In the past few weeks, a (polite) battle has been raging in the media about the use of filler words—“umm,” “uh,” “so,” “like,” “I mean,” “you know.”
In The New York Times, Christopher Mele argued that we should stop using them.
In Quartz, Jena Barchas-Lichtenstein replied that we should stop demonizing them.
On Twitter, Wharton professor Adam Grant took Barchas-Lichtenstein’s side:
Don’t drop “um” & “like”: they help us soften disagreements, introduce delicate topics.
Weak language has strengths: https://t.co/AseL5VLMbl— Adam Grant (@AdamMGrant) March 25, 2017
Stanford professor Tina Seelig disagreed, replying:
I really disagree with this. From my perspective, these filler words are sloppy. There are always more accurate ways to communicate. https://t.co/41lg1DdcqZ
— Tina Seelig (@tseelig) March 25, 2017
Where does The Science of Speaking stand?
As I note in the book, there are two different contexts here, which everyone seems to be conflating. One is that of formal presentation (standing and delivering in front of an audience), the other, that of informal conversation (personal communication, one-on-one).
I discovered this distinction several years ago while teaching public speaking at Stanford. While I have succeeded in almost completely eliminating filler words from my formal speech over the years (to good effect), I noticed that I still used them extensively when giving my students feedback in one-on-one speaking tutorials, in the way that Barchas-Lichtenstein and Grant note—to introduce delicate topics. I also noticed that as an introvert, I often use filler words in conversation as a way to indicate that I’m thinking and not quite ready to cede the floor.
So, umm, what’s the, you know, bottom line: are filler words good or bad?
As someone who has seen thousands of speeches over the years, I completely agree with Mele and Seelig in the formal context: when you’re standing and delivering, filler words should generally be eliminated. This is because research (and common sense) has shown that they make you sound less knowledgable. In one study, for example, researchers found that even when a speaker paused silently for a full five seconds, they were rated as more knowledgable than when they said “um” or “uh.” (Of course, a shorter silent pause was even better.)
In Your Perfect Presentation, public speaking coach Bill Hoogterp presents a great analogy for what filler words can do to your message:
Let’s try a little experiment. Fill a glass or cup one-fourth full with a beverage you like—coffee, soda, something flavorful. Now add plain water to the same glass until it is three-fourths full. How appetizing does it look now?
In theory, it shouldn’t be a problem. Water has no taste, so it should have no effect. The same should be true for all the ums, basicallys, and other weak language. They don’t mean anything, so what’s the harm.
Take a sip of the watered-down drink. How did it taste?
That is what it tastes like to other people’s brains when we use weak language. It dilutes and weakens the power of your message.
In an informal context, however, I will grant it to Grant and Barchas-Lichtenstein. In Give and Take, Grant cites research showing that “powerless language” can actually have benefits in one-on-one settings. For example, when telling a partner which items you think are most important in a survival situation, adding weak language like questions and hedges can actually increase your status and influence. Specifically, saying “Do you think the flashlight should maybe be rated higher? It may be a pretty reliable night signaling device” leads to higher status and increased expectations of success in a future cooperative task than saying “The flashlight needs to be rated higher. It is the only reliable night signaling device.” To adapt Hoogterp’s analogy, sometimes black coffee needs cream and sugar to be palatable.
In the end, it seems, this disagreement is really much ado about nothing. As always, the best speaking strategy depends on your audience and what you’re trying to achieve. If you’re trying to confidently deliver your message, and filler words keep creeping in uninvited, by all means, stomp them out. But if your goal is actually to soften your message, by all means, use the power of powerless language. Both strategies can be effective at different times—you don’t have to pick just one.
Update (3/28/17): I tweeted at everyone involved:
The value of filler words depends on your goal. https://t.co/WH7wP9lTYx @MeleChristopher @neoyorquinanerd @AdamMGrant @tseelig @billhoogterp
— Science of Speaking (@scienceofspeak) March 27, 2017
Adam Grant and Tina Seelig liked the tweet, and Barchas-Lichtenstein replied:
Good analysis, tho I’d say there’s slippage — I suspect @MeleChristopher and I don’t agree on whether most workplace conversation is formal https://t.co/FGRrSBykeD
— Barchas-Lichtenstein (@neoyorquinanerd) March 28, 2017
and
@scienceofspeak @MeleChristopher @AdamMGrant @tseelig @BillHoogterp I dig the analysis … but what is and isn’t formal = always contested.
— Barchas-Lichtenstein (@neoyorquinanerd) March 28, 2017
To which I replied:
@neoyorquinanerd @MeleChristopher @AdamMGrant @tseelig @BillHoogterp Absolutely. For this reason, I’m starting to prefer my “goal” approach.
— Science of Speaking (@scienceofspeak) March 28, 2017
and
@neoyorquinanerd@MeleChristopher@AdamMGrant@tseelig@BillHoogterp i.e., adapt strategy based on goal, rather than based on context.
— Science of Speaking (@scienceofspeak) March 28, 2017
Update (3/28/17): A few weeks later, Lindsay Dodgson wrote in Business Insider about some of the other reasons we use filler words.
Nick:
What psychologists think also is important, and often ignored. See the October 2012 article by Duane G. Wilson at Psychological Science Agenda titled More than words: Disfluencies, emphasis and gesture aid in communication. I blogged about it at Joyful Public Speaking in a February 13, 2014 post titled Adding a few ahs and ums improved recall of plot points in stories.
Thanks for the article, Richard – I’ll definitely cite it in the future.
What I want to see now is the same study repeated with the control being a silent pause instead of a cough, as this study leaves open the possibility that a silent pause > a filled pause > no pause > a cough. Or maybe a filled pause would still come out on top (or there won’t be a difference) – we won’t know for sure until we try it!