Engen says:

    Hvor … hen. But, yeah, it’s used. With hvor it feels somewhat oldfashioned, but colloquially it works as an interrogative even without the real interrogative: Hen ska vi gå i da?. Same with other specifiers: Ti ska vi møtes?

  64. I think it just sounds a lot like “devastate” and people got confused.
    Or ‘dessicate’.

  65. I think it just sounds a lot like “devastate” and people got confused.
    Or ‘dessicate’.

  66. Trond Engen says:

    Oldfashioned in writing, I meant. It’s a strange thing becoming oldfashioned while still being unremarkable in speech. I think it has to do with the recent lexical replacement by many speakers of the eastern colloquial forms of the interrogatives, starting in å, with the standard forms.

  67. Trond Engen says:

    Did you get that, or do you want me to repeat it?

  68. Those of us who didn’t get it the first time will need you to say it differently the second time.

  69. I think it’s easy enough to explain the modern sense of decimate without resorting to sound-alike words. You can imagine it being said (by someone who knows the classical sense of the word, but who is using the word in a slightly looser way) that a plague or a war has decimated the population of a city or a region, and it’s easy to imagine someone else (who does not know the word) understanding it, in some sense correctly, to mean “killing many” without thinking about how many.

  70. I should explain that I deleted a half-dozen or so copies of Trond’s comment immediately preceding “Did you get that, or do you want me to repeat it?”
    You can imagine it being said (by someone who knows the classical sense of the word, but who is using the word in a slightly looser way) that a plague or a war has decimated the population of a city or a region, and it’s easy to imagine someone else (who does not know the word) understanding it, in some sense correctly, to mean “killing many” without thinking about how many.
    Yes, exactly.

  71. Trond Engen says:

    Well, now it reads as it’s meant for Bathrobe’s puny two. I like that too, but it’s unintentional.

  72. marie-lucie says:

    Jesús: As far as I know, “decimare” derived from “decem”.
    This is the generally accepted derivation, but the existence of decumatus as an older alternative for decimatus raises questions. That’s why I suggested above that decumatus was older, and decimatus (and other forms of the verb) a folk reinterpretation linking the sequence decum- with the word decem ‘ten’. I don’t think there is any evidence that decum ever meant ‘ten’.

  73. I think I’d talk about the Mid-Atlantic Isles, and call Great Britain the East Island, Ireland the Middle Island, Iceland the North Island, and the Azores and Canaries the South Islands. “Western Islands” means the Hebrides to me.
    certainly Iceland & I think the Faroes are considered and consider themselves Scandinavian (but not Nordic countries)
    Exactly backwards: they are Nordic countries (part of Norden) but not in Scandinavia.
    Queen is also the Duke of Normandy, it’s in the list of her official titles
    Nope, as others have pointed out. Elizabeth II is the head of state of the Channel Islands in right of her crown as Queen of the United Kingdom, but this fact is not reflected in her official titles either in the U.K. or elsewhere (she has different titles in Canada, Australia, and the other “realms and dominions” from those she has in the U.K.) In any case it is clear that even though Elizabeth may be in some unofficial sense the Duke, Prince Philip is in no sense the Duchess.
    But what she is is Lord of Mann [sic], another title not reflected in the royal style. And if AJP wants to change the Isle of Man to the Isle of Person, he will have to decide whether he feels able to face the wrath of Personannan mac Lir, son of the sea-god for whom King Lear was named. (Properly, that should be “Cersonannan”, Manx being on the K-side of the P/K split.)

  74. Bill Walderman says:

    “since army hasn’t kept up that practice, there’s been no need for a need for a word of that particular definition (outside of the classics) for fifteen hundred years or so,”
    According to this, decimation was practiced as recently as WWI:
    http://www.crf-usa.org/index2.php?option=com_content&do_pdf=1&id=447

  75. John McIntyre says:

    One of the most impressive demonstrations of topic drift I’ve seen in some time.

  76. There’s something here which needs to be kept in mind: most users of English or French won’t spontaneously associate DECIMATE/DÉCIMER with a meaning “ten”/”tenth” (We who leave comments here are *not* a representative sample, I think there is nothing immodest in saying this).
    But I strongly suspect that to a native speaker of Latin the link between DECIMARE and the numeral DECEM (and its allomorph DECIM in such forms as UNDECIM, DUODECIM…) would have been utterly transparent (whatever its actual etymology is), so much so that a “non-etymological” meaning would have been impossible.
    Indeed I would almost advise a novelist writing a historical novel set in the days of the Roman Empire to coin a new verb, “to tenth”, to render to English ears the meaning which DECIMARE must have had to a Latin native speaker (and using latinate words in English to render hellenisms in Latin, perhaps…hmm, has anyone ever tried doing this?)

  77. David: There’s no such word as “devastate” in German – other than the native verwüsten
    Dunno about “native”. Not only does
    verwüsten mean “devastate”, but the words are etymologically related: OHG wuosti, Latin vastus, English “waste” and “vast”, see Grimm on the adjective wüst.
    Perhaps you merely intended to say that there is no German word devastieren. But there is: see the WiPe article on Devastierung. It may not be present in the active vocabulary of the man in the street, but then neither is a significant percentage of the words you and I use, probably.

  78. Wüstenei is a nice standard (although literary) word in this connection. Stressed on the last syllable, as usual, it means “desolate vastness”. Stressed on the first syllable it would mean “desert egg”.

  79. I know the word Wüstenei from one of the gloomiest of the gloomy songs in Schubert’s Die Winterreise.

  80. Sister_Ray says:

    A bit late to the party, but here is what my German dictionaries have to say:
    When I look up decimo in my trusted Stowasser it sends me to decumo and it is glossed as “punish every tenth soldier, decimate” and it has a little T which means Tacitus must have used it. If I understand the notes correctly it also informs me that forms like decumus for decimus are vowel changes because of stress/pronunciation. They also cite “optumus maxumus”as a variant of “optimus maximus”.
    My etymological Duden tells me that Germans started using dezimieren in the 17/18th C in its original meaning of “punishing every tenth man by death” from lat. decimare, but that it now means “make a lot smaller, decrease”.

  81. >Marie-lucie
    In Spanish we have a word pretty much similar: “diezmar”, also derived from “decem”. And related to tax: “diezmo”, as the French “dîme” (tithe); by the way, English “dime” (U.S. 10 cent coin) is also derived from this French word. I’ve found “decumanus” and “decimanus” as synonyms of “decimalis” meaning: belonging to tithe.

  82. Paul (other Paul) says:

    Jamessal: “The majority of linguistic change owes to people either being confused or lazy; it’s the rare change that’s conscientious.” That’s just what I’m against ! Just because people are lazy about a word does not, IMHO, make their misuse of it eventually the correct one. I know, I know, the argument against, but I think you’ve proved my point ;-)

  83. I think you’ve proved my point
    On the contrary, you don’t actually seem to be taking on board the “argument against.” Could you just stop for a moment and reflect on the fact that “The majority of linguistic change owes to people either being confused or lazy; it’s the rare change that’s conscientious”? Don’t just nod and move on; think about it. How can you possibly maintain that the vast majority of word meanings in English are incorrect? That’s as pure an instance as I’ve seen of ideology trumping reality, on a par with creationism or climate change denialism.
    To take a concrete instance, do you think the current meaning of bead is incorrect, and if so, what will you do about it? Stop using it? Use it only to mean ‘prayer’?

  84. Max Planck wrote that “A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.” (I’ve heard this rendered more pithily as “Science progresses one funeral at a time.”)
    I think it’s the same idea with language. Eventually, opponents to any novel usage die out, and the new meaning becomes standard. It’s never the case that the new meaning is so perfectly apt that it triumphs overnight. Clearly, “decimate” is on that trajectory, but not so far along that there isn’t still some opposition.
    I don’t like the bead/prayer example that much because I feel like “bead” shifted as a kind of metonym, and so the mechanism is totally different. Maybe it’s arbitrary, but I think that numbers live on a different conceptual plane–they exist so that we can be precise about things–and so it is inherently weird to use one number metaphorically about much larger, implied number. I also think that language will do what it does and I was never here to say that the evolved usage was “wrong”…but I think there will still have to be quite a few more funerals before the matter is settled.

  85. 1) “Mid-Atlantic Isles, and call Great Britain the East Island, Ireland the Middle Island, Iceland the North Island, and the Azores and Canaries the South Islands”: geographically repellent. The British Isles are on the European continental shelf whereas Iceland is on the mid-Atlantic Ridge, and there are similar objections for the others. There’s also the problem of your scandalous mistreatment of Madeira.
    2) Years ago I read about a psychologist who had looked at the evidence for Planck’s proposition: he found Planck to be wrong.

  86. And for those of you who know a bit of science: one of the oddest things I’ve ever heard at an examiners’ meeting (devoted to approving draft examination questions) was the enquiry “Planck’s constant – why have you written it with a little bar through the ‘h’?”
    Not the stupidest or most ignorant, but the oddest.

  87. As this is languagehat, not constitutionhat, I had refrained from commenting on the political aspects of Crown dependencies, but… it’s rather a case of things and the titles of things. The Queen, our Duke, is the person, “Elizabeth Windsor”. She’s called “Her Majesty Elizabeth the Second, by the Grace of God, of the United Kingdom of Great Britain and Northern Ireland, and of Her other Realms and Territories, Queen”. In Jersey she’s called “Duke of Normandy” but her constitutional capacity is as Queen of the United Kingdom (by the constitution, the monarch is the same person as the monarch of the United Kingdom). However, and here we come to the tricky bit, what is the Crown? Current definition seems clear: the Crown acts in Jersey as the “Crown in right of Jersey”. But is the case so clear when the Crown acts for Jersey outside Jersey? Can the Crown in right of the United Kingdom act for Jersey outside Jersey? It’s clear from cases in Australia and Canada that there may be a conflict between the Crown in right of a state or province and the Crown in right of the federation. Still, this isn’t constitutionhat, so the point is that talking about “the Crown” doesn’t necessarily mean an undivided entity.
    On to more definitions by language: British Isles. The peoples and governments of the Channel Islands generally believe themselves to be in the British Isles. We find it the most helpful description of where we are. Politically we’re in the British Islands (I’ve got that on my passport). However, there’s a difficulty in that in the English language a distinction is made between geographic “British Isles” and “British Islands” – but it’s very difficult to express that distinction in French (in Jèrriais, the dictionary terminology proposed is “Région Însulaithe Britannique” for “British Islands”). A similar difficulty arises with the distinction made between “prime minister” and “first minister” in English, which both are rendered as “premier ministre” in French. The head of government in Jersey is titled “chief minister” in English and “premier ministre” in French (“chef minnistre” in Jèrriais, although some people use “preunmié minnistre”).
    Back to the Channel Islands: in French in the Islands they are “Îles de la Manche” according to longtime usage (I used to have it in my old-style passport). In French of France, the name “îles Anglo-Normandes” is of comparatively recent invention. To English-speaking Channel Islanders though, “Channel Islands” is geographic (because there are a number of separate entities in the Islands) and can include the French Channel Islands (which cannot be included within the term “îles Anglo-Normandes”).
    And returning to the ducal question, the power of the title of “Duke of Normandy” feels rather different to me in English, as opposed to when I refer to “not’ Duc”; and singing “God Save the Queen” in English feels different from when I sing the Jèrriais version which uses both ducal and royal titles: “Dgieu sauve nouot’ bouôn Duc”.

  88. The Queen, our Duke, is the person, “Elizabeth Windsor”. She’s called “Her Majesty Elizabeth the Second, by the Grace of God, of the United Kingdom of Great Britain and Northern Ireland, and of Her other Realms and Territories, Queen”. In Jersey she’s called “Duke of Normandy” but her constitutional capacity is as Queen of the United Kingdom (by the constitution, the monarch is the same person as the monarch of the United Kingdom). However, and here we come to the tricky bit, what is the Crown? Current definition seems clear: the Crown acts in Jersey as the “Crown in right of Jersey”. But is the case so clear when the Crown acts for Jersey outside Jersey? Can the Crown in right of the United Kingdom act for Jersey outside Jersey? It’s clear from cases in Australia and Canada that there may be a conflict between the Crown in right of a state or province and the Crown in right of the federation.
    This may not be constitutionhat, but I found that extraordinarily interesting (as I do many things connected with the Isles of the Sleeve).

