Node Test.

LinguisticsNetwork is “an interactive online resource for linguistics and language-related studies”; it’s a for-profit thing, and I’m not pushing it here, but Bathrobe sent me an e-mail saying:

They have a number of free exercises where you can test your knowledge of linguistics. It starts with this real beauty about nodes. Having done the exercise, I now know that I know NOTHING about linguistics. Perhaps the readership of LH might find the same (although I’m willing to swallow my pride and admit that I’m wrong).

I can reassure him that plenty of linguists don’t bother with the details of this particular theoretical approach, but it’s always fun to take tests; they’ve got them for phonology and phonemic symbols as well. Enjoy!

Comments

  1. Throwback to over 10 years ago, and yet I managed 9/10!

  2. Bravo!

  3. January First-of-May says:

    I couldn’t tell node theory from rocket science, and in fact I suspect I would have scored better on a rocket science test.
    As far as I was concerned all the questions were along the lines of “Whom are the doshes distimmed by? A) the kuzdra, B) the gostak, C) the borogove.”

    Even so, 2/10 is probably a little low (in particular, it appears to be worse than chance).

    Might try the other tests later.

    EDIT: 18/20 on phonology, 7/10 on minimal pairs. Three of those errors were for the same reason – I thought the English plural suffix was /s/ instead of /z/. I legitimately flubbed on par/per, and I’m not sure what went wrong with “Thomas”.

    (For the record, the doshes are distimmed by the gostak. The kuzdra budlated the bokr, and the borogoves were all mimsy.)

  4. David Eddyshaw says:

    The other tests are ridiculously simple, in fact, at least for a literate first-language English speaker. They don’t rely on much more than being able to relate English spelling to actual pronunciation.

  5. David Eddyshaw says:

    It is too little appreciated that borogoves were never characterised as mimsy, prior to Carroll’s populist account. Many of us feel that his work seriously misrepresents the traditions of an ancient and proud people.

    There are no well-documented accounts of actual outgribing among raths, come to that. It’s all very well to speak of poetic license, but there should be some limits …

  6. ridiculously simple

    You might think so. Seemingly easy. But there’s one question that requires you to press TWO ANSWERS out of the possible three. Otherwise it’s nuls points, and back to your dreary old day job.

  7. For the question about finding a ferret under the kitchen sink, I would say the correct answer is that you call animal control first, then a plumber.

    Also, it was a stoat, not a ferret (they are easily confused by non-experts).

  8. re nuls points
    For those on the Western side of the Pond, it’s a reference to the annual Eurovision TV sillility where pop musicians from Europe, including Israel and Australia, compete to get zero points (nuls points) from the jury and the national viewing public, who pay to vote.
    I think Norway is the current leader in nuls points, isn’t it, AJP?

  9. I’m not sure Norway’s even in it at the moment, is it? What I’d never realised was that some smaller countries like Norway or Malta or Greece don’t always get to take part. They have to do an earlier qualifying round. Whereas England for instance gets in automatically and then comes last. Nowadays Russia usually wins; something to do with Putin, according to the British press. Israel won once with a bearded chap in a long, skintight evening dress.

  10. John Cowan says:

    the traditions of an ancient and proud people

    Now I’m not a racist, but “a thin shabby-looking bird with its feathers sticking out all round–something like a live mop” is simply well outside the normal range of variation for Homo sapiens.

    (Q: What do you call a deep-frozen corpse? A: Homo snapiens.

    bearded chap in a long, skintight evening dress

    A vision indeed, Euro or not.

    (Q: Why can’t Australia participate in Eurovision? A: Because it is neither in nor near Europe, and making it so would be far beyond the powers of all the dwarves put together, if they could all be collected again from the four corners of the world.)

  11. 2/10 is probably a little low (in particular, it appears to be worse than chance).

    Come on, now, January First-of-May is the only one being honest here. No one else is owning up to their scores. I, too, left it to chance but still got only 2 out of 10. Something about the I-node, I think, a mythical entity that seems as exotic as the kudzra and even harder to find on the Internet. But I didn’t learn my tree diagrams in the last ten years. A question to Yuval: how useful is Node theory to NLP? (Just curious.)

    The rest of the panel has dodged the issue and wandered off into flippant discussions of borograves, ferrets, and Eurovision.

