Reagan vs Obama and McCain

I thought it would be interesting to look at the level of spin in Reagan’s speeches. He shares some characteristics with Obama; not in political opinions but in his ability to motivate an audience, and to be resistant to potentially embarrassing factual issues.
Here is the plot from yesterday’s post comparing Obama and McCain since their conventions, with five campaign speeches (all I could find) by Reagan between the convention and the 1980 election.

Comparing the spin of Reagan, Obama, and McCain

The points with red stars are Reagan’s speeches. As you can see, his level of spin is much higher than either of today’s candidates. The ability to use high levels of spin without coming across as phony is, of course, what makes an actor, so this is not entirely surprising. And I’ve argued all along that high levels of spin pay off for a politician, and the ability to give high-spin speeches especially to people who do not already like you is a key asset for a politician. Reagan is a good example of this in action.

Spin scores to the end of October

Here is the analysis of levels of spin in Obama and McCain’s speeches up to a few days ago. Usual labelling (refer to previous posts for background).

Spin scores (red - McCain, blue - Obama)

The most obvious thing to see in this plot is how McCain’s speeches all tend to lie on one side of the deceptiveness axis while Obama’s tend to lie on the other side. This is because McCain has started using motion words at high rates (and Obama does not). This has a small effect on deceptiveness score, but rates of use of motion verbs are not all that important to signalling deception.

The individual levels of spin from the convention to the end of October are here:

McCain

Obama

The last time I posted was during the period where Obama’s level of spin was quite low. As you can see, it has risen sharply again in the past week. This suggests that he is not as confident of winning now as he was then — he has consistently shown a pattern of stepping out from his facade and using lower spin when he feels confident about winning.

Update — Spin in US Presidential Election

I’ve looked at the spin in the speeches in the last few weeks, more or less since the convention. The overall picture has remained very similar to earlier stages: Obama’s levels of spin are relatively high compared to McCain’s.

Here is the overall spin plot:

Spin between the conventions and the last week of October

Here are the plots over time for McCain:

McCain's spin

and Obama:

The most interesting thing here is how much Obama’s level of spin has dropped in the last few speeches. This is very similar to what happened in the weeks (late February) when it became clear that he would get the Democratic nomination. When he feels sure of himself, he steps out from behind his election facade and presents himself much more openly. The extremely low-spin speech is his comedy routine at the Al Smith dinner — presumably the expectation that he should be funny rather than serious made him feel as if he had permission to be himself.

McCain’s high-spin speeches are those when he gives economic history lessons, unlike his typical speech in which he puts more of himself.

Canadian Federal Election – 1 day to go

The results for the three English-speaking party leaders are shown below. I have not tried to track Duceppe because we don’t fully understand how to move the deception model across languages. I think there’s a fair amount of confidence that the same (types of words) are significant in other languages, but the details are difficult. In the case of French, the pronoun “on” sometimes plays the role of “I” and sometimes not, and the differences are hard to pick out, and might perhaps be especially significant as a distancing mechanism.

The Canadian federal party leaders have all tried to run presidential style campaigns (Vote the man, rather than the party, or the local member). But they can’t bring themselves to talk about themselves, so the speeches they all give are extremely abstract blue-skies policy speeches, with hardly an “I” to be seen. Relative to the U.S. election, all of the speeches would rank as high spin. This may be partly a perception (probably accurate IMHO) that Canadians are not ready for prime minister = president and so the leaders are trying to be the face of their party, but not the single leading figure. This middle-ground approach seems to be a bit clunky, and produces some odd speeches. Can you imagine an American politician at any level giving a speech titled “New Support for Apprentices”?

Here is the spin ranking for the speeches available up to today:

Spin rankings over the entire election campaign

The pattern has been reasonable consistent over the campaign. Stephen Harper’s (blue) speeches tend to be high-spin; Stephane Dion’s (red) speeches are more moderate in spin, but much more variable; and Jack Layton’s speeches (such as they are) (magenta) are relatively low spin. This would indicate that Stephen Harper will tend to do better than the other two, but these speeches play such a small role in the campaign that not much should be read into this prediction.

Election spin results covered by New Scientist

There’s an article on, among other things, my work on spin the US election campaign in this week’s issue of New Scientist. You can find it online here. There are also some interesting results from the people who look at facial expressions (pioneered by Ekman) and so voice analysis.