  89. jamessal says:

    do you think the current meaning of bead is incorrect[?]
    How about the pronunciation of ask? Originally pronounced “axe” — the way it is now in AAE — it changed via metathesis, i.e, the lazy, garbling of sounds within words?
    Max Planck wrote that “A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.” (I’ve heard this rendered more pithily as “Science progresses one funeral at a time.”)
    I think it’s the same idea with language. Eventually, opponents to any novel usage die out, and the new meaning becomes standard. It’s never the case that the new meaning is so perfectly apt that it triumphs overnight. Clearly, “decimate” is on that trajectory, but not so far along that there isn’t still some opposition.
    This is a poor analogy, because the changes railed against, by people like you, are utterly arbitrary. If some soi disant expert had singled out the current meaning of decimate as wrong because of its etymology, then the current meaning would have been accepted — gone unnoticed — like all the other words in the past four hundred years or so to have changed in a way inconsistent with their etymologies.
    I also wish you’d take the time to address Hat’s last comment to you, in which he made my argument even more explicit. To repeat, “How can you possibly maintain that the vast majority of word meanings in English are incorrect? That’s as pure an instance as I’ve seen of ideology trumping reality, on a par with creationism or climate change denialism.” If you think your analogy was an answer, then you need to consider 1) just how many linguistics changes have occurred in the past few centuries and 2) what I said earlier — what you thought, wrongly, made your own point — that the vast majority of these changes occur because of ignorance or laziness.

  90. jamessal says:

    If some soi disant expert had singled out the current meaning of decimate
    Sorry, that should have been hadn’t singled out . . ..

  91. Ettiene – worse still. I’d really have a hard time reading “to tenth the legion” as anything other than to reduce it to a tenth its size.

  92. jamessal says:

    Oh, Paul, hi — I hadn’t even realized I was arguing with you; I thought two new prescriptivists had wandered onto LH and needed to be shown what’s what. Congrats on starting your blog again! Still wish you’d address LH’s point, though (at long, long last!).

  93. Dearieme: Technically, Planck’s constant is written h. The symbol ħ represents the reduced Planck’s constant, or the Dirac constant, and is equal to h/2π. I assume that the bar in ħ represents the fraction bar.
    Geraint: I suppose, though, that if the monarchy were to be abolished in the U.K., the monarchy in Guernsey and Jersey would continue to persist in Elizabeth and her heirs forever, if no alternative arrangements were made by the States of Guernsey and Jersey respectively, because it is a personal capacity as you say.
    What is the Crown?
    Technically, nothing is the Crown: it is not one of the persons known to the law. The Crown does not sue or prosecute or give the Royal Assent: the Queen does it all. But it’s true that the Queen of Canada may take one point of view and the Queen in Ontario another, due to conflicting advice from her respective ministers (to say nothing of the Queen of Australia, whose attitude may be far more bolshie). Similarly, you Channel Islanders may or may not be subject to the Queen in Council outside judicial matters, but the Queen in Parliament has no power over you (at least in theory, if not always in practice).

  94. Ah! Perhaps we should take our advice from Geraint rather than the other way around, but John and I have history on this matter. As John knows, I disagree fundamentally with his firmly stated last sentence. De jure, both the Queen in Council and the Queen in Parliament have authority over the Channel Isles. De facto the authority is heavily restricted. That’s my story, anyway.

  95. I’m with Picky on that.

  96. marie-lucie says:

    Latin -imus/-umus
    Jesús, Sister_Ray, thank you for your comments. It looks like I was probably wrong about my hypothesis on decim- ~ decum-, but I would like to know more about when -umus was most used.

  97. marie-lucie says:

    Geraint, very nice to see you here again. You certainly live in an interesting place.

  98. Trond Engen says:

    So now I was just making silly jokes after all?

  99. Bill Walderman says:

    “One of the most impressive demonstrations of topic drift I’ve seen in some time.”
    You should read this site more often. This is probably the only site on the internet where topic drift is encouraged.

  100. Greg Lee says:

    languagehat: “How can you possibly maintain that the vast majority of word meanings in English are incorrect?”
    Without wishing to seem argumentative, what is wrong with that? Maybe they are all incorrect.

  101. Well, it obviously depends on your definition of ‘incorrect’. A working hypothesis that all word meanings in English are ‘incorrect’ seems to lead to only one useful conclusion: language changes so that what was once incorrect is now regarded as correct.
    There are certain mindsets that might sympathise with this view: for example, a cosmological vision that regards the universe as having been correct in its original form but since degraded by ongoing decay.
    From a linguistic point of view I’m not sure where it would lead. How about this: the urlanguage was a model of immaculate perfection that has been destroyed by succeeding generations? There are perhaps secluded parts of society that might even believe this, but I don’t think you will get much sympathy here.

  102. GL: If I understand rightly, your “maybe they are all incorrect” could be read as a short form of the statement: “All word meanings in English may be incorrect in that they are not the meanings they should be (in some sense of should). Nevertheless they are the meanings that most people know, which is good enough for communication”.

  103. Ø: I know the word Wüstenei from one of the gloomiest of the gloomy songs in Schubert’s Die Winterreise.
    I just picked up piano a couple of weeks ago. The first piece of music I’m (stil) trying to reproduce is the Sign-post.

  104. jamessal says:

    Well, it obviously depends on your definition of ‘incorrect’. A working hypothesis that all word meanings in English are ‘incorrect’ seems to lead to only one useful conclusion: language changes so that what was once incorrect is now regarded as correct.
    There are certain mindsets that might sympathise with this view: for example, a cosmological vision that regards the universe as having been correct in its original form but since degraded by ongoing decay.
    From a linguistic point of view I’m not sure where it would lead. How about this: the urlanguage was a model of immaculate perfection that has been destroyed by succeeding generations? There are perhaps secluded parts of society that might even believe this, but I don’t think you will get much sympathy here.

    Well put, Bathrobe! Just to make things clearer to Greg Lee, the reason such a view wouldn’t get much sympathy in these parts is that many of the readers or commenter here are either linguists or people with an appreciation for linguistics — a fruitful branch of science, which your** hypothesis would do much to undermine. “And for what?” is essentially what Bathrobe is asking. What do we gain? Why, exactly, should we trade an entire branch of science for your a useless hypothesis? (**Please excuse the word your if you hold no brief for that hypothesis.)

  105. Rodger C says:

    How about this: the urlanguage was a model of immaculate perfection that has been destroyed by succeeding generations?
    Isn’t that exactly what the Renaissance believed with its search for the Language of Adam? Umberto Eco has a whole book about it.

  106. Isn’t that exactly what the Renaissance believed with its search for the Language of Adam?
    Yes, indeed. The Renaissance believed all manner of silly things. (As do we, of course, as our descendants will be happy to point out.)
    Without wishing to seem argumentative, what is wrong with that? Maybe they are all incorrect.
    I’m hoping this is a poker-faced joke, because if it’s serious, I just don’t know what to say. (Fortunately, I don’t need to, since Bathrobe put it pretty well.)

  107. And yes, topic drift is one of my favorite things about this place. Serendipity forever!

  108. marie-lucie says:

    maybe they are all incorrect
    The history of the word which started this thread, decimate, shows that its meaning when first “borrowed” (adopted and adapted) into English has changed over the centuries and some people call the current meaning (although attested for quite some time) “incorrect” since it differs from the Latin meaning. But there have been many such changes not just within the history of English words but within the history of Latin, Greek, French, or whatever language the borrowed words were borrowed from. If we go back to Proto-Indo-European, the meanings of whatever entire words have been reconstructed to the earliest time, and especially the meanings of reconstructed roots, indicate considerable changes throughout the descendant languages. And PIE may be the most securely reconstructed language, but it must have had an ancestor too, common also to other language families. So how far should we want to go back to find the “correct” meanings? And if we are trying to learn those meanings, perhaps we should also try to go back as far as possible to learn (and use!) the original forms? Historical linguistics is a quest for learning about the history of languages, their relationships to each other and to their hypothesized common ancestor, not a call for present-day speakers to go back to using those archaic forms and meanings.

  109. David Marjanović says:

    Hvor we take something hin is used in Norwegian. I can’t remember if it’s split in two also in German.

    There’s a strong trend of blaming all directions on the verbs rather than the prepositions; wohin gehst du is less idiomatic than wo gehst du hin, where hin is the separable prefix of the verb hingehen. That’s easier to see in the past, where the prefix doesn’t separate: wohin bist du gegangen, wo bist du hingegangen.
    Is that what you mean?

    you merely intended to say that there is no German word devastieren. But there is:

    Huh. I had no idea.
    I’m not surprised, though. Sometimes I think every Latin/French word is automatically acceptable as a German one as soon as you slap -ieren or the like on it, just like how every Chinese character is theoretically acceptable to use in Japanese.

    Years ago I read about a psychologist who had looked at the evidence for Planck’s proposition: he found Planck to be wrong.

    It is mostly wrong, but not always entirely. It does happen that a few scientists stop being scientists, ignore the new evidence, and keep their opinions till they’re buried with them. But the universal acceptance of seafloor spreading in the late 1960s happened almost like scripted by Popper: by the mid-70s there seem to have been no fixists left – not even behind the Iron Curtain. A mass extinction of geologists around that time is not documented.

  110. marie-lucie says:

    Max Planck wrote “A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.”
    but
    a psychologist found this to be wrong
    Planck’s attitude is also found in Kuhn’s The structure of scientific revolutions, but in less strict words: a new scientific theory arises when the problems associated with an older one are coming more and more to the fore and some of the more creative scientists try other possible explanations. In order to succeed, a new theory needs to catch the minds of (usually) younger scientists by solving problems which were not well handled by the old theory, but those scientists who are wedded to the old theory (which has worked for them up to a point) may remain devoted to it until they die. By the time the new theory has become generally accepted and is actually taught to students as the state of the art, many of the older practitioners have retired and passed on, and the new generation may not even have heard of it.
    You could see why the psychologist in question would get different answers depending on which scientists he or she interviewed and whether or not there was serious dissension among the scientific community.

  111. Greg Lee says:

    I gather that most of you think it is absurd to think that most words are used incorrectly. I’m simply asking what your reasoning is. Why is it absurd? It’s no good imagining what theory of correctness I might have that would lead me to ask that, and then attacking that theory. I don’t have a theory of correctness. I’m a descriptive linguist — I don’t do the correctness thing. At the moment, I honestly have no idea what you might mean by this contention that most words must be used correctly. Explain yourselves.

  112. marie-lucie says:

    GLee, if you don’t have a theory of correctness, then you can’t have a theory of incorrectness.

  113. Paul (other Paul) says:

    Jim: I’ll post again shortly about being snow-bound for a few days, unusual in these parts.
    What I was trying to say is that I deplore the idea that words change not because of conscious decisions that they should have a new meaning, or that distinctions between similar words should be ignored, but because, as you said, people are lazy.
    This upsets me because my education, which was more than half a century ago, stuck, and part of that was “old fashioned” insistence on the immutability of words in their then current form. One may or may not like that – and clearly many of the LH readers do not – but that was ingrained in me, and reinforced by decades in journalism sticking to styles, in such fashion that it is now too late to change, I’m afraid. (I was only exposed to the descriptivist prescriptivist argument when I started reading LH a few years ago).
    And I mourn the loss of shades of meaning (and not so fine shades either), eg my favourite example – insure, assure and ensure. The fact that these may have been used interchangeably at some time in the past surely does not mean that their basically different meanings no longer count.
    Meanings here sometimes do overlap, but the fundamental meanings do not. The Oxford Dictionaries online say: There is considerable overlap between the meaning and use of insure and ensure. In both British and US English the primary meaning of insure is the commercial sense of providing financial compensation in the event of damage to property; ensure is not used at all in this sense. For the more general senses, ensure is the more usual word, but insure is also sometimes used, particularly in US English, e.g. bail is posted to insure that the defendant appears for trial; the system is run to ensure that a good quality of service is maintained.
    In UK English, I would say bail is posted to ensure thast the defendant appears for trial. That may be more an across-the-pond than a prescriptionist thing.
    Of assure, they say : 1 [reporting verb] tell someone something positively to dispel any doubts: [with object and clause]:Tony assured me that there was a supermarket in the village [with object and direct speech]:‘I quite understand,’ Mrs Lewis assured her [with object]:they assured him of their full confidence
    (assure oneself) make sure of something: she assured herself that he was asleep
    2 [with object] make (something) certain to happen:victory was now assured
    Of course (LH) I’m not suggesting we should now use bead only as prayer. But I think it is reasonable to be a bit upset at (IMHO) unnecessary loss of meaning in current words (and I know it will be said that that was true at any point in time – however, I can only talk about my own time).

