Without a doubt

A student (a native speaker of German) used the expression "without doubts" to start a sentence. That sounds wrong to me; I would prefer "without a doubt". But instead of just marking it as wrong, I did a quick bit of research.

I searched the NOW Corpus at the Brigham Young University site with linguistic corpora. The NOW Corpus contains 3.3. billion words of data from web-based newspapers and magazines from 2010 to the present. And by the present, they mean the present: it's updated with 4-5 million new words every day.

There were 27 hits for "without doubts" and 7752 hits for "without a doubt." So I can correct the student's usage in good conscience.

And that is how a usage point can be resolved: check actual usage using a linguistic corpus like the NOW Corpus (or any of the other corpora at the BYU site).

Using "that" to refer to people

I saw a complaint about the supposedly increasing use of "that" to refer to people (instead of "who").

Since I have a digital copy of Jane Austen's "Emma," it took me only a few minutes to find an example of such a usage by Austen: Mr. Woodhouse "was a nervous man, easily depressed; fond of every body that he was used to, and hating to part with them; hating change of every kind."

But the Corpus of Historical American English also provides some evidence. I searched for the phrase "man who" and the phrase "man that". Amusingly, "the man that" seems to be decreasing in frequency. From the 1830s to the 1930s (by decades), it appears with at least 200 hits per decade, while since then it has been decreasing, with only 94 hits in the 1990s and 116 in the 2000s.

But the complaint was made by someone from Britain. The BYU corpora do not include a corpus of Historical British English like the COHA for American English, but you can use the BYU corpus site to search Google Books. And again, "man that" has decreased considerably since the 19th century: over 20,000 hits per decade from the 1830s to the 1900s (and up over 30,000 sometimes). Oddly enough, there is then a sudden drop in hits between the 1900s and the 1910s, from over 36k hits to about 12,500. And then it's been pretty steady since then.

So whatever is going on here, it is highly unlikely to be a matter of increasing frequency of use of "that" to refer to people (or at least to refer to "man"). This is what's known as "the frequency illusion."

"deemed as"? "based on" vs. "based off of"?

A Facebook friend who was grading papers wrote:

Fellow language/grammar enthusiasts: I need another verdict so I know if two more of my pet peeves are justified. My students write that something is "deemed as" rather than simply "deemed" the adjective that follows. They also write that things are "based off of" rather than "based on" other things. I hate both. May I correct them or are these now so common as to be legitimate?

I responded:

This is the kind of thing I love to go a little bit crazy with …

I looked up the phrases in the Corpus of Contemporary American English (COCA) and the Corpus of Global Web-Based English (GLOWBE). These two corpora can be found at the BYU site for linguistic corpora.

For based on vs. based off of:

COCA
based on = 61860
based off of = 9

GLOWBE
based on = 335832 (US 76525)
based off of = 736 (US 392)

I think you can safely say that “based off of” is not yet Standard English.

Deemed (as): this pair is a bit harder to deal with, because it’s not two distinct phrases (i.e., all hits for “deemed” as are also counted for “deemed”, and many of the uses of “deemed” might be in contexts where “deemed as” would not be used). But here are the numbers:

COCA
deemed as = 56
deemed (including the above) = 5141

GLOWBE
deemed as = 1419 (US 136)
deemed (including the above) = 39523 (US 5512)

While these ratios lean quite strongly in favor of not using “as” with “deemed,” I think there’s enough uncertainty about the numbers to make “deemed as” worth accepting, as well as enough grounds for understanding where “deemed as” comes from (parallel to “regard as” and “consider as” — which latter form I don’t like, actually, but have to admit exists).

This is only about current usage (1990-2012 with COCA, 2012-2013 with Glowbe), so it doesn’t even address the history of “deemed as” and “based off of.” I bet the former has a long history, while the latter doesn’t. But linguists have a term to refer to people’s sense that a construction is new when it is not: “recency illusion”.

And the Google Books Corpus shows that I’m right: “base d off of” does not appear until the 1990s in American English (though it has risen sharply since). In contrast, “deemed as” has been in steady use for 200 years now.

So mark “based off of” as wrong (and discuss it?), but accept “deemed as” (and talk about the “recency illusion”?).