The Speech Pattern That Undercuts You Before the Argument Does
By Mira Lindqvist · Voice & Pace
How do you sound more confident when you speak? Cut the habits listeners read as doubt, not the words that carry real uncertainty. Do not stack hedges and hesitations on one claim, keep qualifiers off your evidence, and let statements end on a falling tone instead of rising like a question. Use a short silence instead of "um" before a key point, and precise qualifiers instead of "sort of".
In 1978, a research team led by Bonnie Erickson took a speech style that one of its members, the anthropologist William O'Barr, had identified in real courtroom testimony, and tested what it did to the people hearing it.1 They called it "powerless speech": frequent hedges ("sort of", "I think"), hesitation forms ("um", "well"), a rising, question-like intonation on statements, extra polite forms, and intensifiers used to shore a claim up ("very", "so"). When the same testimony was presented with and without these features, the witness who used them was judged less credible and less favourably overall, although the facts had not changed.
The style is separable from the substance
What makes the finding useful rather than just alarming is that style could be changed while content was held still. The same testimony, built in two versions, produced different judgements from people who were given nothing else to go on. The evidence was not weaker. The delivery pattern alone moved how much it was believed.
That is a sharper result than "confident people are more persuasive". It says that a handful of specific, learnable habits carry a cost of their own, separate from anything about the speaker's actual knowledge.
Which parts do the damage
Powerless speech is a bundle, and later work pulled it apart. In two studies, Lawrence Hosman tested hedges, hesitations and intensifiers separately and in combination.2 Hedges and hesitations each lowered how authoritative a speaker seemed on their own, without needing the rest of the bundle. Hosman also found that the speaker's status changed how the style was judged, which is a reminder that the same words land differently depending on who says them.
A 2008 study by Amanda Durik and colleagues asked a more precise question: where in an argument does a hedge hurt?3 People read an editorial arguing for a university policy. When the hedges were placed on statements of data, the factual claims the argument rested on, readers thought less of the policy, the source and the argument. When hedges were placed on the interpretation of the data, they did little harm, unless the hedges were colloquial in tone rather than the professional kind a careful writer uses.
So the cost is not spread evenly. A hedge on your evidence undercuts the evidence. A measured hedge on what the evidence means can be read as care.
Hedges aren't inherently a flaw
It would be a mistake to read any of this as "never hedge". Janet Holmes's 1990 study of hedges and boosters (words like "really" and "definitely" that intensify a claim) found that both do several jobs in conversation.4 A hedge can mark genuine uncertainty, and it can also soften a statement out of politeness or keep a conversation cooperative. A word that looks like weakness on a transcript is often doing social work.
Ken Hyland's research on hedging in scientific writing makes the stronger case: careful hedging is often the honest choice.5 A claim stated with the uncertainty it actually has is more accurate than the same claim overstated, and it is no weaker for that. Overclaiming has its own cost, because a listener who catches you overstating once tends to discount what comes after.
Density and function
What separates powerless speech from careful hedging is density and function, since both contain hedges. Erickson's pattern is a cluster of markers stacked on top of each other, applied by reflex rather than where real uncertainty exists, often on claims the speaker is sure of. A single "I think" on a claim you are tentative about is calibration. Two hedges, a hesitation and a rising tone on a fact you know well is the pattern that cost Erickson's witnesses their credibility. It is a habit, and it says little about how sure the speaker is.
How our scoring treats it
Exprea's hedging-rate scoring counts common hedging phrases, such as "I think", "maybe", "kind of" and "sort of", as a rate per hundred words. It does not penalise a single hedge, and it cannot tell a warranted hedge from a nervous one. It matches phrases. What it can show is density, which is what Erickson's pattern is made of. In the Coach's clarity scoring, about one hedge per hundred words sits in the good band, two in the acceptable band, and four or more in the poor band. A high rate on material you know well is worth a look. A high rate on genuinely uncertain material may simply be accurate.
What to do
Watch the cluster, not the word. Listen back to a recording and mark the places where two or more markers stack up. Those are the sentences to work on.
Keep hedges off your evidence. State what you know plainly: the figure, the result, what happened. If you need a qualifier, put it on the interpretation, where it reads as judgement.
Use precise qualifiers. "In the two teams we measured" carries the same honesty as "sort of" and none of the cost. Durik's results suggest the type of hedge matters as much as its presence.
Let statements end down. A rising tone turns a claim into a question. Letting the last stressed word fall is often enough to remove the question mark.
Replace "um" with silence. A hesitation form before a key claim signals searching. A short pause in the same place signals the opposite. The difference is covered in what "um" tells your audience.
The goal is not a speaker who never qualifies anything. It is one whose hedges track their actual confidence, rather than running on autopilot.
References
- Erickson, B., Lind, E. A., Johnson, B. C., & O’Barr, W. M. (1978). Speech style and impression formation in a court setting: the effects of “powerful” and “powerless” speech. Journal of Experimental Social Psychology, 14(3), 266–279. https://doi.org/10.1016/0022-1031(78)90015-x
- HOSMAN, L. A. (1989). The Evaluative Consequences of Hedges, Hesitations, and Intensifies Powerful and Powerless Speech Styles. Human Communication Research, 15(3), 383–406. https://doi.org/10.1111/j.1468-2958.1989.tb00190.x
- Durik, A. M., Britt, M. A., Reynolds, R., & Storey, J. (2008). The Effects of Hedges in Persuasive Arguments: A Nuanced Analysis of Language. Journal of Language and Social Psychology, 27(3), 217–234. https://doi.org/10.1177/0261927X08317947
- Holmes, J. (1990). Hedges and boosters in women’s and men’s speech. Language & Communication, 10(3), 185–205. https://doi.org/10.1016/0271-5309(90)90002-s
- Hyland, K. (1998). Hedging in Scientific Research Articles. John Benjamins. https://openlibrary.org/works/OL85756W