Which vs That--Problem 2--and Why It's Important

[leonhard_paul]

Hello,

I just wanted to point out that if problem 2 is interpreted as written, one must assume that the Fibonacci sequence has no even-valued term after four million.

Written: "Find the sum of all the even-valued terms in the sequence WHICH do not exceed four million."

"Which" can only be used with non-restrictive clauses (and its clause must be surrounded by commas). In restrictive clauses, you must use "that".

For instance:
Sum the natural numbers, which are positive. (All natural numbers are positive)
Sum the integers that are positive. (Some integers are negative, but we want to sum the positive ones.)

This may seem trivial in this example, but this mistake is frequently made in science and can change the meaning in non-trivial ways.

Here is a good description for scientists, and someone else's views on its importance in technical writing: http://home.earthlink.net/~llica/wichthat.htm

Thanks for reading!

[jaap]

It is unfortunately an oft repeated grammar rule without much foundation.
First of all, things are more complicated than the rule suggests:
http://itre.cis.upenn.edu/~myl/language ... 02146.html
Secondly, there are many counterexamples in literature:
http://itre.cis.upenn.edu/~myl/language ... 00918.html

[leonhard_paul]

These are good counterpoints. This is also a continuation of the battle between linguistic prescription and description. While I can get behind the Sprachgefühl of a native English speaker needing to use a restrictive which, in the end, most of Academia, dictionaries like the OED, most publisher/editors, style guides like the Chicago Manual of Style, all prescribe this general distinction between 'which' and 'that', however. For some, prescriptive use of a language by these institutions may carry no foundation, but this is hardly something to dismiss outright. I also tend to side with prescriptions that do not muddy meaning.

The counterexamples could tell a good story. He just mentions the first usage, but not how they compare to one another proportionally. This later point would be more telling. In addition, these are examples from literature, and not technical writings, where there is arguably less leeway for many kinds of poetic license, whether you are looking to include this which/that usage or not.

Anyway, I find it's better to knowingly transgress somebody else's rules than to do so unwittingly. The author of that blog seems to be much more passionate about his pet-peeve than I am, but linguistics is his domain. At any rate, thanks for providing a dissenting voice to the reactionary semantic-governor in my mind.