    (Incidentally, what is the pronunciation of ‘-ed’ in ‘poked’? The phonology test seems to hark back to a linguistics that is much older than the node test, by at least 60 years. I think it might have been structuralists who pointed out that voicing is neutralised in both ‘-ed’ and ‘-(e)s’ endings.)

  12. David Eddyshaw says:

    @JC:

    I am saddened by this blatant borogovism.

  13. January First-of-May says:

    Even if they are not, in fact, Homo sapiens (which I admit they might indeed not be), this doesn’t actually prevent them from being an ancient and proud people (which I admit they might indeed be).

    (It doesn’t help that they’re regularly confused with the borograves, a different ancient and proud people whose Homo sapiens status was never really in doubt.)

  14. David Eddyshaw says:

    Their most famous scion was assuredly Max Borograves, the well-known entertainer.

  15. David Marjanović says:

    Even so, 2/10 is probably a little low (in particular, it appears to be worse than chance).

    It’s close – 3/10 would almost be as much as expected from chance.

    I thought the English plural suffix was /s/ instead of /z/.

    It has three allomorphs: /s/, /z/ and /ɪz/. These can be united as a morphophonemic transcription |z|. Naturally, the exercise is rather confused about sounds, phonemes and morphophonemes.

    No one else is owning up to their scores.

    I don’t understand a single one of the node questions. Phonetics 20/20, phonology 10/10 (though the makers overlooked the fact that won and one are a minimal pair for some people).

    Incidentally, what is the pronunciation of ‘-ed’ in ‘poked’?

    Before a vowel you can tell that it’s [t], to my surprise, and thus /t/, even though it’s |d| underlyingly.

    I was also surprised when Katy Perry ki[st] a girl – not because she liked i-it, but because I natively have a contrast between /st/ and /sd/, and likewise between /kt/ and /kd/. It never occurred to me that past and passed are homophones either before I saw native speakers misspell them en masse. But it all makes sense: English has enough phonemic voicing left to assimilate such clusters; southern German doesn’t.

  16. David Eddyshaw says:

    I too didn’t understand any of the node questions initially, but then I had a moment of enlightenment and realised that in every case the correct answer was “merge.”

    (However, the marking scheme appears to be defective, and I unaccountably got nul points. Back to the day job indeed …)

  17. January First-of-May says:

    It never occurred to me that past and passed are homophones either before I saw native speakers misspell them en masse. But it all makes sense: English has enough phonemic voicing left to assimilate such clusters; southern German doesn’t.

    Russian, of course, just devoices all the final consonants it can, including clusters (so съезд “meeting, congress” and съест “will eat (3sg)” [or thereabouts – Russian aspect is weird] are homophonous [except perhaps occasionally before vowels]).

    I still get confused by the implied homophony of past and passed, but for a different reason – because the latter word appears to be bisyllabic in my idiolect (or possibly disyllabic… well, it seems to have more than one syllable, anyway).
    …Well, that, and I still don’t entirely get that English doesn’t really have consonant length. [Apparently it technically does – a minimal pair is unaimed/unnamed.]

  18. David Marjanović says:

    English has phonemic consonant length across morpheme boundaries, and only there. I think this is a very widespread arrangement, though it’s not universal: PIE clearly lacked long consonants even across morpheme boundaries.

    Most of southern German has consonant length (behind stressed vowels anyway), so I do need to pay attention to avoid introducing it into English and French from the spelling.

  19. Before a vowel you can tell that it’s [t]

    Whether it’s [t] or [d] doesn’t matter to me. [d] sounds better, but that could just be some kind of lenition.

    “He poked it in the hole” /hi: pəʊkd əd ɪn ðə həʊl/, although /hi: pəʊkt əd ɪn ðə həʊl/ is probably just as good, and also, of course, super articulately, as /hi: pəʊkt ət ɪn ðə həʊl/. (I’ve enclosed this in phonological slashes although it’s a bit dicey saying what level it is. No stress marking.)

  20. Comparing past and passed is complicated by the fact that the words are etymologically identical, and there has never been a time when native writers consistently distinguished them by spelling. Essentially, they are a single lexeme, which authorities decided should be spelled differently in different usages, and that distinction has therefore become part of the standard, meaning that it predominates in edited prose.

