Tuesday, July 25, 2017

Social Psychology and Legal Writing: Hallmarks of Truthfulness

*This is the second post based on the work of Dr. James W. Pennebaker. Check out my first post for background on his research.

Social psychologist James W. Pennebaker has found that the language a person uses can suggest whether a person is lying or telling the truth. While no person, machine, or program currently in existence (including polygraphs) can detect lies at a rate any better than about 65%, Pennebaker’s studies suggest some hallmarks of truthfulness. According to Pennebaker, in general, people telling the truth use:

  • More words and details
  • Longer and more complex sentences (including use of     exclusive words such as except, but, and without)
  • Fewer emotions
  • Fewer verbs
  • More self-references (I-words)

So why does the use of more words and details and more self-references signal honesty?

Pennebaker suggests that those telling lies speak and write more simply or straightforwardly because they are making up statements about things they did not experience. Truthtellers, on the other hand, use more words and greater detail because they are “distinguishing what did happen versus what did not, what they were thinking and what they were not thinking about, what was in the category and what was not in the category.”

And according to Pennebaker, I-words signal to the listener or reader that the speaker is paying attention to himself, and studies suggest that those who are more self-aware are more humble and honest.

Pennebaker notes that liars sometimes use performatives (statements about statements) to hide their lies. For example, consider this sentence: 

  • Let me be clear: I did not collude with the Russians.

In that sentence, “let me be clear” is the performative. According to Pennebaker, a statement with a performative cannot be assessed for truthfulness because the part of the statement that we care about (here, the assertion about a lack of collusion with Russia) is not directly asserted.

Pennebaker offers a statement by President Bill Clinton as an example. At a January 1998 press conference, Clinton said:

  • I’m going to say this again: I did not have sexual relations with that woman, Miss Lewinsky.

The nation later learned, of course, that Clinton did have sexual relations with Lewinsky. But he was not lying when he said, “I’m going to say this again” because he did, in fact, say it again.

Self-deception is also an issue. People (whether intentionally or unintentionally) deceive themselves, often by expressing overconfidence, says Pennebaker.

  • I am absolutely certain that I can settle the case for $50,000.

To most, this sentence appears (and likely is) less truthful than the following sentence:

  • There is a good chance I can settle the case for $50,000.

Interestingly, juries may be less likely to believe the very language that signals truthfulness. Take, for example, the following sentences from a crime victim:

  • I am 100% sure that the defendant is the man who entered the store and robbed it.

  • The defendant is the man who entered the store where I was working at 9 p.m. on August 31, 2016, brandished a 9 mm gun, and robbed me of all the money in the cash register, about $350.

All other things equal, jury members may find the definitiveness of the first statement more credible than the second, even though the second is more detailed and appears more reasonable (who, really, can be 100% sure of anything?), and, in the second, the speaker refers to himself more.

The Lying Words chapter of Pennebaker's The Secret Life of Pronouns may be of particular interest to criminal lawyers. In that chapter, Pennebaker describes a test he and a private investigator performed. The investigator gathered sworn testimony from two groups of criminal defendants—those who were convicted but later exonerated by DNA or other overwhelming evidence and those who were acquitted but later convicted of perjury. Members of the first group, thus, were “bad” at telling the truth while members of the second were “good” at lying.

The results of that study were consistent with Pennebaker’s earlier findings, specifically with respect to I-words. In their testimony, the later-exonerated defendants used more first-person singular pronouns; that is, I-words appeared to signal innocence, as did bigger words and more detailed descriptions. The defendants later convicted of perjury, on the other hand, used more third-person pronouns, in an apparent effort to shift blame to others.  

So, the next time you're assessing the writing or speech of others, look for Pennebaker's hallmarks of truthfulness. They might help you sniff out a liar! 

Monday, July 17, 2017

Social Psychology and Legal Writing: Part I

*This is the first of a two-part post based on the research of Dr. James W. Pennebaker.*

Dr. James W. Pennybaker
A Twitter friend recently introduced me to the work of Dr. James W. Pennebaker, a social psychologist who researches the relationships between people’s language and their psychological states. Dr. Pennebaker’s findings have interesting implications in legal writing. But first—some necessary background.

