Do you speak Canadian?

« previous post | next post »

Flash! From the Toronto Star on 2 June: "Language test spells trouble for newcomers", in which Lesley Ciarula Taylor (the Star's immigration reporter) tells us that all immigrants to Canada would soon be required to take a specific "rigorous language test", the International English Language Testing System (IELTS) exam, widely used in Britain and Australia and already used in Canada for foreign students seeking to go to Canadian universities.

This much is accurate. But the story leads off with an especially tricky grammar question:

Think you speak English? Try this test.

Find the grammatical (or syntactic) error in this sentence: The standard of living has increased.

Stumped? Soon, that will count against you if you're hoping to immigrate to Canada. The rigorous language test that will be a requirement is vital to be fair to the influx of newcomers or vastly discriminatory and fatally flawed, depending on whom you talk to.

The correct answer is: The standard of living has risen.

And that, as it turns out, is just wrong. I wasted considerable time trying to find this sample question on the IELTS site, until I realized that there weren't any grammar questions at all on the exam. Then, illumination from Brett Reynolds (Professor of English for Academic Purposes at Humber College) on his English, Jack blog the same day, under the heading

Language tests for immigrants & Honesty tests for newspapers

Yes, more reportorial mischief.

(Hat tip to Randy Elzinga.)

Brett asked about the standard-of-living question on a mailing list, and Lynda Taylor of University of Cambridge ESOL Examinations replied that

IELTS does not have (and has never had) a distinct section testing explicit grammatical knowledge …

The grammar question example given in the Toronto Star article does not come from an IELTS test paper nor does it come from the IELTS Official Practice Materials. Instead, it seems to have been taken from another source entirely – probably one of the many test preparation coursebooks produced by publishers around the world.

(The offending test preparation book has not yet been identified.)

Meanwhile, the Star did a follow-up mocking story on the 3rd ("Daring to ask the fluff questions", by Vinay Menon), and people (mostly Canadians) began to complain angrily, in a variety of forums, among them Digg ("Do you know English?"), The Impudent Observer ("Could You Pass Canadian Immigration Test?"), and the Star's own letters page ("Test must be from another planet"). Yahoo! Answers introduced the increase/rise question as: "Why is the first sentence wrong and the second correct?", which of course invited respondents to accept the premise of the question and try to justify rise over increase, and some (but not all) of them did. Commenters on other sites almost all said that there was nothing wrong with increase. (I'll get to actual usage in a moment.)

And there were objections to the once-size-fits-all policy, and to the use of a British-made test.

Then on the 4th, Lesley Ciarula Taylor reappeared, with the story "Ottawa drops English exam" (subhead: "Move on immigrant test follows Star story"). The immigration section of the Canadian Bar Association had objected strongly to the change in policy, and the government relented. Taylor goes on:

… the status quo remains and prospective immigrants can produce their own documents to prove how well they speak English, or take the IELTS test [or a French equivalent].

What about the syntax of the increase/rise sentences? A prohibition against increase is news to me; I can't find it in any advice manual, and don't recall anyone having mentioned it to me. The increase version sounds fine to me (as it does to so many baffled Canadians), as does the rise version. You can google up plenty of examples of increase in combination with standard of living in serious writing, including some from British sources (like the Daily Telegraph and the Guardian). One of NOAD2's definitions for rise in fact glosses it in terms of increase — 'in number, size, amount, or quality'.

But another of the uses of rise is in the sense 'improve', as in NOAD2's example living standards have risen (though this sense isn't easy to distinguish from the 'increase in quality' subsense). This use is remarkably restricted; all sorts of things that can be said to have improved — health, fitness, eyesight, disposition — cannot be said to have risen. That is, it's easy to find collocations of rise 'improve' with standard(s), but not with other nouns. My hypothesis is then that it's not so much that standard (of living or whatever) requires rise, but that rise 'improve' requires standard. Maybe someone got the relationship backwards.

Otherwise, both rise and increase can be used to mean 'increase', qualitatively or quantitatively. There might be some preference to use rise for qualitative increase and increase for quantitative increase, but the usages can overlap, as in these two different descriptions of the same study reporting a quantitative increase:

Her results, which were based on California statistics, showed that in the first year after divorce, the male standard of living increased 42%, … (link)

… in the year following divorce, women and children underwent a 73% drop in their standard of living, while men's standard of living rose by 42%. (link)

The two versions are semantically equivalent, but differ slightly in tone: increased is simply literal, while rose has vestige of a metaphor left in it. So some writers might prefer rose as a bit more vivid — but that doesn't make increased syntactically unacceptable.




