<p>A long campaign season with genuine gaffes and alleged misstatements began its culmination with the first presidential debate. Like many citizens, linguist Geoff Pullum, a visiting professor at Brown, was watching.</p>

Geoff Pullum: The Gerard Visiting Professor of Cognitive, Linguistic and Psychological Sciences at Brown University andProfessor of General Linguistics, University of Edinburgh
Geoff Pullum The Gerard Visiting Professor of Cognitive, Linguistic and Psychological Sciences at Brown University and
Professor of General Linguistics, University of Edinburgh
Grammarian Geoff Pullum is the Gerard Visiting Professor of Cognitive, Linguistic and Psychological Sciences at Brown University and professor of general linguistics at the University of Edinburgh. He is also a frequent blogger on language and politics on Language Log and Lingua Franca.

In a conversation with David Orenstein, he cited several specious analyses of word choice and syntax that have been used unfairly against candidates. The often ill-informed critiques stand in stark contrast to the way people are typically inclined to overcome the misstatements of others as they extract understanding from clumsy speech. Pullum will be listening closely to the presidential and vice presidential debates.

Are there clues in linguistics that can help us discern whether a gaffe is a matter of poor phrasing or poor thinking?

Language doesn’t usually matter all that much if you look at the intent and the actual facts of the matter, but what happens is that the press picks up on it as if it was the most important thing in the world, and you get this fake bubble of newsworthiness out of an incident that is an unremarkable and unimportant slip of speech. Sometimes it isn’t even true at all: The bubble of publicity isn’t even filled with the gas of misinterpretation, there’s just nothing there.

Everybody was always laughing at George W. Bush for his misstatements. Sometimes they were just classical malapropisms (you reach for a word and pick out the wrong one); others were slips in sentence planning, which are completely unimportant — everybody makes a few of those every day. But they just used to hunt for them in Bush’s speech.

“There’s a lot more “Grammar Gotcha” than there is genuine relevance of grammar to the ebb and flow of politics and the ups and downs of people’s careers”

When it comes down to it, what linguistics reveals is that people are astonishingly tolerant, adaptable, constructive, and brilliant at screening out the mistakes. It is a constructive process, not a subtraction process. To screen out the mistakes that people make in speech, you have to not just ignore some of the things that occurred, like cutting out the “ums and “ers,” you have to positively construct a sentence that does make sense. We all do that all the time; we’re not constantly laughing at our friends for disfluencies and utterances that had a second interpretation.

But then when it comes to politicians and the press, suddenly the ordinary social-psychological contract is torn up, and it’s time to catch them in errors and ridicule them and drag things out of context. And of course, much more so when it’s a campaign time and one side is looking for missteps by the candidate of the other side. That’s what you find when Barack Obama says, “You didn’t build that.”

What it sounded like he was saying is that you didn’t build your company at all. So the Republicans jumped on this and soon had an attack ad ready for TV that had Barack Obama repeating that phrase five times over — “If you’ve got a business, you didn’t build that.” But he was referring back to the other things he just said about the Internet and the freeway system and all of that. You didn’t build all that. You built your business on the basis of an infrastructure that was here already.

He was jumped on for it. That’s the way things happen.

It sounds like language is flexible. It’s really more a matter of the listener’s disposition whether they will do constructive repairs or strip out the intended meaning of words and reuse them hostilely.

In addition to what you can do with genuine minor errors like Bushisms, you’ve also got the outright lying — what I refer to on Language Log as “making stuff up.” With language you seem to be able to do that ad lib. The most extreme case (and it really does strike me as extraordinary every time I review the Language Log postings on this topic) is the allegation started by George Will to the effect that Barack Obama is an egotistical president and that his speeches are stuffed with first-person pronouns because he talks mainly about himself. Mark Liberman, in more than a dozen Language Log posts, has checked this out thoroughly. He has gone over speeches by Obama and compared them with Bush Jr. and Bush Sr. and Reagan, and has occasionally looked at other figures. He’s counted those pronouns. He publishes the details of where he got the text of the speech, and he publishes the script that he uses to hunt for instances of I, me, my, mine and I’m, and he shows the figures and tables. The fact is, for what it’s worth, Obama uses fewer first person pronouns than other recent presidents.

“There are a lot of myths drifting around about grammar that really are myths and always have been; things that are alleged to be incorrect whereas in fact they are perfectly correct”

The percentages of course are low. They are down there in the region of 2 percent. The point is you find him using 2 percent and George W. Bush using 3 or 4 percent.

This applies not only to Obama, but also to other figures. One nice recent case and quite instructive was the case of the speech given at the Republican convention in Tampa by Gov. Chris Christie, which was widely criticized for being egotistical because Christie seemed to be bragging about his achievements and publicizing himself. Only in the last third of his speech did he even mention Romney and he didn’t mention him very much.

A newspaper column asserted that he used “I” 30 times and this was a measure of his egotism. So of course Mark Liberman of the University of Pennslyvania pricks his ears up and says, “Hello, nobody’s done the counting yet.” He did the counting. He took the speech, he counted the pronouns and then he did the same for four other speeches at the same Republican convention. He counted I, me, my, and I’m, and he got results. For Chris Christie, a little bit over 2 percent of his words were these pronouns. Paul Ryan used more. Ann Romney used significantly more than that. Clint Eastwood was up at almost 5 percent.