Canadian Federal Election — first analysis

I’ve done some analysis of the party leader speeches from the first week of campaigning. I am somewhat hampered by lack of data; only one speech by Jack Layton has been posted, and there are only three by Dion.

About the only conclusion so far is that language patterns are all over the place. There is no consistency in any of the campaigns. This might be because speechwriters are still settling in (but surely Harper’s at least must have known the election was coming?). Or it may be that all of the campaigns are unsure of their strategies in some less-conscious way. But it’s interesting, and slightly puzzling.

I’ll await more data.

Spin in the US Presidential Primaries — Summary

As we enter what looks like it might be the end phase of the primary season, I thought I would summarize what I’ve written about spin during the process.

1. What is spin? People often talk about spin as messing with the content of a communication: leaving bits out, or changing the emphasis. What I’m talking about here is a mental (unconscious) process where a person presents themselves or their content in a way that does not reflect what they know to be true about it. Politicians (and the rest of us) do this, to some extent, all of the time — trying to make a good impression. For a politician, outright lying is a poor idea (recall Clinton under sniper fire) but there is a lot of pressure to be “all things to all men”. Because the communication is not the speaker’s natural persona, this kind of spin produces a detectable signature in the communication.
2. What is the model of spin (deception)? This work is based on Pennebaker’s empirically-derived model of word-usage changes when people are being deceptive. This model is characterized by (a) reduced rates of first-person singular pronouns; (b) reduced rates of exclusive words, words that mark the beginning of a phrase or clause that qualifies or refines what has gone before; (c) increased rates of negative-emotion words; and (d) increased rates of action verbs. These changes are unconsciously produced, so cannot be directly altered by a speaker, even one who knows the model. Although the model was developed for plain deception (outright lying) it seems to detect deception across the full range from lying, through spin, to negotiating and dating.
3. Why does there have to be a context? Because the model relies on increases and decreases in word-usage rates, there must be some kind of context of similar communications or documents to be able to tell whether a given frequency represents an increase or a decrease. Therefore, absolute spin scores cannot be determined — instead we can only rank a set of communications from most to least spinful. Even within the context of the presidential primaries, underlying language use has changed, most obviously from a ”getting to know me” phase to a “getting to know my policies” phase.
4. The early primaries. From the beginning of 2008 until the 3rd week of February, all three candidates were introducing themselves. In the speeches given during this period, McCain has the least spin, followed by Clinton, followed by Obama, with noticeably higher levels of spin. McCain generally used (and uses) high rates of first-person singular pronouns, justifying his ’straight talk” claim; Clinton generally used high rates of exclusive words, adding refinement and qualification to many of her statements. Obama’s speeches were lacking in both: he used “we” at extremely high rates (and “I” hardly at all), and his statements were simple and declarative. This makes for speeches that are light on content but, when well delivered, emotionally uplifting. (Reading Obama’s speeches rather than hearing makes one wonder what all the fuss is about — the speeches themselves are rather dry, and the delivery is everything.)
5. Obama decides he’s won. Over the weekend of February 24th, Obama’s language patterns changed dramatically, becoming very similar to Clinton’s. I conclude that this weekend his campaign did the calculations and decided that Clinton could not win the nomination (which seemed, and seems, mathematically true). He cannot have consciously altered his speech patterns, so this must the result of reframing what’s going on to himself — presumably stepping out from behind the persona he had been using before that and presenting something closer to his real self.
6. The past month. In the past month, both Obama and Clinton show higher levels of spin whenever the pressure on them has increased, and they have become defensive. For example, Obama’s levels of spin jumped back to January levels when the Wright controversy became public. In such situations, Obama’s level of spin is characteristically higher than Clinton’s.
7. Responses versus statements. It is difficult to analyze and compare the debate statements of the candidates with their speeches. The question and answer form naturally changes the rates of word usage: for example, if the question is “Would you…” it’s much more likely that the answer will begin “I will…”. And, to make matters worse, debates are not really question and answer since candidates have prepared statements for likely questions and they will use them regardless of the form (and sometimes the content) of the question. It is not yet clear how applicable the deception model is in question and answer situations, so I have not analyzed the debates, except for the most-recent — where Obama still shows up having higher spin than Clinton.
8. Does spin work? Spin is a two-edged sword. On the one hand, it does work in the sense that it can make a speaker appealing to people who would otherwise not be attracted to him/her — which is why,  of course, politicians use it. On the other hand, if a candidate steps out of character, even briefly, people may realize that it is a facade and react in a strong negative way. And, to make it harder, the facade and the language usage are largely subconscious, so a candidate may misstep without realizing it.
9. “I” versus “we”. There’s a lot of (positive) discussion of Obama’s high rates of use of “we”. This pronoun is irrelevant to deception — people who are being honest or deceptive may use “we” at either high or low rates. However, the difference between using these two pronouns is partially understood. People who are being open and not status conscious typically use “I” a lot, while those who are being closed and status conscious typically use “we” a lot. In particular, “we’ is often code for “you” — commanding without creating the impression of command. In other words, “we” is a weasel word.
10. Growing into a persona. It’s possible that Clinton and McCain have been in politics so long that a persona that they originally assumed has now become so much a part of them that it has become their real personality; and that is why their levels of spin are low. By this explanation, the reason that Obama has such high levels of spin is that he’s a relative newcomer to the US national arena, and so he still “puts on” a persona. This doesn’t seem all that convincing — first, he has a long history in public life, although on a smaller stage; second, he seems to be able to step out from behind the persona when things are going well.