  114. I gather that most of you think it is absurd to think that most words are used incorrectly. I’m simply asking what your reasoning is. Why is it absurd? It’s no good imagining what theory of correctness I might have that would lead me to ask that, and then attacking that theory. I don’t have a theory of correctness. I’m a descriptive linguist — I don’t do the correctness thing. At the moment, I honestly have no idea what you might mean by this contention that most words must be used correctly. Explain yourselves.
    Is this performance art? If you’re a descriptive linguist, you must be aware that the only scientific sense of “correct” in this context is “used by (a significant number of) native speakers.” If English speakers use X to mean Y, then that is the correct meaning of X, whatever it may have meant at some earlier time. If you don’t want to use “correct” at all (which I guess is what you mean by “I don’t do the correctness thing”), that’s fine, but it seems bizarre for you to pretend you don’t know what the rest of us are talking about.

  115. David: Is that what you mean?
    Yes, exactly. That’s great to know. Thanks, David!

  116. Possibly jamessal’s use of the word “lazy” has had an unnecessarily provocative effect on Greg Lee.
    About the “loss of distinctions” theme in prescriptivist complaints: Yes, it’s always possible to see examples of distinctions being lost by one’s lazy contemporaries, but it’s equally possible to see new distinctions being made by new usages. We may have our moments when we are offended by new uses of old words and also our moments when we are offended by new words; in the former case we have “loss of distinctions” available as a plausible explanation for our feelings, but in the latter case I think that all we’ve got is “I don’t like it”.

  117. (Greg Lee, and others, too)

  118. I gather that most of you think it is absurd to think that most words are used incorrectly.
    To quote myself: “Well, it obviously depends on your definition of ‘incorrect’” and m-l “if you don’t have a theory of correctness, then you can’t have a theory of incorrectness”. It would be useful if you could explain what your reasoning is. Then we might know what you are talking about.

  119. The modern British distinction between (commercial) insurance and assurance was invented by Charles Babbage in 1826. Before that, insurance had almost completely displaced the older term assurance except in the names of certain institutions. The distinction has never obtained in American English, where insurance has become universal.

  120. Languagehat writes: If you’re a descriptive linguist, you must be aware that the only scientific sense of “correct” in this context is “used by (a significant number of) native speakers.”
    I’m not aware of that. Reference please?
    Maybe a phonological parallel will be useful. There is a notion of “phonemic form” that makes it represent the intention of speech, so that allophonic variants of phonemes are not pronunciations we intend. As, for example, my saying a glottal stop for the “tt” in “button” or a flap for the “tt” in “batter” is not because I meant to say the glottal stop or flap. These allophones are unintentional.
    The unintended allophones that turn up in pronunciation are often interpretable as sloppiness — people just don’t go to the extra effort that would be required to say the phonemes. Suppose we say that the phonemic pronunciation is correct and the sloppy substitution of allophones is incorrect. Phonologists don’t actually say that, but if you could talk them into using this correct/incorrect terminology at all, maybe then they could be talked into it.
    The advantage for clarifying this discussion of how many word of English are used correctly is that the matter becomes perfectly concrete. What proportion of words in English are pronounced phonemically? All of them? Some of them? None of them?
    Well, it turns out to be none of them. We never pronounce words the way we intended, so in a certain sense, we always use words incorrectly.
    Maybe this will help explain why I have such resistance to accepting without evidence or rational argument this notion that words must necessarily be used correctly by speakers. Frankly, I don’t see any linguistic reasoning going on here at all.

  121. marie-lucie says:

    GLee: Suppose we say that the phonemic pronunciation is correct and the sloppy substitution of allophones is incorrect. Phonologists don’t actually say that, but if you could talk them into using this correct/incorrect terminology at all, maybe then they could be talked into it.
    It is good that you gave an example from phonology in order to clarify your use of correct/incorrect. I think that given the initial topic of this thread, most of your readers here thought that by words used correctly/incorrectly you referred to the meanings or the contextual uses of those words. But your use of these two terms as potentially used in phonology do not correspond at all to the attitudes of descriptive linguists (even though you claim to be one of them), instead they seem to be entirely prescriptivist.
    Phonologists would never say that when there are two pronunciations (allophones) of the same phoneme under different conditions, one of them is correct and the other incorrect: both uses are correct when used in the same way as competent native speakers use them (“competent” speakers are those whose peers judge them to be speaking just like them, without unusual peculiarities attributable to immaturity, speech defects or other personal characteristics affecting their speech). If you are learning Spanish, for instance, you need to learn to pronounce /b/ differently at the beginning and in the middle of a word. Using [b] in both contexts will mark you as a foreigner, not as a “correct” user of the language according to the unspoken norms of the Spanish-speaking community. The most that a phonologist could say would be that (for instance) such or such an allophone is not “considered” correct within the speech community, but a properly trained phonologist would never make such a value judgment on their own.

  122. Bathrobe says:

    The unintended allophones that turn up in pronunciation are often interpretable as sloppiness…We never pronounce words the way we intended, so in a certain sense, we always use words incorrectly.
    I don’t see how this applies to prescriptivism in relation to the meaning or usage of words.

  123. And as m-l points out, your phonological example is flawed as an example of correctness vs incorrectness.

  124. When a Chinese speaker says /ju: hæv tu: rʌn fastə/ for ‘You have to run fast’, every phoneme is being pronounced according to intention.
    Unfortunately, this pronunciation of the sounds exactly as intended is incorrect, as it is more likely to be interpreted as ‘You have to run faster’ (depending on your dialect, maybe). Why should a fastidiously correct pronunciation that ignores allophonic variation be misunderstood? Because to insist on fastidiously ‘correct’ pronunciations as being ‘correct’ is not descriptivist thinking at all.

  125. Greg Lee says:

    To minimize misunderstanding, I should have said I was referring specifically to optional rules of allophony, which create stylistic pronunciation variants in casual speech. The examples I gave were of this sort. I can say [t] in “button” or “butter” instead of glottal stop and flap, resp., but ordinarily I don’t bother. When I hear [t] pronounced in those words, it doesn’t make me think I’m listening to a non-native speaker of English; it just sounds stilted (or British).
    In fact, I think some prescriptivists would call the pronunciations with glottal stop and flap “incorrect”, since they need not occur in formal speech. So there is some natural association between “phonemic pronunciation” and “correct pronunciation”.
    It doesn’t seem fair to me to accuse me of not being a true descriptivist when I use the term “correct”. I don’t, ordinarily, and it wasn’t me who introduced it into this conversation. It was you guys. I just being my usual helpful self and trying to help you figure out what you might have meant by it.

  126. But the phonological example is a red herring.
    The proposition, if I remember rightly, was that ‘the vast majority of word meanings in English are incorrect’.

  127. Greg Lee says:

    No, Bathrobe, look back and you will see that this subdiscussion began this way:
    Jamessal: “The majority of linguistic change owes to people either being confused or lazy; it’s the rare change that’s conscientious.” That’s just what I’m against ! Just because people are lazy about a word does not, IMHO, make their misuse of it eventually the correct one. I know, I know, the argument against, but I think you’ve proved my point ;-)
    Posted by: Paul (other Paul) at March 27, 2013 08:47 AM

    So there was no mention of meaning, and you do not remember rightly.

  128. Once again, you are being belligerently disingenuous; the quoted sentence from jamessal is immediately followed by “The fact remains, decimate denotes mass destruction even in written Standard English,” so it is in fact meaning that was under discussion.

  129. @Greg Lee: According to your quoted sentence, we are talking about the use (or misuse) of words, not their pronunciations. That is why phonology is a red herring.
    The statement that came later in the discussion still cries out for clarification:
    “How can you possibly maintain that the vast majority of word meanings in English are incorrect?” — Without wishing to seem argumentative, what is wrong with that? Maybe they are all incorrect.

  130. Greg Lee says:

    @Bathrobe: Yes, I asked what was wrong with maintaining that the vast majority of word meanings in English are incorrect. I still don’t know. (I’m not counting languagehat’s response to the effect that anything anyone ever says is correct. Really, no one using the term “correct” would ever mean by it no more than that something was said.)
    I thought the phonological analogy would be helpful, but I guess it’s not working. So, tell me, why is it absurd to think that the vast majority of word meanings in English are incorrect?

  131. First, we need to define ‘correct’ and we’re still not getting there. Are you talking about prescriptively-defined ‘correctness’ or something else again? Hat has offered the position that usage defines correctness. You disagree.
    If the vast majority of word meanings are ‘incorrect’ but the words work all the same, what exactly is this ‘incorrectness’ of which you speak? You are the one who has taken the position that a key element of language (the meaning linguistic forms) is ‘incorrect’. If that’s not absurd, please explain why not.

  132. marie-lucie says:

    I did not think that the phonological example was a red herring, because phonology has a more definite organization than semantics, so that it was easier (at least for me) to understand what G Lee meant by “correct” and “incorrect”. At least I thought I understood. Earlier he said “I am a descriptive linguist, I don’t do correctness”, while taking a position with is stereotypically prescriptivist (making value judgments about correctness). Now he reiterates that most words are being used incorrectly, while refusing to say what he means by that or by “correct”. This discussion is going nowhere at all.
    As for “phonemic pronunciation = correct pronunciation”, such a statement assumes that a phoneme (a) is always strictly determinable and (b) optimally represented orally by a single specific sound. Neither of these general statements is true. Anyone who has undertaken phonemic analysis with the goal of setting up an alphabetic writing system for a hitherto unwritten or inadequately written language will have had to struggle with how to define and represent this language’s phonemes. Sometimes an arbitary decision, or one based on some extralinguistic criterion, is unavoidable.

  133. I expect certain contributors here to experience a fit of exasperation at the following comment, for the reason that it seems to contain more general bla-bla than attention to specifically linguistic matters. So I may just mention that, in my view, discussions of any particular kind are always based on a considerable amount of general bla-bla, i.e. broad assumptions about “reality”, “what is a description” etc that are not made explicit, but if made explicit could be seen to be relevant to the particular discussions.
    To say that a meaning is “correct” doesn’t make much sense, to me at any rate. Statements (or any kind of “claim”) can be correct or incorrect, right or wrong. I’m talking here about the predicative use of these words. A thing or group of things is not a statement or claim. “The vast majority of word meanings in English” is not a statement or claim, nor is “the universe” (Bathrobe’s example).
    To use the word “correct” is (at least surreptitiously) to use a concept of adequation or representation. In terms of being “correct” or not, a statement is thought of as being comparable with an imagined “reality that the statement is about”. A particular statement of the form “this round thing is an apple” is correct exactly when this round thing is an apple.
    Not only statements, but also “states of affairs” can be called correct when compared with something else. When you’ve assembled an Ikea cupboard, someone may state that you’ve incorrectly mounted the handles. This statement is correct when, upon comparison with the Ikea assembly instructions and the catalog picture, it is found that the handles are incorrectly mounted.
    It seems that “X is correct”, where X is not a statement, is a common way of expressing the results of comparing X, or a context in which X plays a role, with something else, to determine whether X is in conformance with that something else. When a person A says “You are correct” to person B, I usually take that to mean that A is saying to B something equivalent to “the statement(s) you are making is/are correct”.
    What could it mean to say that the meaning of a word is “correct”, although meanings are not statements ? Suppose I use the word “beads” to refer to beads, say as opposed to prayers. Where could any “correctness” be found in this context ? Surely it is my use of the word “beads” to mean beads that is correct or not, by comparison with an imagined “reality of how people use (today or in the past) the word ‘beads’”. There is no such thing as “the meaning of ‘beads’” independent of conventions to which people adhere (or did in the past) as to what “beads” means.
    Or “should” mean … Prescriptivists claim “people should speak like this”, descriptivists claim “people just speak like that”. Defenders of descriptivism tend to claim in addition that “it should not be said that people should speak like this instead of like that”. All are into the should game. My view is that some people do misunderstand and misuse words (as referred to some set of conventions). Such phenomena are capable of description without immediately being subjected to moral evaluation.
    The above sketchy account of meaning and reality is itself “correct” only when compared with an imagined “reality of how statements represent reality”. Most people have no trouble imagining the reality of beads, but balk at imagining the reality of statements about reality. On my view it’s merely a matter of what you’re accustomed to, in particular whether you’re accustomed to put quotation marks around reality, and to refrain from insisting that reality is real (or that imagining is imaginary).
    I find it useful to switch around among viewpoints, rather than cling for dear life to a single one. That is one solution to the Gynecologist’s Dilemma.

