You are currently browsing the tag archive for the ‘coha’ tag.

To a linguist, there is an obvious difference between verbal and oral: only the first word can be used to mean “pertaining to a verb”. But for people who don’t talk about parts of speech so often, the more relevant question is whether verbal can refer to spoken language (as opposed to written language), or if it can only refer to the more general sense of all language:

(1a) The written warning is primarily the verbal warning put in writing […]
(1b) […] general verbal skills, such as verbal fluency, ability to understand and use verbal reasoning, and verbal knowledge.

Some people insist that verbal can’t be used as in (1a). Verbal is derived from the Latin verbum, meaning “word”, and that means that it only distinguishes things involving words from things not involving words. This is the usage in (1b), where verbal reasoning is implicitly differentiated from mathematical reasoning, or spatial reasoning, or any other form of reasoning that is not based in words. Clearly this is a valid usage of verbal.

And, while we’re at it, we can quickly agree that oral would be inappropriate for the usage in (1b). “Oral knowledge”, for instance, is specifically knowledge that is spoken aloud, and I really can’t see that being the intended meaning. I think we can also all agree that oral is definitely appropriate for the usage in (1a). So what we have a is 2×2 chart, with three of the values filled in:

using words spoken
verbal YES (1b) ? (1a)
oral NO (1b) YES (1a)

The only remaining question is whether verbal is allowable in that last cell, with the meaning “spoken”. And the answer is yes, and it has been almost since verbal‘s first appearance in English. The Oxford English Dictionary first attests verbal in 1483, but at that point it modifies people. William Caxton writes:

We be verbal, or ful of wordes, and desyre more the wordes than the thynges.”

The first attestation of verbal meaning “composed of words” comes between 50 and 100 years later, in either 1530 or 1589.* And the first attestation of verbal meaning “conveyed by speech” comes in 1617:

“The Chamber of the Pallace where verball appeales are decided […]”

This meaning has persisted. I looked at the Corpus of Historical American English (COHA) for the most common nouns to follow verbal over the past 200 years. The two most common collocates were communication and expression, each with 43 hits. Unfortunately, looking at the contexts in which these were used, it’s hard for me to tell which meaning was intended. But the third most common collocate, message, appears 40 times, spread out over the past two centuries. And these are pretty unambiguously examples of the “spoken” meaning, because it’s rare that you’d need to distinguish message delivered in words from those that aren’t. For instance:

“His reply was this verbal message: ‘Wait — and trust in God!'” [1875]
“The verbal message is the key to the written one.” [1909]

I don’t have numbers on the relative usage of the two meanings of verbal, so I’m not going to try to say that one is more common than the other. But it is pretty clear that the “conveyed by speech” meaning is valid.

Does this acceptability mean that you should unquestioningly use it in this way? Not necessarily; there is a potentially significant ambiguity here, so it’s not the best choice in all situations. On occasion, it will matter whether verbal means “conveyed by speech” or “involving words”. If I write to a tutor and ask them to improve my verbal skills, it may be ambiguous as to whether I’m looking for instruction in public speaking or vocabulary building. That’s a trivial example, but in legal contexts, it’s probably better to refer to oral contracts, warnings, etc. than verbal ones, just to avoid the ambiguity.

In most cases, where this ambiguity is small or unimportant, you can and should use whichever feels better to you. You can freely swap between the two meanings in different contexts, as I do. A lot of the time, the context (especially what noun verbal is modifying) will clarify things. So in the end, our chart becomes:

using words spoken
verbal YES YES
oral NO YES

Summary: Verbal can refer either to anything delivered in words or something that is specifically spoken. This latter usage is sometimes condemned as modern sloppiness, but it’s been persistently attested for 400 years. The ambiguity is generally not sufficient to be problematic, so it’s only in cases where precision is paramount that the latter usage should be avoided.


*: The 1530 attestation is listed under this definition, but its usage seems to me identical to Caxton’s usage, modifying people. The 1589 attestation is unambiguously referring to language, referring to “verbale sermons”.

I hate when someone starts a monologue by needlessly invoking a dictionary definition for some word. Few openings can ruin a graduation speech faster than “Webster’s defines ‘scholarship’ as …”. (Even the Yahoo! Answers community knows this.) For most common words, the dictionary definition is just a simplified, neutered form of the rich definition that native speakers have in their heads. There’s no need to tell me less about a word than I already know.

Unfortunately, I simply can’t come up with another way to start today’s post. I recently ran across this analysis of can’t help but, an idiom that (if you can believe it) the author finds illogical:

“Try to avoid the can’t help but construction. While it has been around for a while, most grammarians agree that it’s not the most logical construction. It’s considered to be a confused mix of the expressions can but and can’t help.”

Before we try to “logically” analyze idioms, let’s reflect for a moment what an idiom is. Here it comes — The Oxford English Dictionary defines an idiom (in its third noun sense) as:

A form of expression, grammatical construction, phrase, etc., used in a distinctive way in a particular language, dialect, or language variety; spec. a group of words established by usage as having a meaning not deducible from the meanings of the individual words.”

