3

This is from an exercise in Minna no Nihongo, ch16.

Full sentence:

神戸は古い物があまりありませんが、町の後ろに山が、前に海があって、素敵な町です。

As above, I don't understand why there is nothing after the が. I can sort of see that two sentences have been contracted into one but I'm still lost:

Machi no ushiro ni Yama ga aru

Machi no mae ni umi ga aru.

If someone could help that would be great :)

senshin
  • 5,645
  • 30
  • 72
user65218
  • 145
  • 6

1 Answers1

15

後ろに山が has been coordinated with 前に海が:

町の	[　後ろに　山が、 ]  
　　	[　前に　　海が　 ]　あって、素敵な町です。

One possible way to describe it is like this:

  1. Start with the following:

    	[　町の後ろに	山があって、 ]  
　　	[　町の前に　	海があって、 ]　　　　　素敵な町です。

  2. Pull out 町の from the left side:

    町の	[　＿＿後ろに	山があって、 ]  
　　	[　＿＿前に　	海があって、 ]　　　　　素敵な町です。

  3. Pull out あって from the right side:

    町の	[　＿＿後ろに	山が＿＿＿、 ]  
　　	[　＿＿前に　	海が＿＿＿　 ]　あって、素敵な町です。

In linguistics, steps 2 and 3 are called left-node raising and right-node raising respectively. We have something like this in English, too, which you can see in phrases like pre- and post-war:

　　[ pre-war ]　and　[ post-war ]　
　　[ pre-___ ]　and　[ post-___ ]　war

Another way to understand this is by analogy to mathematics. When we pull 町の out from the left, it's like we're factoring it out:

Let a = 町の
Let x = 後ろに山が
Let y = 前に海が

　ax + ay = a(x + y)

Of course, the explanations above are just metaphors to help you understand how the words fit together. It's not necessary for you to believe that anyone actually starts with more complex sentences and reduces them this way. And this certainly isn't the only possible way to explain these particular examples.

Whichever way you choose to think about it, in your example the 町の on the left connects with both coordinates at the same time, and so does the あって on the right. So I think your description is basically correct.

Community
  • 1
  • Re: "we don't have left-node raising in English", isn't that what happens with things like "[John's car] and [John's dog] are brown" > "John's [car] and [dog] are brown"? – Eiríkr Útlendi Jul 10 '14 at 16:16
  • Shûichi Yatabe makes a compelling case for Japanese having LNR, but English is rarely analyzed that way. Your example could be analyzed instead as Conjunction Reduction. But there are many ways to analyze coordination, and I don't think I can make a principled argument in the space of a comment. –  Jul 10 '14 at 16:56
  • Likewise, above I analyzed the missing あって as RNR, but some linguists would call it (backwards) gapping. I didn't think the theoretical consequences of the difference were important enough to mention here. –  Jul 10 '14 at 17:00
  • 同じ疑問を持ったんですが、この論文にたどり着いた辺りで理解するのを諦めました… – naruto Jul 10 '14 at 17:03