  1. Jonathon said,

    June 4, 2008 @ 8:02 pm

    Just recently I heard an opposite prescription—that rise can only be used to refer to physical location, while metaphorical uses had to resort to increase. Thus, prices don't rise; they increase. I was just as baffled by that usage advice as I am by this, and it seemed just as baseless.

  2. JJM said,

    June 4, 2008 @ 8:45 pm

    It's The Toronto Star for God's sake.

    A rag of a paper…

  3. Robert said,

    June 4, 2008 @ 9:21 pm

    I am also baffled that anyone could object to the use of increase. However, the source of the notion may be that increase can be used transitively whereas rise cannot. Some people assert that meanings must be in bijective correspondence with words, and so cannot accept that increase, used intransitively, and rise mean (nearly) the same thing.

  4. Tom said,

    June 4, 2008 @ 10:39 pm

    If I strip away the "of living", I prefer "risen" to "increased". To "increase" a standard sounds odd to me & suggests that the scope of the standard has changed rather than the level of the standard (i.e. that we're measuring more, not expecting more). With the full phrase (standard of living) I don't notice the oddity. This makes this similar to other cases where prescriptivist usage has us ignore the larger NP (and me vs and I is a striking example; [sing noun] of [plural]… [ver-in-plural-form] is another)

  5. Linda Walsh said,

    June 4, 2008 @ 10:57 pm

    I had to think about this one — but the phrase they had for 'incorrect' seemed ever so slightly 'off', but I couldn't quite place it. I'm running off at the brain here, a bit, so try to bear with me. Had to think it over after I read the 'answer'. What I came up with is that increase is for numeric quantities. Risen refers to levels or qualities. Numbers don't rise..levels rise. The standard of living is a level or quality. It rises or falls. But…does this clarify or aid: "I increased the level in the bathtub [sic]". That's one step further down the path of 'level' — it sorta grates on my senses even though the meaning is fairly clear. Standard [level] of living…hmm….inflation…thats a 'rate' which is expressed in numbers, so it would increase or decrease. But "standard" of living .. isn't so much a numeric quantities as a 'quality'… Oh…dread…
    I think temperature can be both … it's a quality and a number depending on context. Well, enough that before I really stick my foot in it.
    Opinions? Insults? I'd hate to think someone would try a question like that on an English test though — that's more along the lines of a 'brain teaser'.
    I don't know if I would have placed my 'finger' on it if I hadn't read people's responses…

  6. Helma said,

    June 4, 2008 @ 11:09 pm

    OK, so I'm not a native speaker. But when I consider a simpler form,
    'the standard has ___'
    I am much unhappier with 'increase' than in the full form given above. Much unhappier about that one than about
    ?'prosperity has increased'
    which doesn't sound great to me either.
    'Salaries have increased by 2%'
    sounds much better to me.
    Now, to get to the potential ambiguity nonsense, surely 'increasing standards' is not found in the sense of 'additional criteria being applied'?

  7. john riemann soong said,

    June 5, 2008 @ 12:37 am

    "Rise" seems more vivid and rhetorically effective. I wonder if you used the same psycholinguistics techniques used to test whether "singular they" imposed cognitive load, on say, "the standard [of X] has risen" versus "has increased" whether you would get a shorter time on "risen" (and not simply because it's the shorter word). I don't think it's entirely to the transitive-intransitive issue; for me my mind jumps more hoops trying to rectify "has increased" to :the standard of living," maybe because I associate "increase" with numbers or numerical augmentation and I'm not sure whether the standard of living is being measured in a certain unit. You wouldn't say for pole vaulters, "the bar has increased" — rather it has really literally risen — but you might say "the minimum height required to clear the vault has increased". I am not prescribing against "increase" for non-quantifiable entities, but I am saying it makes my mind jump through less hoops.

    If you say, "the HDI [of country X] has increased" it doesn't make me hesitate.

  8. john riemann soong said,

    June 5, 2008 @ 12:37 am

    Ahh, that funny summative "it" in the last sentence of my first paragraph — I meant to refer to the encouragement of "risen" and not the prescription.

  9. Jane said,

    June 5, 2008 @ 4:11 am

    Yes, language is changing, but I still prefer fewer hoops to less :-) Now, after a smiley, how close should I put the fullstop?

  10. Dick Margulis said,

    June 5, 2008 @ 5:11 am

    Unsubstantiated hunch: If you're looking for the source of the example sentence in test prep materials, I think you want to focus on parts of the world where prescriptivism in English instruction is still very much in fashion. My candidate is India.

  11. Quendus said,

    June 5, 2008 @ 7:15 am

    The real surprise here is seeing a newspaper named "The Star" publishing a pro-immigration article.