Dr. Pennebaker developed a computer program, Linguistic Inquiry and Word Count (LIWC), to analyze the psychology behind what we write. Through his research, Dr. Pennebaker has learned that the words a person uses and the frequency with which the person uses them provide insight into the psychology of the speaker.

Specifically, Dr. Pennebaker is interested in function or style words—pronouns, articles, prepositions, and a few other categories of small words. According to Dr. Pennebaker, these words, which most of us barely notice, account for 60% of the language we use. These stealth words can tell us about whether a person is organized, emotionally stable, social, and many other things.

And Dr. Pennebaker’s research shows that men and women, the old and the young, and people of different socioeconomic statuses use language differently. Specifically, Dr. Pennebaker has discovered that men tend to use more big words, nouns, and words per sentence than women. Women, on the other hand, use more personal pronouns, verbs, and hedge phrases (such as “I think”) than men. Further, according to Dr. Pennebaker, people tend to change their language based on setting. In formal settings, for example, people speak with fewer pronouns, more articles, and fewer social words; that is, the research shows that women adjust their language in these settings to talk like men.

Dr. Pennebaker has also found that our language use changes as we age. Younger writers use personal pronouns and past-tense verbs at higher rates than older people, who use more articles, nouns, prepositions, and cognitive words that reflect insight.

Finally, Dr. Pennebaker has found a “sound of power”—that those with power and status (generally, men, older people, and those of higher socioeconomic status) use more noun clusters and fewer pronouns and verbs. The reasons for these findings aren’t entirely clear, but Dr. Pennebaker suggests that those in power tend to focus on tasks (they are, after all, the deciders) while those with less power tend to pay attention to others (e.g., the deciders).

Given that background, I used Dr. Pennebaker’s website, secretlifeofpronouns.com, to analyze my own writing.

Dr. Pennebaker’s first writing prompt, a perceptual style test, required me to spend five minutes describing a picture. According to Dr. Pennebaker’s research:

I scored above average on functional thinking, tactile sensitivity (appreciation of texture, contour, and dimension), contextual thinking, and visual sensitivity (colors and styles).

And, unsurprisingly, I scored above average on verbal thinking. According to Dr. Pennebaker’s research:

“[I] devoted more time than most describing the words and word fragments. Words grab [my] attention and [I] view them as important anchors to [my] reality.”

That’s certainly true!

The second prompt, a thematic apperception test, required me to spend ten minutes writing a fictional story based on a different picture. My results reflect a higher-than-average need for achievement (unsurprising), use of big words (also unsurprising), and positive emotions. My results show a lower-than-average need for power and negative emotions.

I’m a pretty self-aware person (I think!), and both I and the people close to me agree that the results of these two tests accurately reflect my personality.

I also used Dr. Pennebaker’s other website, analyzewords.com, to analyze my @ladylegalwriter tweets. The results are interesting. According to Dr. Pennebaker’s program:

My emotional style is only of average upbeatness (I’m a little surprised by this.)

My social style is highly personable, plugged in, and arrogant/distant (Maybe I am?)

My thinking style is highly analytic (no surprise there).  

While I was at it, I also analyzed tweets from Barack Obama (@barackobama), who, according to Dr. Pennebaker’s program, has a very upbeat emotional style but only average style in personability and analytics.

Finally, I looked at tweets from lawyer Jeffrey Toobin (@jeffreytoobin) who, Dr. Pennebaker's program says, has a highly upbeat emotional style and a highly personable but spacey social style.**

Dr. Pennebaker notes that many factors that cannot be controlled outside a research setting (distractions, level of tiredness, etc.) may affect the results of all his tests. Thus, cautions Dr. Pennebaker, the results should be taken with a grain of salt.

But the outcome of my writing assessments appears to accurately reflect some of my psychological strengths and weaknesses and has gotten me thinking about the implications of Dr. Pennebaker’s research in legal writing. I’ll discuss this in my next post.

**Caveat: I noticed is that the program only analyzes the tweeter’s most recent 1,000-1,500 words or so and factors in retweets, both of which could skew the results.