  134. jamessal says:

    Stu, given how rare it is to tightly nail someone down for belligerent disingenuousness, as Hat and Bathrobe have done with Greg Lee, I do hope your varying view points don’t muddy the waters, as otherwise salutory those view points might be. I don’t at all intend that as an insult; it’s just rare indeed that bullshit be called for what it is, with so little room for the bullshitter to further prevaricate.

  135. Jim, I don’t find what Greg Lee wrote to be in any way belligerent or disingenuous. Your own contributions to this comment thread have been pretty belligerent so far, though. “Bullshit”, “prevarication”, “disingenuousness” … It’s as if you imagine descriptivism to be a damsel in present distress.
    Not myself being much of a gentleman, I leave the defence of virtue to others, and merely comment on the fisticuffs. I guess you could say I am a grouch potato.

  136. I haven’t called him disingenuous etc., but I would sorely like him to explain what he means.

  137. I see now that it was Hat, not Jim, who introduced “belligerent disingenuousness”. For what it’s worth, as a past master at displaying such behavior I don’t find in GL’s comments anything that measures up to my standards.

  138. Bathrobe, your provisional take on GL’s meaning was: “… the vast majority of word meanings are ‘incorrect’ but the words work all the same …” That’s essentially what I had speculated just following a previous comment by you to the same effect.
    It’s up to GL to say whether this is what he means. In the context of my last long comment, I would guess that “incorrect” here is intended to mean “incorrect in terms of older conventions”. In terms of newer conventions, the words work (“all the same” but in a different way). I think GL meant this more or less as a tease, but now the juggernaut of prescriptivism has been sent out his way, and Parlament is asking questions about whether the expense can be justified in the eyes of the public.

  139. Oops, I mean juggernaut of descriptivism. I do tend to confuse these two old imperial dreadnoughts (or do I mean “imperious” ?)

  140. There may be something in what GL says, but it’s not getting through. Since the commenters on this blog tend to be well educated, literate and familiar with the norms of prescriptive grammar, the written language, and dictionaries, we possibly have a more enduring faith in ‘standard language’ than we let on, even if we are not prescriptivists. Perhaps words in action (everyday speech) have a more protean quality than we are willing to admit. But it would be nice if GL could tell us exactly why he made that sweeping statement about most words being used incorrectly.

  141. Trond Engen says:

    I hadn’t meant to enter this, but I think it’s important that Greg didn’t make that statement. It came up as the absurd end of a reductio, and he asked why it’s absurd. There may be an unstated premise behind that question, but, OTOH, it may not.
    I’ll take this from another angle: As much as I agree with the absurdity, I think much discussion until now has missed the main prescriptivist view: It’s change itself that’s bad and has to be stopped. It doesn’t really matter if Caesar did it, or Shakespeare. It’s irrelevant if the forms defended came about as errors or if they even are linguistically accurate. The point is, they’re part of an acquired norm, a style one has internalized or learned how to emulate, and the more narrowly the norm is defined, the less likely it is to change. That’s why outspoken defenders of the old order so often want to change it by adding new “zombie” rules.
    So, essentially, to the prescriptivist, the fact that all words could be wrong proves their point.
    The non-prescriptivist argument, then, is rather that language is a dynamic process, and it’s always been, like a river. Trying to interfere with that is futile at best, destructive to the qualities they love or even fatal to the language itself at worst.

  142. So, essentially, to the prescriptivist, the fact that all words could be wrong proves their point.
    But GL is claiming not to be a prescriptivist! That’s the whole point. and why I call him disingenuous. If he were a standard-issue peever, irrationally but understandably lamenting the alleged loss of precious distinctions and the coming reduction of all language to meaningless babble, one could chuckle and ignore him. But for someone who claims to be a descriptive linguist to claim not to understand why it doesn’t make sense to say the way English-speakers use English words is incorrect smacks of trolling. Note that despite repeated requests he has not clarified why he thinks it could possibly make sense to say that most words are now used incorrectly; he reminds me of a clever high-school sophomore who has just discovered solipsism and goes around asking other people why he should believe they exist. Me, I’m not going to bother engaging him any further unless he says something I consider intelligible.

  143. Let me expand on Trond’s first paragraph.
    Jamessal wrote:
    The majority of linguistic change owes to people either being confused or lazy; it’s the rare change that’s conscientious.
    The context was different senses of decimate, so let’s ignore other kinds of linguistic change and confine ourselves to jamessal’s statement as it applies to changing senses of words (“meanings”). That’s what the subsequent exchanges were about, except for the red herring bit.
    Paul responded, seeming to agree with jamessal’s assertion but also saying:
    Just because people are lazy about a word does not, IMHO, make their misuse of it eventually the correct one.
    Hat responded:
    How can you possibly maintain that the vast majority of word meanings in English are incorrect?
    It’s clear in context that by “word meanings” Hat meant “senses of a word”, and that by “incorrect word meanings” Hat meant what Paul meant by “misuse of a word”, so I don’t we need Grumbly’s bla-bla here.
    That was Hat’s reductio.
    Then along came GL, asking Hat what’s so absurd about the statement that “the vast majority of word meanings in English are incorrect”.

  144. I do think that some part of this little tempest in a pickled herring jar can be traced to jamessal’s use of the words “confused” and “lazy”. This can be read as saying that language change begins when people misuse words.

  145. Prescriptivists and Descriptivists have divided control exclusively between themselves for too long, like Democrats and Republicans in the USA. I would like to put in a word for the Nudgers. I learned about this in Kahneman’s “Thinking, Fast and Slow”. “Libertarian paternalism” combines equal amounts of descriptivist and prescriptivist dogma:

    Life is more complex for behavioral economists than for true believers in human rationality. No behavioral economist favors a state that will force its citizens to eat a balanced diet and to watch only television programs that are good for the soul. For behavioral economists, however, freedom has a cost, which is borne by individuals who make bad choices, and by a society that feels obligated to help them. The decision of whether or not to protect individuals against their mistakes therefore presents a dilemma for behavioral economists. The economists of the Chicago school do not face that problem, because rational agents do not make mistakes. For adherents of this school, freedom is free of charge.

    In 2008 the economist Richard Thaler and the jurist Cass Sunstein teamed up to write a book, Nudge, which quickly became an international bestseller and the bible of behavivioral economics. This book introduced several new words into the language, including Econs and Humans. It also presented a set of solutions to the dilemma of how to help people make good decisions without curtailing their freedoms. Thaler and Sunstein advocate a position of libertarian paternalism, in which the state and other institutions are allowed to nudge people to make decisions that serve their own long-term interests.

    Of course these authors have merely adapted modern, take-it-or-leave-it therapeutic theories for the field of politics.

  146. jamessal says:

    Jim, I don’t find what Greg Lee wrote to be in any way belligerent or disingenuous. Your own contributions to this comment thread have been pretty belligerent so far, though. “Bullshit”, “prevarication”, “disingenuousness” …
    Quoting Hat, I called him “belligerently disingenuous,” and if you haven’t found him to be so, I suggest rereading his comments, along with the ones he was purportedly responding to. As for “bullshit,” I have no problem calling it when I see it. And as for your damsel-in-distress metaphor about descriptivism, that’s just silly.

  147. empty: … so I don’t we need Grumbly’s bla-bla here. … Then along came GL, asking Hat what’s so absurd about the statement that “the vast majority of word meanings in English are incorrect”.
    You’ve got the chronology wrong. My bla-bla came after GL’s question, addressing it and the various huff’n’puffings it occasioned.
    I do think that some part of this little tempest in a pickled herring jar can be traced to jamessal’s use of the words “confused” and “lazy”. This can be read as saying that language change begins when people misuse words.
    I essentially do read it that way, except I think he means that some language change begins when people misuse words. It would be hard to deny the truth of that with impunity.

  148. jamessal says:

    This can be read as saying that language change begins when people misuse words.
    Much of it does.

  149. jamessal says:

    Stu, that political quote confirms it: you really don’t understand descriptivism. I’d be willing to break it down for you any time.

  150. The prescriptivist position is: “Let us describe speech patterns and how they change. Let us not try to interfere in that with rules and regulations”. Would that be a fair break-down ? If so, it seems only natural to pin the epithet “libertarian” on it.

  151. I meant “descriptivist position”. Just can’t keep the right words in mind. That shouldn’t bother a descriptivist, though.

  152. Merely because I put up a pretence of confusing the words “descriptivist” and “prescriptivist” doesn’t imply that I am a prescriptivist with a liking for sophism. I am a Nudger in practice. When someone in a particular situation uses what I think is a wrong word (wrong in terms of what they apparently meant to say, against the background of current usage), then I don’t see myself as trying to halt the progress of language if I suggest that maybe another word would make more sense in that particular situation.
    I am not concerned with setting up rules to be followed in all situations. I have little sympathy with prescriptivist attitudes, but even less with refusal to acknowledge that people can make mistakes and that it is OK in principle to point these out.

  153. The [descriptivist] position is: “Let us describe speech patterns and how they change. Let us not try to interfere in that with rules and regulations”. Would that be a fair break-down ?
    Really depends on who you mean by us.

  154. Sorry, I would have responded earlier; I’m currently selling ice cream, and must stop typing for each customer.

  155. “Us” means the descriptivists who are setting out their position in the words I surrounded with quotes, as if quoting.

  156. Merely because I put up a pretence of confusing the words “descriptivist” and “prescriptivist” doesn’t imply that I am a prescriptivist with a liking for sophism. I am a Nudger in practice. When someone in a particular situation uses what I think is a wrong word (wrong in terms of what they apparently meant to say, against the background of current usage), then I don’t see myself as trying to halt the progress of language if I suggest that maybe another word would make more sense in that particular situation.
    I am not concerned with setting up rules to be followed in all situations. I have little sympathy with prescriptivist attitudes, but even less with refusal to acknowledge that people can make mistakes and that it is OK in principle to point these out.
    You sound like a descriptivist who doesn’t really know what a descriptivist is.

  157. Let’s do a deal: if you describe the best ice cream you currently have on offer, I won’t insist that the blueberry kind is better.

  158. You’ve got the chronology wrong.
    OK, but still, through all the huffings and puffings I don’t think there was any confusion, any talking at cross purposes, attributable to different senses of the word “meaning”.

  159. You seem to have ascribed to descriptivists a false dogma, as if any and all matters of prescribing were anathema to us, whereas really we’re just as reasonable as you make yourself idiosyncratically to be; we just respect the science of linguistics, and think about language with the tenets of that science in mind.

  160. If the course of this comment thread after GL’s remark did not involve confusion or talking at cross purposes, or it did but this was not attributable to different senses of the word “meaning” – then what do you think it did involve, and to what do you think it is attributable ?

  161. My last question was addressed to empty.

  162. Roasted Strawberry is really fantastic. Because we broil to the point of blackening the strawberries before adding them to the ice cream, the berry flavor really pops through in a way it doesn’t in any other strawberry ice cream we’d encountered before. As opposed to a hint of strawberry drowned by cream, this is the delicious berry in full followed by cream: strawberries and cream as it was meant it be.

  163. Because we broil to the point of blackening the strawberries before adding them to the ice cream, the berry flavor really pops through in a way it doesn’t in any other strawberry ice cream we’d encountered before.
    Damn, that’s description at its best ! I would prescribe a helping for everyone.