The analysis on which this summary is based (and the figures that go with it) can be found in earlier postings.

Update: US presidential election spin

I’ve updated the spin analysis to reflect the speeches given up to the last weekend in March.

McCain: red dots, Clinton: blue dots; Obama: blue stars.

You can see a transition in the type of speeches, as candidates move away from speeches intended to introduce themselves as people to speeches that express policy. These look inherently more spinful compared to the earlier speeches, at least for McCain and Clinton. Arguably some of Obama’s movement is also because of the same transition.

The trends in spin are shown in the following figures:

McCain:

Clinton:

Obama:

Although there is a general trend to speeches that score higher on spin than earlier in the campaign, Obama still leads the pack in levels of spin. I will pull out the recent speeches and repeat the analysis to get a new baseline from current content, and see whether this pattern holds true. Results in a day or two.

Changing spin over time

I’ve updated yesterday’s results to include the prepared text of Obama’s speech on Iraq yesterday. The new speech, numbered 39 in the figure, is one of his most spinful.

I’ve also looked at how Clinton’s and Obama’s spin has changed over time.

Clinton started out talking fairly straight, but her spin has been increasing (lower on the graph) recently.

Obama’s level of spin has always been quite high. However, after a patch of much less spin, his last two speeches, on race and on Iraq, have had very high levels of spin.

It’s also interesting how much variation there is from speech to speech. Some of this is probably due to the topic; speeches that involve telling one’s own story naturally produce lower spin scores, and all three candidates spend some time helping people get to know them (or their personas ). There may be some variation because of which speechwriter(s) actually created the text. We were surprised, initially, that there were typical levels of spin because any speech is the product of several people, not just one. But it seems like a good speechwriter must produce something that the candidate is able to say in a natural way, and so aligns with what the candidate would have said.

I don’t know why these figures render so poorly. Any suggestions appreciated.

This is the beginning of the Easter weekend. I won’t post again until Tuesday, unless something time-sensitive happens.

Obama — return of the spin

The big speech on race that Obama made yesterday stimulated me to bring my analysis of the level of spin the candidates are using up to date.

Small clips of Obama talking about issues around his pastor sounded like they contained very high level of spin, especially wonderful phrases like “we need to think about that” (code for “YOU need to think about that”)

The figure below shows projections of all of the candidates speeches (as they appear on their websites), projected onto a axis that represents spin. The red dots are McCain speeches; the blue dots are Clinton speeches; and the blue stars are Obama speeches.

For each candidate, the speeches are in data order: 1-5 McCain; 6-21 Clinton; and 22-38 Obama.

Link to a copy for better readability (maybe)

Obama had a major reduction in spin in all of his speeches around Feb 24th. Although his speeches contain more spin than either of the other two candidates, even his speeches are of two different kinds. Some, such as 37 (US Admirals and Generals) and 28 (Reclaiming the American Dream — autobiography) have quite low spin. Others, such as yesterday’s (38), his speech when he lost Texas and Ohio (36), and his Great Need of the Hour speech (23) have high levels of spin.