I’ve bolded that last bit because that’s the key point: an idiom is an idiom when its meaning is well-known among users of the language but does not come from strict interpretations of the words themselves. If you say someone has idiomatically kicked the bucket, there’s no bucket, there’s no kicking motion, and it actually means they died. Logical analysis of kick the bucket won’t get you anywhere near the actual meaning.

With that in mind, let’s look at can’t help but. Surely, most fluent English speakers — including those who disparage it as “illogical” — know what it means. If that meaning can be deduced from the words and syntax of the construction, then hooray, it’s fine, because it’s grammatical. If that meaning cannot be deduced from the words and syntax of the construction, then hooray, it’s still fine, because it fits exactly the definition of an idiom. It doesn’t matter if the meaning is deducible or “logical”, whatever that means. (For some thoughts on why I put “logical” in quotation marks when talking of grammatical logic, see Emily Morgan’s post on the logic of language.)

You might think that I’ve done some rhetorical sleight-of-hand in the last paragraph by saying that can’t help but either makes sense or is an idiom. What if it isn’t an idiom, but just an illogical corruption of can help but? I’ve got two thoughts on that.

The first is a simple matter of history. The OED records the use of can’t help but starting in 1894, but I’m finding it in Google Books further back than that. Here are examples from 1852 [Uncle Tom’s Cabin], 1834, and 1823. Similar investigation antedates can help but around the same time, 1842 and 1834. There’s no clear evidence that one form predates the other, so there’s no evidence that cannot help but is a corruption of the correct form.

The second point is that the supposedly logical alternatives can help but and can’t help make no more sense than cannot help but. I don’t understand the above claim that can’t help but is “not the most logical construction”. Maybe it isn’t; I’ll grant that it’s not as immediately interpretable as “I am walking” or something. But if can’t help but isn’t logical, why are the alternatives can help but and can’t help logical? What meaning is there for help that makes can’t help eating the cake mean “can’t stop myself from eating”? Whatever it is, it’s strictly idiomatic; you couldn’t, for example, write “I am helping eat the cake” with the meaning “I’m stopping myself from eating the cake”. In fact, it means exactly the opposite!*

For confirmation, I checked in the OED, and this meaning occurs only in these idioms. So can help but and can’t help aren’t “logical” either; they’re the result of people applying idiomatic knowledge to the interpretation of the construction. As soon as you expect help to mean something other than its standard aid-related usages, you’re going idiomatic, and logic pretty much goes out the window.

This is a long way of arguing that can help but and can’t help but are both grammatically reasonable. Shouldn’t we decide on one form over the other? Well, no. I know that prescriptivists love doing that, but it’s not the way language really works. The fact of the matter is that both are common, and in the opinion of the Merriam-Webster Dictionary of English Usage, both are standard.

But if that still won’t placate you, if you simply must be told which one is better, the perhaps surprising answer is that it’s the “illogical” can’t help but. The Corpus of Historical American English has 243 examples of can not help but to a mere 6 of can help but, and Google N-grams shows cannot help but dominating since 1840. (And personally, can help but doesn’t exist in my idiolect.) If you want to write the more common form, go with can’t help but. If can help but seems better to you, go with that.

Summary: Can’t help but is a perfectly standard idiom, meaning “can’t stop myself from”. It’s also the more common choice, historically and contemporarily, over can help but, even though both options are grammatical and standard in English. (Can’t help Xing is fine too, of course.)


*: Furthermore, doesn’t can’t help Xing have the potential to be even more confusing than can’t help but? If I say “I can’t help putting together your bike today”, am I saying that I can’t do it or I can’t stop myself from doing it?

The explosion of data available to language researchers in the form of the Internet and massive corpora (e.g., the Corpus of Contemporary American English or the British National Corpus) is, I think, a necessary step toward a complete theory of what the users of a language know about their language and how they use that information. I became convinced of this with Joan Bresnan’s work on the dative alternation — which I’ve previously fawned over as the research that really drew me into linguistics — in which she and her colleagues show that people unconsciously combine multiple pieces of information during language production in order to make probabilistic decisions about the grammatical structures they use. This went against the original idea (which many grammaticasters still hold) that sentences are always either strictly grammatical or strictly ungrammatical. Furthermore, it showed the essential wrongness of arguing that one structure is ungrammatical on analogy to another structure. After all, if (1a) is grammatical, by analogy (1b) has to be as well, right?

(1a) Ann faxed Beth the news
(1b) ?Ann yelled Beth the news

That’s not the case, though.* There are a lot of different factors affecting grammaticality in the dative alternation, including the length difference between the objects, their animacy and number, and even the verb itself. But this conclusion was only reached by using a regression model over a large corpus of dative sentences. This regression identified both the significant features and their effects on the alternation proportions. In addition, having the corpus allowed the researchers to find grammatical sentences that broke previously assumed rules about the dative alternation, showing that the assumed rules were false. Prior to having a corpus study on this alternation, people thought they mostly understood it, but now that we have the corpus study, the results are much different from what we’d been saying.