  164. Flattery will get you everywhere.

  165. Is there anybody out there who does not agree with the following?
    It happens every day that words get used in new ways through mis-hearing, misunderstanding, muddle-headedness, laziness.
    When it first happens, when one speaker uses a word in some odd new way, not consciously innovating but just screwing up, it’s fair to call it an error. Most of these errors, or accidental innovations, disappear leaving not a ripple behind.
    But a few of them catch on. It may even reach the point, many years later, where the “new” usage, now old, strikes nobody as wrong. In fact, just about every usage we have that now strikes nobody as wrong was once an error.

  166. I still can’t let you off the hook for this bit of snark, though: “Us” means the descriptivists who are setting out their position in the words I surrounded with quotes, as if quoting. “Us” still could have referred to a larger group of people, as if the speaker — himself a descriptivist, to be sure — were delivering a broader injunction. Considering you only wrote to sentences to define “descriptivist,” I thought it important to understand exactly what they were intended to convey.

  167. Is there anybody out there who does not agree with the following?
    Not I!

  168. what do you think it did involve, and to what do you think it is attributable ?
    Differing senses of the word “incorrect”, maybe.
    I carefully wrote and posted a comment a little while ago, and it didn’t appear. It was going to be the first of two or three related comments. Damn.

  169. Oh, there it is at last. Now (Greg Lee) if you do agree with my last comment, can you (with a straight face) say that it would be a good use of words to describe these usages that no longer strike anybody as wrong as being “errors”, or “incorrect”, or “misuse of words”?

  170. Of course there is also the in-between case, where a usage has been accepted by a large number of speakers but still strikes others as wrong. Hardly anybody would claim to be able to draw an absolute line between the two extremes, except as a matter of personal taste or preference.
    (And I have to mention the bandwagon peeving phenomenon: people who never saw anything wrong with a well-established usage, but when they hear someone else complaining about it then they think that it will mark them as superior if they start complaining about it themselves.)

  171. It happens every day that words get used in new ways through mis-hearing, misunderstanding, muddle-headedness, laziness.
    When it first happens, when one speaker uses a word in some odd new way, not consciously innovating but just screwing up, it’s fair to call it an error. Most of these errors, or accidental innovations, disappear leaving not a ripple behind.
    But a few of them catch on. It may even reach the point, many years later, where the “new” usage, now old, strikes nobody as wrong. In fact, just about every usage we have that now strikes nobody as wrong was once an error.
    That is an excellent description which (fair warning) I am likely to steal.

  172. flattery and dividing people to us and them in whatever context is the two things i can’t stand, the descriptivist and prescriptivist divide seems to me something not very that serious, life or death deciding matter, whatever side wins wins though i suspect the problem between them is like unwinnable
    i saw recently the map of the people who changed their fb icons to various red icons supporting the prop 8, thought ah so that many people seems like live well and idle enough to be concerned with it and feel strongly enough to do something about it like changing icons. i wish the same display of solidarity was shown against the iraq war but at that time there was no facebook i guess
    i mean i of course support the marriage equality for all, just that so much collective craze like movement feels not to “my” tastes, marry whoever you like but why everybody, the state, the court, should be concerned about who one loves or not is just strange, people cite all the reasons which sound something materialistic, health benefits, inheritance, life insurance
    well should support whatever side is the more progressive side i guess

  173. That is an excellent description which (fair warning) I am likely to steal.
    Ditto! My “Not I!” was meant to convey to the same enthusiasm, but I realized that, with the double negative, not only might the enthusiasm get muddled, but also the meaning might be misconstrued.

  174. marie-lucie says:

    jamessal, No muddle at all, I understood you perfectly! And I am sure most of the hatters also did. I also agree with LH about your excellent description.
    language changing through errors, eg misuse of words
    I agree that this may happen sometimes, but error is certainly not the main cause of linguistic change.
    Words are not normally used in isolation but in a context, both social and linguistic. Change of meaning may happen when a word comes to be used in a context different from its usual one, often in a larger or a more restricted context, and the change of context influences the specific meaning of the word. Eventually the new context may become the more common one and the old context may disappear, along with the older meaning of the word. Confusion of words happens when words are both similar in shape and used in related contexts (like insure/ensure/assure), and ordinary people who are unfamiliar with those specific contexts are aware of a general shared meaning but not of the often professional or technical contexts in which the subtle differences in such words make sense. Another reason for change is the deliberate “misuse” or just different semantic use of a word by a group in order to differentiate itself from the majority by making its speech difficult to decode by non-members: for instance using BAAAD to mean ‘nice’. For phonetic or phonological change, which affects the sounds of a language, laziness may explain the use of a less forcefully uttered sound (as in the usual North American pronunciation of the tt in Patty or pretty), but “laziness” may be mistaken for simply “relaxation” under informal circumstances where participants in a conversation know each other and each other’s mode of speech, so that they do not have to accentuate the usual features of their pronunciation. Speech fashions can also arise which deliberately exaggerate normal pronunciation features (either towards weakening or intensifying), again in order to show one’s membership in a specific group and one’s non-membership in another. It is almost a truism that teenagers, who tend to be creative and experimental in many ways, do not want to talk exactly like their parents or grandparents any more than they want to dress exactly like them. Changing even small details of their speech in the same manner as their own peers (however they define them) makes sure they cannot be identified with an older generation.
    These are just a few reasons for language change, but there are many more which cannot be simply attributed to errors in phonological form or semantic usage. There are many textbooks in linguistics which spell out such changes in detail. In the vast majority of cases, change occurs because of sociolinguistic reasons, related to speakers’ more or less conscious understanding of their place relative to others in a given speech community. As the generations succeed each other, what had become the norm for one generation becomes archaic after a few others.

  175. jamessal says:

    Here’s a list from a linguistics textbook under the subheading “Creating New Words and Changing the Meaning of Words.” It’s in the morphology chapter, so it doesn’t deal with changes stemming from phonology, e.g., syncope, apocope, metathesis — the types of changes, I should mention, I initially had in mind when I first used the word “lazy.” (John McWhorter called such phenomena “laziness of the mouth.”)

  176. jamessal says:

    I also agree with LH about your excellent description.
    Thank you! (Although it was Grumbly who complimented my ice cream description, and Ø who wrote the description that Hat complimented.)

  177. syncope, apocope, metathesis — the types of changes, I should mention, I initially had in mind when I first used the word “lazy.”
    So I stand corrected. GL was justified in bringing up phonology since jamessal originally had that in mind. However, comments by Paul (other Paul) (‘Just because people are lazy about a word does not, IMHO, make their misuse of it eventually the correct one’) and Hat (‘How can you possibly maintain that the vast majority of word meanings in English are incorrect?’) both steered the discussion in the direction of semantics and changes in meaning.

  178. marie-lucie says:

    jamessal, I too would compliment you on your ice cream description – it sounds “to die for”. Too bad I am not in your neighbourhood.

  179. jamessal says:

    So I stand corrected. GL was justified in bringing up phonology since jamessal originally had that in mind.
    You stand no such thing. And GL was in no was justified in breaking out that nonsensical analogy. I merely mentioned that in passing, because McWhorter’s quote always comes back to me whenever the subject of language change arises. The subject at hand was clearly meaning.

  180. jamessal says:

    Too bad I am not in your neighbourhood.
    Too bad indeed! Far too many Hatters with whom I’d love to share ice cream live too far away!

  181. Trond Engen says:

    Seen from here, the two of you live pretty close!

  182. jamessal says:

    I’m sorry, a few comments up I said “phonology” when I meant “phonetics.”

  183. jamessal says:

    Well, if you ever found yourself close enough to drop in, Trond, please do!

  184. Greg Lee says:

    I want to reply or comment on some of this discussion, and I’ll try to, soon, but there is an awful lot here. You all are typing faster than I can read.
    I’ll make a preliminary comment that I didn’t ever intend to propose that most English words are used incorrectly, but only that I doubt it is absurd to think the contrary. I offered the phonological analogy (not intended as an account of correctness) to make my doubt seem reasonable.

  185. Trond Engen says:

    Close to eachother, I meant. But if I ever do, I will.

  186. Once Lord Kames asked Lord Monboddo if he had read Kames’s latest book. “No, my Lord, said Monboddo, “you write a good deal faster than I am able to read.”

  187. Trond Engen says:

    Oh, I might reciprocate with an open invitation to anyone who finds themselves lost in the less spectacularly spectacular parts of Norway. I can’t promise homemade icecream, though.
    Greg: The ox is slow but the earth is patient.

  188. I’ll make a preliminary comment that I didn’t ever intend to propose that most English words are used incorrectly, but only that I doubt it is absurd to think the contrary.
    I understood that, but I don’t understand why you doubt it is absurd to think the contrary; it seems to imply that you think it makes sense (= “not absurd”) to say that most English words are used incorrectly, and I don’t understand what it would mean to think that (for someone trained in linguistics). I look forward to clarification.

  189. marie-lucie says:

    If you think that most English words are used incorrectly (presumably in sync rather than in contrast with most French, Italian, Russian, Chinese, etc words), try to test this statement with any text and find the words that are indeed “used incorrectly” (this would mean with non-standard lexicon, morphology and syntax, since phonology is ruled out by the written medium). I think that you will find very few texts (except for the spam obviously written by non-native speakers) where even a small minority of words are “used incorrectly”.

  190. m-l: Note that he said “I didn’t ever intend to propose that most English words are used incorrectly, but only that I doubt it is absurd to think the contrary.”

  191. marie-lucie says:

    LH, when I write “you” I don’t necessarily mean just the person I am responding to. It often means “anyone who”.
    GL, I have now reread the entire thread (excepting the digression on the Channel Islands) in order to try to understand what is your actual position, but I feel I have failed. I am confused by the differences between the several rewritings of your statements (sometimes in question form), and by the abundance of negatives (actual or implied), all of which make your position or your change(s) of position hard to figure out.

  192. Greg Lee says:

    Hat: … it seems to imply that you think it makes sense (= “not absurd”) to say that most English words are used incorrectly, …
    The most conventional way a descriptivist could take issue with what you said is to take the position that it is nonsense to speak of language expressions as being either “correct” or “incorrect”, so that when you proposed that the majority (or all?) English words are obviously used correctly, that was nonsense. One must not speak of correctness.
    However, it’s more interesting and fun to make up senses of correctness that make it sensible (not necessarily true) to say most words are used incorrectly. Here are a few that occur to me …
    The way words were used in olden days was correct. (It is evident that “correct” is sometimes used this way.) Most words are no longer used as they were in olden days. Ergo: Most words are not used correctly.
    Words change in meaning and pronunciation through the perpetuation of mistakes. (In historical phonology, this is the thesis that regular sound change results from imperfect child acquisition of language.) What is mistaken is incorrect. Words have changed. Ergo: words have come to be used incorrectly.
    In transformational grammar, deep structures more closely represent our logical thought than surface structures. When we vary from speaking in deep structures, we speak less logically and therefore less correctly. We never speak in deep structures. Ergo: we always speak incorrectly. (This is a variation on the phonological analogy I gave earlier.)

  193. One must not speak of correctness.
    I take it that’s your basic position, underlying all the fun suppositions, and I have no quarrel with it.

  194. marie-lucie says:

    GL, Thank you, at last, for a clear explanation of your position, including a definition of the words “correct” and “incorrect” within your analytical framework, hopefully putting an end to the misunderstandings.
    It often happens (it has happened to me) that a person who is working and thinking within a certain framework does not quite realize that what seems crystal clear to them is not so clear to other people. I think that must have been the case here. In any case, I agree with LH’s remark above.

  195. Greg Lee says:

    marie-lucie: In the vast majority of cases, change occurs because of sociolinguistic reasons, related to speakers’ more or less conscious understanding of their place relative to others in a given speech community.
    I don’t agree with this, or at least not the “vast majority” part. In their inception, lenitive phonological changes are not part of speakers’ consciousness at all. American English speakers don’t notice when they are replacing t, d, n by flaps, for instance, or replacing [t] with [?]. I can conceive of them doing such things to mark their solidarity with some social group, I suppose, but it just doesn’t happen that way.