You can see that the amount of spin in yesterday’s speech worked. The pundits are impressed; it’s the greatest speech about race in a generation, apparently. It was a good speech. But, perhaps, it wasn’t quite the real Obama.

I stick my my interpretation of the changing spin. Around Feb 24th Obama became convinced that he had won the nomination. Every time since then that something has happened to shed doubt on that, any kind of pressure, and he retreats behind his higher-spin language.

Clinton is also feeling the pressure. 3 of her 4 most recent speeches have had much higher levels of spin than she usually uses.

Levels of spin are not consciously controlled (see earlier posts), so they provide an insight into the mental landscape and framing that is driving each of the candidates.

Segment: CBC Radio’s “The House”

I did an interview with CBC Radio’s parliamentary news program “The House” which will be on air on Saturday morning from 9 a.m. on CBC Radio 1 (in Canada and the northern tier of the U.S.).

I talked about spin and deception in the U.S. presidential campaign and the Ethics Committee hearings about the so-called Airbus affair (Mulroney and Schreiber).

Results from the analysis of the Ethics Committee hearings are inconclusive so far, but I will eventually do some more analysis, and perhaps write about the results here.

The audio is
here

U.S. versus Canadian political spin

I’ve gone back and done a little work analyzing the party leaders’ speeches in the Canadian federal election in 2006, comparing the language patterns to those of the current U.S. presidential contenders.

In the overall ranking of spin, 38 of the 50 Canadian speeches I’ve collected rank higher on spin than any of the U.S. speeches.

This probably doesn’t mean that Canadian politicians spin more than U.S. politicians, at least not directly. In general, the Canadian speeches have much shorter sentences and are shorter overall. This limits the opportunity to use exclusive words at all which, in turn, means that they can’t play much of a role in marking deception.

It may be that the Canadian electoral system, in which election campaigns are measured in weeks, rather than the gruelling years of American races, makes it easier for politicians to maintain a facade. It may also be that the Canadian media put less pressure on politicians.

It goes to show the importance of measuring spin (and deception) within a set of contextually appropriate texts. Failing to work out what the norms are in a particular domain can make a particular speech look particularly sincere or deceptive even when it is not.

“I” versus “we” revisited

Many people intuitively think of “I” and “we” as two ends of a continuum, where the ends are something like:

“I”, Hillary Clinton, egotistical, one-person band

and

“we”, Barack Obama, inclusive, working together

In all of the work we’ve done analyzing text, the frequencies of “I” and “we” are uncorrelated, so that knowing one tells you nothing about the other. In other words, there are people who use high rates of both, some who use “I” a lot but not “we”, some who use “we” a lot but not “I”, and some who use very few of both. So any idea of a continuum is rather misleading.

From the point of view of deception, it’s the frequency of “I” that matters: high rates equal openness, low rates equal a facade.

Until a week ago, Obama was using very low rates of “I” and so I’ve argued that he was spinning hard; while Clinton uses much higher rates, and so is presenting something closer to her “real self”. (Since last weekend, Obama has changed his use of “I”.)

It also matters what kind of context the speech is made in, most importantly how much the speaker is constrained in how they can speak. The signals of deception are much clearer in free statements than they are in responses to questions. There are quite large differences in language use between the opening statements in debates, and the subsequent answers to questions, even though politicians are good at using questions as a platform for saying what they wanted to say anyway.

It doesn’t seem to matter whether the speech is scripted, as most official speeches are, or extempore (off the cuff). Obviously a formal speech is not usually written by the speaker. It seems, though, that speech writers must capture language patterns that the speaker is comfortable with to be successful at the job. So scripted speeches tend to have the language patterns of the speaker, and not of the speech writer. It may also be that a speaker adjusts the use of the the little words that are important for detecting deception on the fly; we think of this as making tiny, unimportant deviations from the script, but they may actually be extremely important.

Improving the process of detecting deception

The main problem with Pennebaker’s deception model, discussed in an earlier post, is that the signal for deception for two of the word classes is a decrease in word frequencies. But a decrease with respect to what? Without some care, a text that does not use any of certain word classes can look unusually deceptive, just because it happens not to use certain kinds of words. For example, formal business writing almost never uses first-person singular pronouns (“I”, “me”, “my”) so an occasional text that used one or two might be considered undeceptive just on that basis.

It’s better to consider the deceptiveness of a group of documents or texts from a common domain, and rank them from most to least deceptive, rather than imagining that deception is a kind of absolute property. Then it’s clear that decrease means “frequencies that are lower compared to the norms of documents in this domain”.

It also turns out to be useful to consider the correlations between word usage across the documents in a domain. There may be conventions about the way in which ideas are expressed that go beyond simple word frequencies. So deception detection is improved by considering the correlation among messages, and using it to pick out documents that are more unusual for enhanced consideration. This is the approach I’ve used to look at Enron emails and politicians’ speeches, for example
here for the US presidential election in February 2008, and
for the Canadian Federal election in 2006.

Can a speaker fool the deception model?

If deception can be detected in text, and the properties that signal deception are known (more or less), can a clever speaker or author use this knowledge to come across as truthful?

The short answer is No. The reason is that language production is a deeply unconscious process. Although we can decide consciously what we would like to say, we have much less control over how we say — much less even than we think we do.

A speaker, with some practice, could start to insert more first-person singular pronouns and exclusive words into their speaking, but only by concentrating. But concentration creates other problems: it tends to make the whole delivery sound more stilted, and it consumes processing resources spent thinking about the form of the speech that come at the expense of the content. In other words, to make the speech sound less deceptive, it’s almost necessary to make it more bland, and therefore less effective for whatever reason it is being made in the first place.

In written text, there is more opportunity to work on the signals in the text to make it seem less deceptive. For example, an author could use our deception detection software to measure the deceptiveness of the text, and change a few words to improve it.

There are two problems with this. First, it’s always better to think of deceptiveness in the context of a set of documents from the same domain. It may not be obvious what the norms for a domain are, and so how much the particular document needs to be adjusted. Second, explicitly manipulating word frequencies tends to create unusual documents because that’s not how documents get edited. These kinds of tinkering run the risk of creating an alternative signature which may also be detected by a different kind of analysis.

The bottom line is that language carries all sorts of information at different levels of abstraction, and it’s all consistent when the language was generated in an ordinary way. Messing around with pieces of language as if they were independent quickly breaks some or all of this global consistency.

Obama thinks he has the nomination

In my previous post, I pointed out that Obama has changed the way in which he uses first-person singular pronouns — increasing them dramatically, a strong signal that he has reduced the amount of spin he is using.

I wasn’t sure why this was happening. In the debate this evening, he’s using the same increased level of first-person singular pronouns.

The change is truly dramatic.

I think perhaps he has decided that he has the nomination, and he’s stepping out from behind the careful facade that he’s used to get this far. Certainly he learned something this weekend that has changed his attitude to the campaign. I don’t think that this much of a change could be managed consciously.

It’s a big change.

“I” versus “we”

A number of people have noticed that there are substantial differences in the way Hillary Clinton and Barack Obama use pronouns. This is part, but not the whole story about deceptiveness and spin in communication, as I’ve talked about in earlier postings.

The conventional explanation goes something like this:

“Clinton says “I” a lot because she’s egotistical, or because she plans to get things done, with or without the help of other people. Obama says “we” a lot because he is inclusive and wants to develop a shared dream that all can be involved in.”

This view is completely wrong, although it’s probably true that Obama, at some time, developed his use of “we” because he thought or was told that it would create an inclusive impression.

The reality is almost exactly the opposite. People who use first-person singular pronouns (“I”, “me”, “my”) do so, unconsciously, because they are being open, warm, and low status.

People who use first-person plural pronouns (“we”,”us”,”ours”) are not being as open, especially if the speaker is a man. Men use such pronouns as a velvet glove around an iron fist, a way to command without the appearance of command.

How do I know this? The models of language have been derived empirically, much of it the work of James Pennebaker and his group. They have studied the language that people use in known situations, and derived word usage profiles that have enabled them to draw these kinds of conclusions.

Obama uses the language he does because it works — people do perceive him as inclusive. But that doesn’t mean that that’s his real view of the situation. Clinton has adopted, consciously or not, an approach that presents her real self much more directly

It’s not all about pronouns. Other important components are exclusive words (“but”,”nor” which both McCain and Clinton have used in the past week!); negative emotion words (“angry”); and action verbs (“going”).