  21. authorities decided should be spelled differently in different usages

    But how about similar usage: un-American English has leant vs leaned, distinguished because leant is pronounced like lent. And there’s learned vs learnt which ought to sound the same according to the past & passed example, but don’t.*

    *(I’m not sure what my point is here or even if there is one.)

    On decoding the node test, it took me some time to recognise that questions were being asked, so claiming 0 out of 20 is a stretch.

  22. Trond Engen says:

    I gave up and did the test. 1/10. And that point was a triumph of tactical reasoning. The second time it asked about a number, I figured that whatever I guessed that was wrong the first time had a more than average chance of being right the second time.

  23. I’m not sure Norway’s even in it at the moment, is it? What I’d never realised was that some smaller countries like Norway or Malta or Greece don’t always get to take part. They have to do an earlier qualifying round. Whereas England for instance gets in automatically and then comes last. Nowadays Russia usually wins; something to do with Putin, according to the British press. Israel won once with a bearded chap in a long, skintight evening dress.

    Crown’s recollections are not entirely accurate.

    1) The last year Norway did not participate was 2002. They won in 2009 with Alexander Rybak, who represented them again last year (and came in 15th place).

    2) The part about qualifying is basically correct. Everyone has to go through the semifinals except a) the “big 5” donors (UK, France, Germany, Spain & Italy) and b) the host country (the previous year’s winner). (Although the UK does not literally always come last, it’s true that they have rarely ranked highly under the current system.)

    3) Though they have often scored highly, Russia has in fact won only once, in 2008. Last year they failed to pass the semifinal for the first time.

    4) The only bearded Eurovision winner in a dress I can recall was Austria’s Conchita Wurst, who won in 2014. Crown might be thinking of Dana International, who won for Israel in 1998. She was transsexual, but did not have a beard. (1998, incidentally, was the last year in which contestants had to sing in their national languages. Nowadays most entries are in English, which makes things much less interesting from my point of view.)

    As for Sashura’s question, going by the lists at https://en.wikipedia.org/wiki/Voting_at_the_Eurovision_Song_Contest#Scoring_no_points it looks like Norway is tied with Austria for most songs with 0 points, unless I’m miscounting.

  24. I gave up and did the test. 1/10.

    I feel better now.

    claiming 0 out of 20 is a stretch

    Not sure what this means but I’ll take it as an encouraging sign.

    I unaccountably got nul points.

    And I’ll take this as a face-saving comment.

    I’m feeling much better than when I read Yuval’s post.

  25. I’m sorry, it seems I’ve conflated two Eurovision transexual entrants. In my day, the Euroskirts were merely on Scotsmen.

  26. John Cowan says:

    It seems to me that Euroskirts, like Eurodollars, ought to be European-style skirts made and/or worn by those outside Europe. My wife rejoices in a pair of blue-and-green Eurosneakers bought in Friesland.

    (Q: What is a kilt? A: A garment occasionally worn by Scotsmen in America and Americans in Scotland.)

    As for the past-tense ending and its relative, the plural/3sg-present ending, their voicing variations are clearly morphophonemic rather than merely phonological. The variants with a voiced consonant plus /t/ are irregular survivals from before the present rules settled down; I myself have drempt, unlike most Americans, who have dream’d regularly. But the variation between regular spell’d and irregular spelt is never matched by a variation in the merely homophonous noun speltTriticum spelta, a species of wheat’.

  27. Athel Cornish-Bowden says:

    (Q: What is a kilt? A: A garment occasionally worn by Scotsmen in America and Americans in Scotland.)

    If you see a man in a kilt in London you think he’s Scottish; if you see a man in a kilt in Edinburgh you think he’s from the Highlands; if you see a man in a kilt in the Highlands you know he’s American.

  28. Athel Cornish-Bowden says:

    I couldn’t make any sense of the first nodes question and didn’t try any further. Maybe I should. If I get as much as 1/10 I’ll report back.

  29. Athel Cornish-Bowden says:

    What a triumph! I got 3/10 without understanding anything. Better on minimal pairs, however: 10/10. I think I’ll start claiming to be an expert on linguistics.

Speak Your Mind

*