    Decreasing the signal-to-noise ratio: I'd prefer "improve" in that sentence, though all three possibilities are Acceptable. *Goes to post youtube coments*

  12. Kathleen Molloy said,

    June 5, 2008 @ 7:17 am

    I can't imagine how a newcomer could feel like they are finally home when arriving in Canada. Not only are the language tests tricky but the tests tend to be designed with an eurocentric bias. I it habit or laziness or institiutionalized us-v-them?

    I tried to poke fun at this a bit in the chapter Keeping up with the Joneses in my novel Dining with Death. The protoganist Zophia learns that her friend had applied for Canadian citizenship after decades in Canada. Following his death, as she roots through his papers to put the last threads of his life in order, she discovers a government issued pamphlet called Canadians are Not Polite Americans. It was to be used as a guide to help integrate her departed friend Kermit. And the citizenship questions leave only confusion and frustration behind. Zophia wonders if the Newbies, the new Canadians will ever get a fighting chance.

    Kathleen Molloy, author – Dining with Death

  13. red said,

    June 5, 2008 @ 8:14 am

    My (BrE) twopennorth: I didn't immediately see anything wrong with 'increased', but on reflection I prefer 'risen' or 'improved', and on further reflection I'm pretty confident I would never produce 'standards have increased' myself. I do read 'increased' quantitatively – 'got bigger'; standards get higher or better.

    The idea I find thoroughly nonsensical, though, is the idea of a '42%' rise in one's 'standard of living'.

  14. James said,

    June 5, 2008 @ 8:30 am

    Find the grammatical (or syntactic) error in this sentence: The standard of living has increased.
    The correct answer is: The standard of living has risen.

    1) Either the reporter thinks that grammar and syntax are different things, or she thinks they are the synonyms. And either way she is so misguided it hurt to read this post.
    2) Even if "increased" somehow was inappropriate there, I'm pretty sure that word-replacement doesn't count as correcting a syntactic error. Isn't this more a question of semantics?

    (By the way, how do you display quotes with the blue bar next to it? That would make the first part of this comment look nicer.)

  15. language hat said,

    June 5, 2008 @ 8:54 am

    I don't know why this has never occurred to me before, but this discussion has made me realize that a mischievous person could pick any construction at random and denounce it just for the fun of watching the opprobrium spread across the prescriptivist world, or take two perfectly good English sentences and state authoritatively that one was correct and the other not and watch people fall all over themselves to provide justifications for the judgment. If you say one person or idea is better than another most people will form their own judgment and either argue or agree with you, but when it comes to grammar (or "grammar"), it seems there's a vast public eager to appropriate any proscription that comes within their ken, whether it makes sense or not. To err is human, to want to feel superior to other people's supposed errors is even more so.

  16. Christopher Craig said,

    June 5, 2008 @ 9:43 am

    James: That was actually the first thing I noticed. Even if we admit that the use of "increase" is an error (an admission I won't make), substituting one intransitive verb for another doesn't strike me as having anything to do with syntax. Maybe, though, this comes from my background in computer programming rather than natural languages.

  17. Adrian Bailey said,

    June 5, 2008 @ 3:54 pm

    Google calls it even:
    The S of L has increased. 192 ghits
    The S of L has risen. 205 ghits

  18. Tom said,

    June 5, 2008 @ 9:41 pm

    Google doesn't call it even if we search more broadly:

    "the standard of * has risen" – 40,100
    "the standard of * has increased" – 4,700

    Clear preference for increased.

    However, some google "research" has failed to confirm my theory that "increase" feels more odd with a naked "standard" (the reasoning being that it feels strange to make a standard "grow", whereas when it's modified into "standard of living" the thing becomes more abstract (less clearly a "level") and therefore more growable)

    I started by trying to make the noun plural to get rid of the "standard of X" construction — here's what I got:

    "standards have increased" – 68k
    "standards have risen" – 39k
    "increasing standards" – 26k
    "rising standards" – 79k

    But many of these are of the form "X standards" (living standards, and so on). So I tried adding "my"…

    "My standards have risen" – 681
    "My standards have increased" – 182

    This seems to support my theory, but the numbers are awfully small. A final attempt using "the"…

    "the standards have risen" – 128; "the standard has risen" – 881
    "the standards have increased" – 79; "the standard has increased" – 614

    Finally, a quick test to see whether my gut sense that "the level has risen" is more natural than "the level has increased"

    Naked "level" phrase…
    "The level has risen" – 11,400
    "The level has increased" – 3,820

    Abstract "level" phrase
    "the level of * has risen" – 97,700
    "the level of * has increased" – 216,000

    Finally the numbers I was looking for, with a clear preference for "rising" with a naked level but a preference for "increased" with a phrase (level of *). If only that had worked for "standard", I might have had a point… :)

  19. Tom said,

    June 5, 2008 @ 9:41 pm

    s/clear preference for increased/clear preference for risen/

  20. Alan Cooper said,

    June 6, 2008 @ 1:43 am

    You clearly don't understand how to evaluate sources. The question was reported in print in a newspaper with professional writers and editors, so it must be real. That the denial comes in a mere "blog" makes it inherently less credible. If you had taken the trouble to read the real book 'Cult of the Amateur' written by Andrew Keen you would have understood this and could have joined happily in the chorus of dismay about the silly test question.