Of course, we don’t really know much about whether pronoun frequency correlates with egotism. But we do know that merely counting the frequency of words associated with something is not likely to be a good index of the inner thinking and attitudes of that person. So it might be that measuring the percentage of first-person pronouns is not giving an index of anything. But if you want to know, the facts are Barack uses first-person pronouns less frequently by apparently a significant amount than other recent presidents, and Chris Christie uses fewer of them than the other people who gave major addresses to the RNC.

“[We’ve] counted those pronouns. ... The fact is, for what it’s worth, Obama uses fewer first-person pronouns than other recent presidents”

It seems that the way that people mishandle their analysis, much less the language itself, is quite broad.

Nearly all of the talk in the media about language in politics is about word choice in politics. It’s quite rare to come upon something that really does relate to syntax. You see it, but again it’s usually fake. It’s a non-issue that’s been fluffed up on the basis of nothing at all. Noreen Malone got hold of a letter that Barack Obama wrote once and had it examined by a Columbia English professor to see how smart this guy really was, and one piece of evidence cited for giving him a B-minus was that he was “confusing that and which.” This is the biggest myth in American understanding of grammar: the idea that it is ungrammatical to say “any job which one would want,” rather than “any job that one would want.” It’s a huge myth, but there is a sort of trope here that if you can find a relative which without a comma before it in somebody’s writing you can crow over it and ding them for making a mistake because they should have used that instead.

But President Roosevelt, talking about the Pearl Harbor attack, said that December 7 is “a date which will live in infamy.” That’s a restrictive relative clause: “which will live in infamy.” He wasn’t doing that because he didn’t know how to speak English properly. He wasn’t some ill-educated oaf.

That’s how it is a lot of the time: There’s a lot more “Grammar Gotcha” than there is genuine relevance of grammar to the ebb and flow of politics and the ups and downs of people’s careers. Your syntax isn’t going to matter very much in reality for the most part, but it will have an effect on whether journalists find things to ding you for. I don’t know if the general public really is less likely to vote for a man who once used a restrictive relative clause that began with which. I’m inclined to doubt it.

What difference does it make when people speak on the fly rather than in more rehearsed or scripted ways?

It makes a huge difference in two respects. First, if you are going to be distracted by the constant flubs and corrections and misspeakings of spontaneous speech you might prefer to listen to a carefully prepared presentation of a point of view. The other thing is that if you want to hear a president give his actual views, then catch him candidly on television because otherwise you’ll just get a scriptwriter.

As for Todd Akin, of course, what a terrible slip to make. It seems clear now that when he said legitimate rape, he did mean “real” rape — i.e., forcible rape. That’s what he has now claimed that he meant. He was talking about forcible rape as if there was some other kind that is perfectly OK. This man deserves the huge outburst of hostility that he got as a result of that ridiculous remark.

Of course, it is possible that in the debates candidates will simply say what they mean and what they mean may sound entirely reasonable.

“President Roosevelt ... said that December 7 is ‘a date which will live in infamy.’ That’s a restrictive relative clause”

There may be just a perfectly ordinary use of language in the presidential debates. Usually it’s the ideas expressed that matter much more. But one can always watch for a classic really interesting syntactic flub. You never know what’s going to happen. One surely wouldn’t have expected the fascinating little piece of syntax to come out of the swearing in ceremony of Barack Obama, but what appeared to happen was that because of a failure to rehearse the speaker-change breaks, Obama started talking too soon and that threw [Chief Justice] Roberts off, and Roberts then unintentionally edited by moving an adverb to the end of the verb phrase. He sort of unconsciously shifted the adverb until later, so they didn’t use the right words for the oath. When they looked back at the horrible mess that resulted, they realized that the right words hadn’t been used, so strictly it was possible to say the president hadn’t been duly sworn in by the chief justice, so they did it again.

You teach a class about grammar at Brown. What do you want the students to understand?

People have been repeating the same old same old from, believe it or not, not just the 20th century, not just the 19th century, but the 18th century. They are repeating Lindley Murray and Robert Lowth from the 1700s -- the same old analyses that don’t make sense. I’m trying to teach it in a way that gives the analysis the kind of update that is needed if you are just going to keep in touch with what’s been discovered about the structure of English over the last century or two. I’m trying to bring 21st-century thinking about how to analyze sentences into play because once you get to be more than a century or so out of date in a subject, I take that to be bad news.

There are two different ways I want to change the students’ lives with this stuff. One is that I want them to be aware that there are a lot of myths drifting around about grammar that really are myths and always have been; things that are alleged to be incorrect whereas in fact they are perfectly correct. These include things like the split infinitive. It never has been a mistake to say “to really understand” as opposed to “really to understand.” It has always appeared in decent literature, all the way through the history of English.

The other thing is, the real motivation for teaching anybody about the syntax of English, the technical details of how to understand sentence structure, seems to me to be this: If there is some sound advice on how to write out there that you want to pay attention to and you want try to follow, you won’t be able to do it if you don’t even have enough technical knowledge to understand what the advice says. Suppose it were true that writing passive clauses, at least with any significant frequency, was a very bad idea. In practice I don’t think that’s true, but suppose it were true. You’d better know how to identify passive clauses or you can’t even begin to take advantage of the warnings against them. Just from the standpoint of whether you want to be able to understand advice about how to write better, you’ve got to have some grounding in technical notions of grammar and you might as well have them on the basis that makes some sense, rather than repetition of 19th and 18th century myths about it.