Obama’s changing pronouns

I pointed out in an earlier post that first-person pronoun use is a signal for warmth and openness, and that Obama uses such pronouns at very low rates — much less often than Clinton and McCain.

This is changing. Here is an extract from a speech on February 13th (first-person singular pronouns in red):

It was nearly a century ago that the first tractor rolled off the assembly line at this plant. The achievement didn’t just create a product to sell or profits for General Motors. It led to a shared prosperity enjoyed by all of Janesville. Homes and businesses began to sprout up along Milwaukee and Main Streets. Jobs were plentiful, with wages that could raise a family and benefits you could count on.

Prosperity hasn’t always come easily. The plant shut down for a period during the height of the Depression, and major shifts in production have been required to meet the changing times. Tractors became automobiles. Automobiles became artillery shells. SUVs are becoming hybrids as we speak, and the cost of transition has always been greatest for the workers and their families.

But through hard times and good, great challenge and great change, the promise of Janesville has been the promise of America – that our prosperity can and must be the tide that lifts every boat; that we rise or fall as one nation; that our economy is strongest when our middle-class grows and opportunity is spread as widely as possible. And when it’s not – when opportunity is uneven or unequal – it is our responsibility to restore balance, and fairness, and keep that promise alive for the next generation. That is the responsibility we face right now, and that is the responsibility I intend to meet as President of the United States.

We are not standing on the brink of recession due to forces beyond our control. The fallout from the housing crisis that’s cost jobs and wiped out savings was not an inevitable part of the business cycle. It was a failure of leadership and imagination in Washington – the culmination of decades of decisions that were made or put off without regard to the realities of a global economy and the growing inequality it’s produced.

It’s a Washington where George Bush hands out billions in tax cuts year after year to the biggest corporations and the wealthiest few who don’t need them and don’t ask for them – tax breaks that are mortgaging our children’s future on a mountain of debt; tax breaks that could’ve gone into the pockets of the working families who needed them most.

One “I”!

Now look at this extract from February 24th:

Our economy has been struggling for some time now. And as I‘ve traveled across Ohio, I‘ve seen the face of this economy – a mother who told me she can’t afford health care for her sick child; a father who’s worried he won’t be able to send his children to college; and seniors who’ve seen their pensions disappear because the companies they gave their lives to went bankrupt.

I don’t have to tell you about this. Folks around here have been directly impacted by the changes in our economy – whether it was the loss of steel jobs over the past few decades, or the closing of the Ford plant that was here for so long. And folks in this area are still worried about whether they’re going to lose their jobs and how they’re going to make ends meet if that happens.

Now, if we’re honest with ourselves, we’ll acknowledge that we can’t stop globalization in its tracks and that some of these jobs aren’t coming back. But what I refuse to accept is that we have to stand idly by while workers watch their jobs get shipped overseas. We need a president who’s working as hard for you as you’re working for your families. And that’s the kind of President I intend to be.

I‘ve proposed a job-creation agenda that starts with making sure trade works for American workers. We can’t keep passing unfair trade deals like NAFTA that put special interests over workers’ interests.

Now, Senator Clinton has been going to great lengths on the campaign trail to distance herself from NAFTA. Yesterday, she said NAFTA was “negotiated” by the first President Bush, not by her husband. But let’s be clear: it was her husband who got NAFTA passed. In her own book, Senator Clinton called NAFTA one of “Bill’s successes” and “legislative victories.”

And yesterday, Senator Clinton also said I‘m wrong to point out that she once supported NAFTA. But the fact is, she was saying great things about NAFTA until she started running for President. A couple years after it passed, she said NAFTA was a “free and fair trade agreement” and that it was “proving its worth.” And in 2004, she said, “I think, on balance, NAFTA has been good for New York and America.” One million jobs have been lost because of NAFTA, including nearly 50,000 jobs here in Ohio. And yet, ten years after NAFTA passed, Senator Clinton said it was good for America. Well, I don’t think NAFTA has been good for America – and I never have.

This represents a huge shift in word usage — a much more direct presentation of himself. It’s hard to know what to make of this. Perhaps he read some of my earlier commentary or perhaps he feels much more strongly about the content of the second speech, and so let’s more of himself come through (although the content doesn’t seem too different).

Notice that his usage of “we” hasn’t dropped much. Usage of first-person singular and first-person plural pronouns has tended to be independent in the datasets we’re studied.

Finding deception in text

When people speak or write deceptively, there are characteristic ways in which their word usage changes. My work has been built on a model of deception developed by James Pennebaker at the University of Texas, Austin.

Pennebaker’s model was developed empirically, by asking people to write in support of ideas that they did and didn’t believe in. As a result, it works best when applied to statements where the speaker or author is free to express him- or herself without constraints. In other situations, for example question and answer, testimony, business writing, and journalism, the model needs modification. For example, in the presidential debates, it’s much easier to detect how much deceptiveness (spin) is happening in the opening statements than in the responses to questions.

Deception results in:

• Reduced frequency of first-person singular pronouns, perhaps as a way of decreasing distance or ownership of the content. In contrast, increased frequency of first-person pronouns signals warmth.
• Reduced use of exclusive words, words like “but”, “or”, “however”, and “whereas”. These words introduce parts of sentences tha modify the rest and so are markers for increased complexity. Those who are being deceptive are, to some extent, controlling the content in a way that doesn’t happen in completely spontaneous speech, and this take cognitive resources. The extra processing required seems to come at the expense of the complexity of the content.
• Increased use of negative emotion words such as “anger”. This seems to be because of some sense of societal disapproval of deception that colours the speaker’s mental landscape.
• Increased use of motion verbs. This may be because of an attempt to keep the narrative moving and so distract from any inadvertent flaws; or it may be a side-effect of decreasing the overall complexity. Verbs like “go” and “going” are particularly important in this class.

We are not good at controlling our use of these classes of words, nor are we any good at noticing their frequencies, either hearing or reading. That’s mostly why the existence of this side-channel of information about a speaker’s mental state has not been noticed until recently.

Computer programs are, of course, very good at computing frequencies and so they can be used to compute deception scores for chunks of text very easily.

Spin in the US Presidential Election

One of the places where I look for traces of bad guys is in the text they produce. An important aspect of this is detecting when someone is being deceptive.

It turns out that deception is signalled quite clearly by changes in the way a speaker uses function words, little words such as prepositions, auxiliary verbs and pronouns. Unfortunately, we as humans are not equipped with the ‘hardware’ to detect these changes, especially in real-time.

Deception covers a wide range from outright lying to socially acceptable forms such as white lies and negotiation. Political spin falls somewhere in the middle of this range — politicians want to reach out to as many voters as possible and they get tempted into saying things they don’t quite believe. Spin is one of those irregular verbs:

I inform my supporters

You spin

He panders

I’ve looked at the amount of spin in the speeches of the three current contenders in the US presidential election. There are substantial differences. The most important signals of the presence of spin are: low rates of first-person singular pronouns (I, me, my) and low rates of exclusive words, those words that introduce a modifying phrase or clause (or, but, however).

McCain’s characteristic style is heavy with ‘I’ and so we can conclude that he really is talking straight. His denial of the NYT story of his alleged relationship with a female lobbyist is entirely convincing because of his word use when he talked about it at the press conference.

Clinton’s spin is also low, mostly because she uses large numbers of exclusive words — wonkishness coming through. She, also, is presenting more or less the real person she is.

Obama, on the other hand, shows all the signs of high levels of spin. I’m not saying that this is deliberate; but it does seem that he is presenting a facade that is not who he really is, at least to a greater extent than McCain and Clinton.

Obama’s speeches depend, for their success, on his delivery. If you read the text of his speeches, they are quite dry. If he did not use the pronoun ‘we’ so often, they would seem even drier.

His use of ‘we’ is interesting. This pronoun is often thought of, intuitively, as conveying a kind of inclusiveness. When women use it, this is often the case. However, when men use it, it is often as a velvet glove to cover an iron fist. In other words, men tend to use it as a distancing mechanism.

Why is Obama so successful? Because he is, at some level, telling people what they want to hear. This is a risky strategy. We don’t have a lot of history to go by, but it seems as if the candidate who spins the least tends to win elections. It also seems that the candidate who uses first-person singular pronouns more than first-person plural pronouns also tends to win — because this conveys greater optimism, likeability, and humility.
Here are some of the results of my analysis of recent speeches