    With regard to the actual answer to the silly question, the one given is in fact wrong since although requiring a correct answer would apparently raise our standard of admission sufficiently to send back home about half of those already here (including the native-born), the standard-of-living is not actually a standard of that sort but rather a measure of quality of life which is properly described as increasing or decreasing rather than rising or falling. On the other hand if we were to establish a minimal acceptable standard-of-living then that would be a standard of the standard sort and so could appropriately be said to rise or fall.

  21. Matthew Stuckwisch said,

    June 6, 2008 @ 2:12 am

    Or we can just go the phrasal verb route: The standard of living has gone up. Has about one tenth as many hits as increased or risen, but it's what I would say in place of increased or risen.

  22. Peter said,

    June 6, 2008 @ 8:29 am

    I think Language Hat's experiment should be tried (just once).

    To my ear, if standards rise, it means the tests/expectations are more demanding; if standards increase, there are more tests/expectations than there used to be (though the latter sounds a bit odd).

  23. Arnold Zwicky said,

    June 6, 2008 @ 11:10 am

    I am hoping that Alan Cooper's comment was entirely tongue in cheek, though the reference to Andrew Keen's book is a bit alarming.

    To anyone who actually believes that everything that appears in a newspaper is accurate and that nothing that appears in a blog can be believed: why are you reading Language Log? (It is, after all, a *blog*.)

  24. mollymooly said,

    June 6, 2008 @ 11:12 am

    I am beginning to think a self-appointed Academy of English would be a good idea. Its prescriptions, and the minutes of the internal debates that produced them, would become the subject of general ridicule and universal flouting.

    It's the implication that the pedants' rules derive from time immemorial that gives them the same perverse force as a conspiracy theory: the lack of supporting evidence itself becomes the evidence.

  25. Victor Latrine said,

    June 6, 2008 @ 8:35 pm

    Surely only a very few people would say "the Dow Jones Index increased yesterday", whilst rather more would say that it had risen.
    A pointer rises and falls, the thing it is measuring increases and decreases (as the pressure in the tire increased, the needle on the pressure gauge rose).
    The argument is more nuanced in the case of the standard of living, and I don't find either sentence particularly difficult to accept.

  26. Stephen Jones said,

    June 7, 2008 @ 5:55 am

    I've does done the google check (since when has Google accepted wildcards, and what are the restrictions?). I get 39,400 to 7,330

    I do suspect, Arnold, that Alan Cooper is being deadly serious. You'd have to do a search on his other posts to see if he really is the master of deadpan humor you take him to be, and not simply some poor deluded schmuck.

  27. Paul D. said,

    June 7, 2008 @ 8:48 am

    "Risen" is correct. Standards can be high or low. Going up is rising. Going down is falling.

    If a standard increases, do you have two standards, or just one really big one?

  28. Ellen K. said,

    June 8, 2008 @ 9:10 pm

    I'm with James. If it's an error, it's a semantic error, not a grammatical or syntactic error.

  29. Bill Poser said,

    June 9, 2008 @ 1:25 am

    In case anyone would like to see what an actual Canadian Citizenship test is like, the Richmond (suburb of Vancouver) Public Library has a practice test. The questions reflect the real exam pretty accurately. On the real exam, in addition to earning an overall score of 60% on the twenty questions, you must get the first two questions, which always deal with the political system, right.

  30. Derek Sellen said,

    June 9, 2008 @ 12:01 pm

    The prohibition of 'standards of living' + 'increase' can be found on page 23 of a very widely used IELTS preparation coursebook, 'Focus on IELTS' by Sue O'Connell (Longman). I'm sure she never imagined it would be the source of such discussion. She explains the example: 'don't use the verbs decrease or increase to refer to the level or standard of something'. But I'd agree with the comment above, that where 'level' is a numerical measurement, 'increase' is fine e.g. sea levels have increased/risen.

  31. Norwegian said,

    June 12, 2008 @ 9:03 am

    Wow – one of the responsibilities of Canadian citizenship is to ELIMINATE injustice!

RSS feed for comments on this post