And this illustrates the power and downright necessity of corpora to descriptivist linguistics (i.e., linguistics). Sure, it might seem obvious that if you really want to describe a language, you need to have massive amounts of data about the language to drive your conclusions. But for almost the whole history of linguistics, we didn’t have it, and had to make do from extracted snippets of the language and imagined sentences, and those are susceptible to all kinds of biases and illusions. Having the corpora available and accessible can save us from some of these biases.

But, of course, corpora can introduce biases of their own. Corpora are imperfect, and in general they still must be supplemented by value judgments and constructed examples. An example that I once had the pleasure of seeing Ivan Sag and Joan Bresnan discuss was that if we go by raw word counts, the common typo teh was as much a word in the 1800s as crinkled. Similarly, if we were to turn linguistics over to corpora entirely and only accept observed sentences as grammatical, then I swept a sphere under the fogged window would be ungrammatical, since it has no hits on Google (at least until this post is indexed). Corpora are treasure troves, but as a quick review of the Indiana Jones series will remind you, treasure troves are laden with pitfalls and spikes.

Yep, this is exactly the sort of danger I face in my day-to-day research.

I was reminded of this when I looked up the historical usage of common English first names to look for rises and falls in their popularity. I looked up Brian in the Google Books N-grams, and found a spike that represents what I like to call the era of Brian:

Hmm, something sent Brian usage through the roof in the late 1920s, only to come crashing back down like the stock market (might they have been linked?!). Time to investigate further in the Corpus of Historical American English (COHA):

Oh wait, never mind, the era of Brian wasn’t in the 1920s; it was in the 1860s (and presaged in the 1830s). Wait, what? Let me go back to Google N-grams:

Oh dear, it’s spreading! What is happening? What is the meaning of Brian?!

The fact is, as you surely already knew, that there was no era of Brian. The variability of the length of the era in the first and third graphs is due to me changing the smoothing factor on the graph. The source of the spike is that in one year the proportion of “Brian” in the corpus shot up to around 10 to 20 times its base level. (This becomes clear if you look at the unsmoothed numbers.) And if we look at the composition of the corpus at that point (1929), it turns out that the Google Books corpus contains a 262-page book titled “Brian, a story”, which seems like it would account for this surge. The COHA corpus has a similar thing going on; two books in 1832 and 1834 have prominent characters with the name Brian, and 1860 has a book titled “Brian O’Linn”.

And that’s one of the problems of corpora. Sure, they’re full of far more linguistic information than the little sampling we used to use, but they’re still incomplete and composed as a not statistically independent sample of the full range of language. If these corpora contained the whole of all writing published in these years, the Brian spike would be negligible, but because of the inherently incomplete nature of corpora, a single book can have an inordinate effect on the apparent proportions of different words.

Corpora are great, but they’re also noisy, and they do require interpretation. I didn’t get that at first, and thought that interpreting corpus data was invariably infecting it with one’s own prejudices. And yeah, that’s a danger, writing off real phenomena that you don’t believe in because you don’t believe in them. But the answer isn’t to accept the corpus data as absolute truth. You have to be as skeptical of your corpora as you are of your constructed examples. And that’s advice, I’m sure, that very few of you will ever need.


*: If you find (1b) to be perfectly grammatical, that’s fine. I think you’ll find other examples in the paper that you consider less than perfectly grammatical but have grammatical analogues. And even if you don’t, the data will hopefully assure you that other people do.

Post Categories

The Monthly Archives

About The Blog

A lot of people make claims about what "good English" is. Much of what they say is flim-flam, and this blog aims to set the record straight. Its goal is to explain the motivations behind the real grammar of English and to debunk ill-founded claims about what is grammatical and what isn't. Somehow, this was enough to garner a favorable mention in the Wall Street Journal.

About Me

I'm Gabe Doyle, currently a postdoctoral scholar in the Language and Cognition Lab at Stanford University. Before that, I got a doctorate in linguistics from UC San Diego and a bachelor's in math from Princeton.

In my research, I look at how humans manage one of their greatest learning achievements: the acquisition of language. I build computational models of how people can learn language with cognitively-general processes and as few presuppositions as possible. Currently, I'm working on models for acquiring phonology and other constraint-based aspects of cognition.

I also examine how we can use large electronic resources, such as Twitter, to learn about how we speak to each other. Some of my recent work uses Twitter to map dialect regions in the United States.



@MGrammar on twitter

Recent Tweets

If you like email and you like grammar, feel free to subscribe to Motivated Grammar by email. Enter your address below.

Join 972 other followers

Top Rated

%d bloggers like this: