Same character, different 'word'...

Wikipedia article on cognates.

It turns out this phenomenon is considered a cognate. People think that Chinese-English doesn't have any true cognates, but Chinese has cognates relative to itself.

One of the more common examples is 长 (chang2, lenght) and 长 (zhang3, to increase). They probably both derive from some common linguistic ancestor. On Memrise, they are different words due to how we define 'word'. I'm guessing the fact that they have different sounds is the delimiter. If the sound was identical, it would probably be considered one word with similar meanings.

Posted by carl_a 6/25/12 , last update 6/26/12 (4 years ago)
  • that's one of the reasons I suggested that Memrise start differentiating characters based on part of speech. Mandarin is an extremely efficient language - from what I can tell. Every sound can be associated with multiple characters (shi4 = room, to sting, conditions, etc). Characters can mean different things based on location - 家 - could be home (a noun) or a measure word, or an adjective (domestic - src - nciku).
    English has several too (but it's just not as efficient as Mandarin)- I lead the people into the lead mines or I will read this book tomorrow but I read that book yesterday or I have to train the trainees on how to operate a train.

    Posted by DarthJen 6/25/12 (4 years ago)
  • Today, I fixed a word that had its part of speech listed incorrectly. The field for that in Memrise is text, but I think it would be more useful if it were a pull-down list to assure consistency. If its a 'verb' it shouldn't be abbreviated 'v' for example. One one hand these seem like edge cases, but they are also some of the hardest parts of the language to learn.

    Should require the part of speech to be entered so that it could be displayed next to the character? Of course, we also have situations where the different words are both verbs....

    Posted by carl_a 6/25/12 (4 years ago)
  • For the drop down - I think Sfrenchie on the spanish forum requested that as well. It would make things a lot more consistent in terms of style.

    As for the second scenario that you listed: that's why I suggested something really stupid: verb1, verb2. I'm sure someone could suggest a more elegant solution. I mean, I would love it if for nouns, the classifier/measure word would show so that some sort of subliminal association could begin. (Someone already suggested this earlier in the forum.) But the wiki database would have to be modified (add an extra field, ensure that the character/meaning associated with the mw is a noun).

    Posted by DarthJen 6/25/12 (4 years ago)
  • I can't think of a more elegant solution within the existing infrastructure and I like your idea because it doesn't cause any harm. As a guiding principle though, I would say that a goal is to memorize the content, and avoid if at all possible Memrise specific information. You're definitely on the right track here. Thanks for replying.

    Posted by carl_a 6/26/12 (4 years ago)

Recent threads

This forum doesn't have any recent activity