  196. jamessal says:

    so that when you proposed that the majority (or all?) English words are obviously used correctly, that was nonsense.
    But he didn’t propose that; he proposed it was absurd to hold that all words are used incorrectly. Which is not at all the same thing. That said, I too have no quarrel with what Hat’s quoted as your basic position. We’re generally not a quarrelsome bunch here.

  197. Greg Lee says:

    [One must not speak of correctness.]
    Hat: I take it that’s your basic position, underlying all the fun suppositions, and I have no quarrel with it.
    No, I said that’s the conventional position — not mine. Whenever I can choose between what’s conventional and what’s fun, I’ll take what’s fun.

  198. jamessal says:

    Well, GL, I very much doubt anyone else here would share your sense of fun, considering all your syllogisms to be facile attempts at solving a problem that doesn’t exist. You could instead, for example, study semantics in earnest and then share your knowledge with the rest of us. Then we’d all be having fun.

  199. In their inception, lenitive phonological changes are not part of speakers’ consciousness at all.
    Probably true. But why a sound-change starts and why it spreads are two different questions. A sound-change may spread because it helps people differentiate themselves from les autres, even if they are not conscious of just how it does so.
    At layers above phonology, it seems extremely unlikely that some of the changes we see can be entirely unconscious. In the conclusion to the Blust paper, he talks about a morphosyntactic change and a semantic change that cannot reasonably be natural:

    George W. Grace has reminded me that in the Uisai dialect of Buin, a Papuan language of Bougainville Island in the Solomons chain, “we find all the gender agreements reversed – that is, all the masculines are feminine and all the feminines are masculine. There is no accepted mechanism for linguistic change which can cause a flip-flop of this kind and magnitude. I believe that at some stage in the past, some influential speaker of the Uisai dialect announced that from now on his people were not to speak like the rest of the Buins. Once the change was adopted, it would become the natural speech of the community within one or two generations.”
    [...]
    in Iban of southwest Borneo, a number of lexical items have undergone what appears to be a systematic semantic innovation (Blust 1980). Examples include Malay haŋit “foul-smelling” (Proto-Austronesian *qaŋeliC “smell of burnt rice, burning hair, etc.”), but Iban aŋit “fragrance, sweet scent (of flowers, etc.)”; Malay ampul “expand, be blown out”, but Iban ampul “soft, unmuscular (of body), soft (of fruit)”; Malay boŋkar “heaving up, raising up something heavy”, but Iban buŋkar “pull down”; Malay hibur “solace, comfort”, but Iban ibur “shocked, distressed, disturbed”; Malay kampoŋ “cluster of buildings making up a large homestead or a small hamlet” [> English compound], and related words of similar meaning in many other languages of western Indonesia, but Iban kampoŋ “forest”; Malay kilau “luminosity; brilliancy; sheet-lightning” and semantically similar reflexes of *kilaw “reflected light” in many other languages, but Iban kilau “dusk”; Malay liut “lithe; leathery; tough”, but Iban liut “soft, silky”; Tagalog namnám “sense of taste” and reflexes of *ñamñam “tasty, delicious” with similar semantics in many other languages, but Iban ñam-ñam “tasteless, insipid”.
    In general, semantic changes are notoriously random and unpredictable, and it is surprising to encounter a case like this, where it can be argued that the meanings of a number of lexical items (seventeen in all) have been ‘reversed’ in some culturally meaningful sense of semantic reversal.

  200. Greg Lee says:

    marie-lucie: I think that you will find very few texts (except for the spam obviously written by non-native speakers) where even a small minority of words are “used incorrectly”.
    For a certain kind of “incorrectness” which descriptivists and prescriptivists seem to mostly agree about, I think this is right. Consider, for instance, this example I just ran across:
    Why doesn’t Canada, or Japan, or England, or a host of other countries don’t have this problem with a fascination for the gun culture?
    The “doesn’t” and “don’t” here are used incorrectly, probably most would agree, and this sort of mistake, due to inattention, is rare in texts.

  201. Less rare these days, when you write a sentence on a device and then decide to rewrite but miss some bit.

  202. The way words were used in olden days was correct. (It is evident that “correct” is sometimes used this way.) Most words are no longer used as they were in olden days. Ergo: Most words are not used correctly.
    I covered that one way back when I talked of the urlanguage.
    Words change in meaning and pronunciation through the perpetuation of mistakes.
    Same as the above, but with the addition of a dynamic process or cause for the incorrectness of present-day language. (Also a ‘reason’ for saying that present-day language is incorrect.)
    In transformational grammar, deep structures more closely represent our logical thought than surface structures.
    So ‘deep structure’ takes the place of ‘prescribed forms’ or ‘correct forms’? Anyway, how is it possible to speak in ‘deep structures’? Deep structures are by definition not the form that language takes in use, which is ‘surface structures’. In fact, there is likely to be more argument among linguists about the correct deep structure than there is about surface structures.
    GL is just playing around with the word ‘correct’, having ‘fun’ as he calls it. There’s nothing there. The problem is, as m-l and I pointed out at the start, is defining what we mean by ‘correct’. We haven’t moved on significantly from that problem.
    You’ll also note that GL continually brings in phonology and syntax, avoiding the central issue here, which was the meaning of words. Now why is that? Again, I would suggest it’s because there’s nothing there. Just having ‘fun’ as usual. The more the merrier.

  203. Greg Lee says:

    marie-lucie: As for “phonemic pronunciation = correct pronunciation”, such a statement assumes that a phoneme (a) is always strictly determinable and (b) optimally represented orally by a single specific sound. Neither of these general statements is true.
    No, it doesn’t assume those things. Phonemicizations might not be unique, or might not be known, and there might be several correct pronunciations, or what is correct might not be known. In my opinion, my analogy stands.

  204. To add to the above, ‘correctness’ is adherence to some standard. GL has proposed at least two standards:
    * The way words were used in olden days
    * Deep structures
    The first is the related to the idea that everything was perfect in the beginning and has decayed since. You can take that or leave it as you like. Some civilisations (like the Chinese) were based on that line of thinking.
    The second is a grammatical theory that no one agrees on. In fact, I thought deep structure was dead. It’s nonsensical to talk of ‘correctness’ here since we don’t even have a standard to go on.
    It all comes down to what standard of ‘correctness’ we adopt. You can adopt whatever standard you like. Take your pick. You could make up an arbitrary rule that languages should not have more than ten vowels. Any languages exceeding that number are ‘incorrect’. It’s all good fun.

  205. The “doesn’t” and “don’t” here are used incorrectly, probably most would agree, and this sort of mistake, due to inattention, is rare in texts.
    On the contrary, I think that poorly-formed sentences are surprisingly common in language. There are plenty of people who write as they speak. Read their prose (which is more and more common in the days of the Internet since everyone is an author) and you’ll often encounter poorly structured sentences, structures run together, etc.

  206. Greg Lee says:

    Bathrobe: Anyway, how is it possible to speak in ‘deep structures’? Deep structures are by definition not the form that language takes in use, which is ‘surface structures’.
    This is a specious argument. There is no requirement in TG that every deep structure should have some obligatory transformation applicable to it. Then, consider some deep structure to which no transformation need apply, and that will also be a surface structure, and will be spoken. So it is, after all, possible to speak in deep structures, since they may coincide with surface structures.

  207. Greg Lee says:

    Bathrobe: In fact, I thought deep structure was dead.
    I think TG is dead, but I also think that it’s okay to invoke it here, because it’s not dead in a relevant sense. I could have constructed the argument using a non-transformational theory I know about, GPSG (Generalized Phrase Structure Grammar), because there are analogs in that theory to transformations, namely, metarules. But I just don’t know whether GPSG is still considered current, even on the West Coast, and I don’t know how familiar it is to those here, either.

  208. This is a specious argument.
    On the contrary, as you say, ‘it is … possible to speak in deep structures, since they may coincide with surface structures.’ In other words, no one is speaking in deep structures. They are speaking in surface structures, which happen in some cases to coincide with deep structures.

  209. I can only see one useful result of Greg Lee’s assertion that it’s not absurd to maintain that all words are used incorrectly: we’ve been taken away from the terms of the narrow prescriptivist/descriptivist debate. Since it’s possible to see correctness in language in wider terms than just a squabble over certain arbitrary rules set up in the 18th century, this is perhaps a useful outcome.

  210. “phonemic pronunciation = correct pronunciation”, such a statement assumes that a phoneme (a) is always strictly determinable and (b) optimally represented orally by a single specific sound.
    I think m-l is correct in making this criticism. You are taking for granted a phonological ‘deep structure’, but how can you have a deep structure when the ‘surface structure’ hasn’t even been figured out, as m-l points out?
    Deciding what is correct ‘deep structure’ is to accept someone’s analysis as standard. With virgin languages even setting up a standard analysis of the surface structure can be difficult.
    That is why I was so scathing about adopting TG syntactic ‘deep structure’ as a standard.

  211. An example of a sentence that is strictly speaking incorrect according to the rules of written English but is perfectly fine in speech (except for the verb ‘keeps’):
    I would add that in your divorce papers, any debts (mortgage, credit card, etc) that you cannot afford to pay but keeps you in your home, financially stable, etc., you want to make sure you create a category Support Obligations and place those debts under that in your decree.
    This is akin to topicalisation in Japanese and isn’t particularly ‘wrong’ in the greater scheme of grammar. It’s certainly not as bad as some examples I’ve seen but neglected to note at the time.

  212. Greg Lee says:

    Bathrobe writes: In other words, no one is speaking in deep structures. They are speaking in surface structures, which happen in some cases to coincide with deep structures.
    You can’t be serious. If you say a form which is identical with a deep structure, you’ve said a deep structure.
    On topicalization, (1) there are no “rules of written English” that would be recognized by any serious linguist, and (2) there’s nothing wrong with topicalization in English — it’s been described and discussed by many grammarians.

  213. Of course I’m serious.
    I used the expression ‘speaking in deep structures’. You said ‘said a deep structure’. This is the kind of confusion that seems to dog this discussion. (See earlier comments about the correct ‘usage’ of words that kept veering off without explanation into a discussion of the correct ‘pronunciation’ of words.)
    The fact that a person every so often ‘says’ a deep structure does not mean they are ‘speaking in’ deep structures. I hope you can see the difference.
    To use a Lee-esque analogy, I might be writing English full of Latinate words. The fact that there are many words that are recognisably ‘Latin’ does not mean that I am writing ‘in Latin’. I am writing ‘in English’ that happens to have many Latinate words. Yes, the analogy is imperfect, but that is the nature of this kind of analogy.
    ‘Written English’ is not a topic that linguists are particularly strong at, unlike old-fashioned grammarians, who had their rules (often prescriptivist in nature, but not necessarily stupidly prescriptivist) about what is acceptable in writing. I’m perfectly aware that topicalisation is found and has been covered in English. The fact remains that the sentence I quoted would, I suggest, be regarded (not necessarily by linguists) as less than optimal written English.

  214. Now that we’ve wandered into descriptivist/prescriptivist territory again, I wonder how Hat would see the sentence I gave. Linguists say they give priority to the spoken language, but usually (not always, but generally) confine themselves to ‘standard spoken English’, which makes things a lot simpler. The sentence I gave would be ok as spoken English and wouldn’t be tagged as ‘incorrect’ by a descriptivist linguist (except for the ‘keeps’), but Hat is professionally engaged in the editing line of work. As a ‘descriptivist’ I’m sure he has no disagreement with the sentence, but I wonder whether he feels it passes muster as ‘written English’…
    Incidentally, slips like ‘keeps’ or ‘doesn’t’ above are dealt with in terms of ‘competence’ and ‘performance’ in Chomskyan linguistics. People don’t always produce perfect English and slips occur all the time, which is accounted for in terms of faulty ‘performance’, not as a problem with their ‘competence’ (i.e., their knowledge of what is ‘well-formed’ in English). Hehe, in an effort to get away from that word ‘correct’, TG came up with the term ‘well-formed’. It seems hard to get away from the concept.

  215. Greg Lee says:

    Bathrobe: The fact that a person every so often ‘says’ a deep structure does not mean they are ‘speaking in’ deep structures. I hope you can see the difference.
    This is progress. Now, if you can see how a person could sometimes speak a deep structure, you’re just a step away from the point. How often a person would be predicted to say a deep structure would depend on the specific properties of the transformational grammar. Start removing transformations to create new grammars — as there are fewer transformations, more and more deep structures will be sayable, because fewer transformations will be applicable. Eventually, most deep structures will be sayable. When all the transformations are gone, all deep structures will be sayable. See?

  216. This is progress
    Why couldn’t you see the difference before?
    When all the transformations are gone, all deep structures will be sayable.
    And this sums up your argument that all English words are used incorrectly? That is your point? That we should all be speaking in some mythical ‘deep structure’? I can see why you so breezily dismissed m-l’s comment on phonemes in surface structure. The abstraction appears to take precedence over concrete reality. So this is your idea of descriptivism?

  217. marie-lucie says:

    GL: If you say a form which is identical with a deep structure, you’ve said a deep structure.
    No one can “say” a deep structure. A deep structure is abstract, anything actually said or written may have a surface structure which closely corresponds to its deep structure, but it cannot “be” a deep structure. Similarly, you cannot “say” a phoneme, you can only “realize” it through physical manifestations which may vary more or less substantially from each other as long as they conform to the most important features of the idealized prototype.
    Deep structures are like patterns or blueprints: you can make any number of clothes, boats, buildings, etc using the same patterns or blueprints, which are not the actual physical objects made in conformity with them.

  218. marie-lucie says:

    Bathrobe: in an effort to get away from that word ‘correct’, TG came up with the term ‘well-formed’. It seems hard to get away from the concept.
    “Well-formed” refers to the conformity of a given clause, sentence, etc with a consistent, intrinsically motivated abstract framework, while the old ‘correct’ concept often appears arbitrary and extrinsically motivated (especially the “zombie” rules formulated in past centuries in order to – more or less explicitly – purge the English language of features which make the language particularly different from Latin).

  219. Are we having fun yet?

  220. Greg Lee says:

    marie-lucie: A deep structure is abstract, anything actually said or written may have a surface structure which closely corresponds to its deep structure, but it cannot “be” a deep structure. Similarly, you cannot “say” a phoneme, you can only “realize” it through physical manifestations …
    I’m familiar with this view, though I don’t share it. Stratificationalists (inspired by earlier similar views) think that phonemic/deep/underlying forms are different types of things from pronunciations, and that it is a “category error” to treat them as though they were the same type. I’ll just ask you to recognize that not everyone believes this, and, specifically, it is not part of TG or of generative phonology. For the latter, underlying phonological forms are represented as feature matrixes and so are surface forms. They differ in detail, but not in type, and so it makes perfect sense to suppose that people could pronounce phonemic (underlying) forms in those cases for which there happen to be no obligatory phonological rules applicable to them.

  221. Ø: Great fun. It’s a barrelful of laughs. Why not join in?
    Thanks m-l for that clarification. When GL seemed to be saying that ‘allophones’ are ‘incorrect’ from the point of view of ‘phonemes’, or that surface structures are ‘incorrect’ from the point of view of deep structure, I could see something incongruous but couldn’t quite put my finger on it. You’ve nailed it down. We’ve had ‘fun’ getting here, but at least now we know where GL is coming from.

  222. marie-lucie says:

    GL, I have no problem “recognizing that not everyone believes” what I believe or what I think I understand. I never said that I agreed with all the details of TG or of generative phonology. I especially don’t believe in “feature matrixes”, which strike me as extremely artificial and of limited usefulness.

  223. A “category error”?
    I challenge you to ‘say’ those deep structures found in the TG literature (apart from the ones that undergo no transformations). And as usual you’ve slid off from addressing issues of syntax and word meaning and brought us back to phonology.

  224. Greg Lee says:

    Bathrobe: I challenge you to ‘say’ those deep structures found in the TG literature …
    I can’t say them. You’re losing track of the point. Recall, I said above: “We never speak in deep structures.”
    I’m following James McCawley, here, essentially, who proposed that the function of transformations is to convert logical forms into linear sequences of words that exist in our language. Logic is not inherently linear, for one thing, but we humans are condemned to saying words one after the other.
    So, you see, the idea is not to prove that we say deep structures, but rather, recognizing that we don’t, to seek some explanation for why that is.

  225. I’m having trouble keeping track of ‘the point’.
    1. In transformational grammar, deep structures more closely represent our logical thought than surface structures. When we vary from speaking in deep structures, we speak less logically and therefore less correctly. We never speak in deep structures. Ergo: we always speak incorrectly.
    2. There is no requirement in TG that every deep structure should have some obligatory transformation applicable to it. Then, consider some deep structure to which no transformation need apply, and that will also be a surface structure, and will be spoken. So it is, after all, possible to speak in deep structures, since they may coincide with surface structures.
    3. Now, if you can see how a person could sometimes speak a deep structure, you’re just a step away from the point. How often a person would be predicted to say a deep structure would depend on the specific properties of the transformational grammar. Start removing transformations to create new grammars — as there are fewer transformations, more and more deep structures will be sayable, because fewer transformations will be applicable. Eventually, most deep structures will be sayable. When all the transformations are gone, all deep structures will be sayable.
    4. the idea is not to prove that we say deep structures, but rather, recognizing that we don’t, to seek some explanation for why that is.
    And I thought the point all along was that “We never speak in deep structures. Ergo: we always speak incorrectly.”
    Can you please tell us where you are coming from? Are you arguing against transformations or against the model used in transformational grammar? Or are you simply saying that language in use is ‘incorrect’ since it is ‘illogical’, based on the idea that deep structure reflects ‘logic’ (which for me suggests Platonic overtones)?
    If the last stab in the dark is true, then your two examples of language being ‘incorrect’ boil down to:
    1. Language in use was correct in the Golden Age but has decayed in modern times.
    2. Language is incorrect since it is an imperfect reflection of logic (= deep structures), which is the golden standard of correctness.
    Am I getting you right? If not, please stop having ‘fun’ and tell us what your ‘point’ is.

  226. Greg Lee says:

    Bathrobe: Am I getting you right?
    Yes, you are. I’m saying these are plausible ideas about correctness which make sense of the proposal that most words are used incorrectly. (I’m not saying they’re true, however.)

  227. Your ‘point’ would have been a lot clearer if you had pointed out from the outset the premise of your arguments, which was, ‘If you follow McAwley, who proposed that the function of transformations is to convert logical forms into linear sequences of words that exist in our language…’. That would have saved us from floating a sea of assertions that made little sense without context.
    In the end you finally came out and said, ‘I’m not saying they’re true, however’. So you were arguing for someone else’s position without telling us what their position was, and every time we got it wrong you told us we didn’t see the point! How could we? You really should learn to clarify where you’re coming from.
    As it is, McCawley’s view (which is not standard TG theory, merely his own surmises) is quite Platonic in nature. It ascribes a greater reality to ‘deep structure’ than it does to surface structure, which is (if you don’t mind me putting it that way) just a shadow of the deeper reality dancing on the wall of the cave. Well, you might accept that as a solid basis for descriptive linguistics, but I doubt that most serious linguists would see it that way.

  228. ‘floating in a sea of assertions’

  229. marie-lucie says:

    Bathrobe, your patience while floating in this sea of conflicting assertions was admirable.

  230. jamessal says:

    Wow. That’s a hell of a compliment coming from marie-lucie (not to mention one with which I wholeheartedly agree), ML being probably the most patient person I’ve ever met.

  231. marie-lucie says:

    Thanks jamessal! I haven’t always been very patient but had to learn, often the hard way. In this particular case, I gave up some time before Bathrobe did. So the prize for patience goes to Bathrobe.

  232. On the contrary, I was highly impatient to figure out what he was talking about and if there was anything really there. The only ‘patience’ involved was that of continuing to give him the benefit of the doubt.

  233. marie-lucie says:

    Bathrobe, perhaps “perseverance” would be a better word? I admire it anyway.

  234. David Marjanović says:

    it just sounds stilted (or British).

    So you do claim it sounds incorrect in many situations (just not all of them).

    As for “phonemic pronunciation = correct pronunciation”, such a statement assumes that a phoneme (a) is always strictly determinable and (b) optimally represented orally by a single specific sound. Neither of these general statements is true.

    Examples for (b) are easy to find, BTW: Is the abovementioned Spanish b/v a plosive that turns into an approximant in many situations, or an approximant that comes out as a plosive if you happen to have had your mouth closed just before? Is the German ch a /x/ that comes out as [ç] half the time, or a /ç/ that comes out as [x] half the time (except in places that extend the range to [χ])? All of these options have, AFAIK, been argued.
    Examples for (a) are perhaps a bit less obvious, but still easy to find: In English and/or German, is [ŋ] actually an allophone of /h/? They’re in complementary distribution after all, aren’t they; this has been seriously claimed. And Mandarin… most sounds occur in so few specific environments that the phrase “allophonic Möbius loop” has been used and that one study has claimed Mandarin has just a single vowel phoneme.

    It doesn’t really matter if Caesar did it

    Oh, that reminds me: I’ll never forget discovering druidae [...] habent instituta sacrificia in Commentarii de bello gallico. That’s the Standard Average European (present-)perfect tense – unlike in Caesar urbem occupatam habet, no alternative interpretation is possible –, which is so not supposed to show up in Classical Latin.
    (This fits the hypothesis that the SAE perfect comes from an Ancient Greek construction that involved the aorist participle… but I digress.)

    Damn, that’s description at its best ! I would prescribe a helping for everyone.

    Strawberries are deeply disgusting, no matter how prepared.
    Point being:

    if you describe the best ice cream you currently have on offer

    He can’t “describe the best ice cream”, because there is no such thing. He can only describe the one he finds best, because he doesn’t know your taste in enough detail. Therefore, de gustibus…

  235. Fine, I’ll have your helping of strawberry ice cream, then.

  236. David, there is in fact an alternative reading of instituta as a noun, the plural of institutum ‘custom, institution’, rather than a feminine participle agreeing with natio, the subject of the whole sentence. Here’s the actual sentence from 6.16:

    Natio est omnis Gallorum admodum dedita religionibus, atque ob eam causam, qui sunt adfecti gravioribus morbis quique in proeliis periculisque versantur, aut pro victimis homines immolant aut se immolaturos vovent administrisque ad ea sacrificia druidibus utuntur, quod, pro vita hominis nisi hominis vita reddatur, non posse deorum immortalium numen placari arbitrantur, publiceque eiusdem generis habent instituta sacrificia.

    I render it thus in accordance with the above theory:

    The entire nation of the Gauls is very subject to superstitious rituals, and for that reason those who are afflicted with very serious illnesses, or who are involved in the perils of battle, either burn human sacrifices or vow that they will be burned, making use of the Druids as ministers for such sacrifices, because they believe that if a human life is not given in return for a human life, the anger [lit. power, sway, will] of the immortal gods cannot be placated; and in public affairs the nation has sacrificial customs of the same kind.

    As for Mandarin vowels, I don’t see any way to reduce it further than a fundamental opposition between /a/ (Pinyin a), and /ə/ (Pinyin e, o), and to do that we have to postulate a third or empty vowel, sometimes realized as the syllabic forms of the glides /j/ (Pinyin i, yi), /w/ (Pinyin u, wu), /ɥ/ (Pinyin ü, yu, u) and sometimes as the syllabic form of various fricative consonants (Pinyin i).

  237. David Marjanović says:

    …Thanks, very interesting. But why do you read sacrificia as a noun (neuter plural) the first time and an adjective the second time?
    (I didn’t think instituta agreed with natio, but with this neuter plural sacrificia. The plural habent clearly does not agree with natio. The subject of almost everything after the first comma are the two occurences of qui.)
    Mandarin vowels: I get pretty far by assuming palatalized and labialized consonants (both of which aren’t really there on the phonetic level), but not down to one vowel either. I’ve never seen the paper and have no idea what exactly it proposes.

  238. an adjective the second time?
    Well, why not? Pretty much any adjective, or at least any normal adjective, can be a noun at will in Latin. You’re right about habent, but I don’t think that the final clause refers solely to “those who are afflicted etc.”: I think it is notional agreement, just as in the other thread, taking natio as here referring to all the members of the nation, and not just those who engage in private human sacrifice.
    I get pretty far
    Well, you can read le, luo and lie with normal, labialized, and palatalized initials respectively, followed by a schwa vowel. But then what of lüe? Are you going to introduce a partial third series of consonants that have /ᶣ/ in them, and so are both labialized and palatalized? That seems to be stretching unnatural phonology pretty far.

  239. David Marjanović says:

    That seems to be stretching unnatural phonology pretty far.

    Definitely; but you can find it in a few places in the Caucasus.
    This would still leave lu and ; distinguishing a schwa from the absence of a phonemic vowel takes it really far… but it would allow us to account for la vs. le and for li vs. lu vs. !

  240. David Marjanović says:

    But it doesn’t seem to explain why *lia doesn’t exist; after all, jie and jia do.

  241. Well, liǎ 倆 ‘two’ does exist colloquially, usually heard (in my experience) as a part of zánliǎ 咱倆 ‘the two of us’, but I guess that’s really just a contraction of liǎngge 兩個. And the zán of zánliǎ always sounds more like zám to me, anyway, which is totally a violation of Mandarin phonology. If I’m hearing it right, I guess that’s probably a contraction, too, of zánmen 咱們 ‘we, us’.

  242. My point being that it exists, but I don’t know that it counts for these purposes.

  243. Missing syllables in Mandarin are weird anyway. The syllable seng should be completely normal in Mandarin, but there are only two syllables with this pronunciation.
    One, the obscure sēng 鬙, shows up out of nowhere in the Tang dynasty, as far as I can tell, as part of the disyllabic compound péngsēng 鬅鬙, which seems to have originally meant something like ‘messy or matted hair’, and picks up a very limited amount of other meanings in compound words over the years. The other is sēng 僧 ‘Buddhist monk’, a borrowing from the first syllable of the Sanskrit saṃgha (the full rendition is sēngjiā 僧伽), unless it comes from Pali or mediated through another language.
    I don’t know the etymology of péngsēng, but it seems likely to be a foreign borrowing, too. Its first appearance that I’ve found (not that I’ve searched too exhaustively) is in Youyang zazu 酉陽雜俎, a ninth-century collection of fantastic stories from an area that to this day contains many non-Han people (and a collection, at that, which includes a version of the Cinderella story, which presumably came from further west). So I’d guess that this very normal-sounding Mandarin syllable only occurs in two foreign borrowings from the first millennium. For whatever that’s worth.

  244. In this view, li lu lü have the glides /j w ɥ/ followed by the empty vowel, which stretches them into full vowels, in the same way that PIE /i u/ are stretched versions of /j w/. The partition a/ə/empty is complete, given this understanding.
    But no, it doesn’t account for the gaps. Lia is not one of them, however: see the Mandarin syllable chart. In fact, Mandarin phonology is so overconstrained that there is little or no evidence to tell one system from another, so we can adopt a five-vowel or six-vowel or three-vowel system pretty much ad libitum, and we can treat j q x as separate phonemes or not, also ad libitum.

  245. I think the sorry fool that you believe to have been “decimated” by John McIntyre was, in fact, me.

    Until now I have not bothered to respond to his response to comments I made on the Carpe Diem blog because it seems generally impossible to get the devotees of common usage to understand that there principal robs language of the possibility of precise communication.

    The reality, however, is that almost everyone speaks several dialects, for example, that which one might use in speaking with the guys down the pub, versus that which one would use writing an application for a faculty position at, say, Harvard. Among ordinary folks, “decimate” may mean “trash” or something like, but to a historian it means, at least in certain contexts, “select one in ten and execute.”

    The silliness of the assumption that “common usage” holds sway in all circumstances was debunked in a seemingly convincing way by Betrand Russell in a paper entitled the Cult of Common Usage.

    Among objections to insistence on the universal adherence to “common usage” Russell enumerates the following:

    (I) It is insincere;

    (2) It is capable of excusing ignorance of mathematics, physics, and neurology in those who have had only a classical education;

    (3) It is advanced by some in a tone of unctuous rectitude, as if opposition to it were a sin against democracy;

    (4) It makes philosophy trivial;

    (5) It makes almost inevitable the perpetuation among philosophers of the muddle-headedness they have taken over from common sense.

    It would be interesting to read John McIntyre’s or anyone else’s attempt to rebut the arguments by which Russell justified these assertions.

  246. Oops: “their principle” not there’s!

  247. CanSpeccy, it’s the notion that there’s something wrong or vague or inherently confusing about common vs. technical usage to which the Hattics (for the most part; we are not a monolith) object. To take a more down-market example, to a plumber a lavatory is a specific type of sink, the kind you wash your hands in (as opposed to the dishes or the laundry). Etymology is on the plumbers’ side here. Yet we do not hear plumbers complaining that it is impossible for them to be clearly understood because most people nowadays use lavatory as a fancy term for bathroom/toilet.

    As for the Russell article: Russell was not complaining about ordinary usage as such, nor to the investigation of language as she is spoke (nowadays territory shared between philosophy and linguistics), but to the claim that the latter activity was the whole of philosophy.

  248. I might also point out that you wrote “principal” when you meant “principle”; I am not in the habit of calling attention to typos and “incorrect” usages, but since this is exactly the sort of “mistake” that persons of your persuasion tend to get upset about and claim is destructive to clarity, it is worth noting that I had no problem understanding what you meant, any more than I (or, presumably, you) have a problem understanding what people mean when they say or write “decimate.” If communication is not impaired, what exactly is the problem?

  249. In addition, Russell is decidedly unfair to the shopkeeper’s colleague, I think. The latter was concerned to nail down the meaning of the question “What is the shortest way to Winchester?” so that he could decide whether he could answer it or not. As it turned out, he could not, but that does not exclude the possibility that someone else could. It was not a matter of his taking “no interest”, as Russell claims, in the question. Indeed, the question seems odd: if asked the “shortest way” from my house to Times Square, I would be hard put to it to answer correctly; though I could easily state a reasonably short route, I would hesitate to claim that it was precisely the shortest. (Looking at Google Maps, it isn’t.)

  250. Oh goodness.

    Let me say that again with fewer typos, lest you think I am a person of a very peculiar persuasion. In fact, perhaps you would be kind enough to delete the first attempt.

    but since this is exactly the sort of “mistake” that persons of your persuasion tend to get upset

    You seem a bit defensive!

    Since I make such mistakes as a matter of course and only correct them on proof reading — if and when I do proof read — why would you assume that I am of a persuasion that tends to get upset by such errors?

    I learnt (past participle of learn, which for some reason your blog software highlights) to read phonetically, and thus tend to write phonetically, although I can spell most words correctly if I stop and think, as I did in the case of their/there.

    Reflecting on the question of common usage further, it occurs to me that most arguments in this connection arise from the sensitivity of people of your persuasion, whatever that may be (or of most other persuasions, for that matter), to criticism of their language use. For most people language use is their most important skill: it largely determines academic achievement, especially if one treats mathematics as a form of language, as well as success in love, business and politics. Therefore, to have it pointed out that one did not know the alternative meanings of a term such as “decimate” is a humiliation to many, which is most easily rebutted, as Russell points out, by arguing that one’s own use, i.e., common use, is the only correct use, and anyone who things otherwise is an ignoramus or a prig.

    Whether that is what you are saying, seems hard to guess. But it would be interesting to know, quite specifically, whether you agree or not with my contention that it is silly, and untrue, to assert that “common usage” should hold sway in all circumstances.

  251. It’s the fact that your browser is set for American English that causes “learnt” to be highlighted; it is nothing to do with this blog in particular.

    As for our host, he is about as sensitive to criticism of his usage as a rhinoceros, without being in the slightest unwilling to learn from actual, as opposed to self-appointed, experts.

  252. Since I make such mistakes as a matter of course and only correct them on proof reading — if and when I do proof read — why would you assume that I am of a persuasion that tends to get upset by such errors?

    Of course you don’t get upset by such errors when you make them. You, after all, are the pinnacle of perfection and a fount of infinite knowledge about English; it would be absurd to think that a mere typo said anything at all about your literacy or intelligence! When other people make them, however, it is a sign of the degeneracy of the modern world, loss of standards, etc. etc.

    You seem a bit defensive!

    Not at all, just amused.

    Therefore, to have it pointed out that one did not know the alternative meanings of a term such as “decimate”

    But I know it perfectly well; I just happen also to know, unlike you, that it is irrelevant in the modern world except in discussions of Roman history.

    But don’t let me stop you; if you enjoy galloping around waving your sword and smiting the windmills of degenerate illiteracy, by all means knock yourself out! There are definitely worse ways of amusing oneself and passing the time.

  253. Like I said, a rhinoceros. Not that actual rhinoceroses have particularly insensitive skin, any more than door-nails are the deadest nails around, “but the wisdom of our ancestors is in the simile; and my unhallowed hands shall not disturb it, or the Country’s done for.”

    However, evidently we must extend the scope from Roman military history to military history in general, given the instances of (technical) decimation people have reported above.)

  254. you, after all, are the pinnacle of perfection and a fount of infinite knowledge about English; etc., etc.

    It seems odd that you engage in what seem to me to be rather lame put-downs, but evade the specific question I raised; namely, whether or not it is both silly and untrue to assert that “common usage” should hold sway in all circumstances.

    But this is your blog so sneer away as you please at any who might agree with me that it would be conducive to clear communication if those who mean “devastate” said so, rather than saying “decimate,” a word that is useful in the sense of kill one in ten pour encouragez les autres (a view I have elaborated on here) — and a view that does not, obviously, deny the validity of common usage in common speech, where precision is rarely of much importance.

  255. whether or not it is both silly and untrue to assert that “common usage” should hold sway in all circumstances.

    Of course it’s silly and untrue, but then again, nobody holds it, so why shouldn’t it be? it’s a strawman (by which I do not mean a trial balloon).

    it would be conducive to clear communication if those who mean “devastate” said so, rather than saying “decimate,”

    In what context? Saying an army was decimated is obviously a context where that had better mean “punished by the (possibly capital) punishment of one man in ten” and not merely that it was devastated (by another army, say). But it would not be particularly conducive to clear communication to insist on the distinction in other contexts, because it is already clear enough for any reasonable person.

  256. Saying an army was decimated is obviously a context where that had better mean “punished by the (possibly capital) punishment of one man in ten” and not merely that it was devastated (by another army, say).

    “That had better mean”? LOL

    We evidently agree that correct usage is context dependent, but you rather spoil the argument by your inappropriate example.

    In fact, and I would have thought rather “obviously,” in common parlance to say that an army was decimated actually means that the army was “devastated.”

    And that is an empirically verifiable assertion, as you can see from a Google phrase search. For example, the first nine Google results for the phrase “Hannibal decimated” all appear to express the idea that Hannibal devastated, destroyed or otherwise crushed, tribes, legions or armies. In no case, does it appear to mean that Hannibal resorted to the exemplary punishment of executing one person in ten.

    So common usage is often incorrect, or at least inexact or ambiguous. However, in common speech such inexactitude is rarely inconvenient since we usually sense the intended meaning of those with whom we are communicating, and even if we don’t, it doesn’t usually matter a whole lot.

    But that’s not to say that common usage can or should dictate correct usage, which it does not, if by “correct usage” we mean that usage which enables the most precise and effective communication among those with relevant expertise to benefit from and who desire precise communication.

    And correct usage may also be defined not by its precision but its propriety. There are different ways to greet the Pope, a mate down the pub or your dog. And correct usage is sensitive to context in many other ways.

    PS: I expect you;re right about my browser, not the blog software, looking askance at my correct use of “learnt.”

  257. Therefore, to have it pointed out that one did not know the alternative meanings of a term such as “decimate”

    But I know it perfectly well; I just happen also to know, unlike you, that it is irrelevant in the modern world except in discussions of Roman history.”

    You’d certainly be a fool if you didn’t know it, since you’ve been discussing it for however many years on this thread if not before.

    But what you seem not to know and to find too painful even to contemplate is the idea that correct usage is dictated not by common usage but by context, which includes many and often subtle factors including the relative status of those in communication, the occasion, and what the speaker intends to achieve, whether to precisely inform, to amuse, to deceive ….

    But you evidently disdain for such subtleties.

Speak Your Mind

*