Language Log

« previous post | next post »

In "Reading Instruction in the mid 19th century" (8/16/2025), I underlined the old-fashioned focus on "elocution", in which readers were trained "to convey to the hearer, fully and clearly, the ideas and feelings of the writer". Much of today's reading instruction turns that into measures of "oral reading fluency", measured as words correct per minute ("wcpm"). This can result in high-scoring readers like those described in this passage from the Introductory Remarks in 1844 edition of McGuffey's Rhetorical Guide, which warns against the consequences of failure to teach "elocution" from the very start:

The habit of defective articulation is generally contracted in the first stages of the learner’s progress, and arises either from indolence, which produces an indistinct and drawling utterance, or from too great haste, which leads to running words together, and to clipping them by dropping unaccented words and final consonants.

Habits of this kind, frequently, indeed, generally, become so inveterate by the time the pupil is sufficiently advanced to use a work on rhetorical reading, or any treatise on elocution, that the most constant and unremitting attention is necessary on the part of both teacher and pupil, in order to correct them. Nothing but a resolute determination to succeed, and faithful practice upon exercises selected with especial reference to the end in view, can accomplish this object. There must be added to this, a constant watchfulness against relapse, when the learner comes to lessons of a more general character.

A monotonous style of reading and speaking, is often formed at the same early age. The little reader is apt to prolong the sound of the word he has just deciphered, until he can " spell out " the one which follows; and if he is hurried from one lesson to another, without having time given him to practice upon that with which he is already familiar, his progress may seem rapid: but he is not learning to read, in the proper sense of the word, that is, to give utterance to words with that modification of voice which their relation to each other demands : he is only becoming familiar with the appearance of words, so as to call their names readily.

There's also widespread current interest in things like the "multidimensional fluency scale", as a way of evaluating student reading performance that goes beyond wcpm. And there are some obvious and easy ways to approach the automatic evaluation of the non-wcpm aspects of student reading, by alignment with a large-enough set of model performances.

But this brings up a problem for linguistic theory. Fluent and rhetorically effective adults don't generally talk the way they read.  Sometimes the differences are minor, and sometimes they're extremely striking.

For example, here's Paul Krugman reading the introduction to his 2020 book "Arguing with Zombies":

Punditry was never part of the plan. When I finished graduate school in 1977, I envisioned a life devoted to teaching and research. If I ended up playing any role in public debate, I assumed it would be as a technocrat—someone dispassionately providing policymakers with information about what worked and what didn’t.

And if you look at my most cited research, most of it is pretty apolitical. The list is dominated by papers on economic geography and international trade. These papers aren’t just apolitical; they’re mostly not even about policy. Instead, they’re attempts to make sense of global patterns of trade and the location of industries. They are, to use the economics jargon, “positive economics”—analysis of how the world works—not “normative economics”—prescriptions for how it should work.

But in 21st-century America, everything is political. In many cases, accepting what the evidence says about an economic question will be seen as a partisan act. For example, will inflation surge if the Federal Reserve buys a lot of government bonds? The clear empirical answer is “no” if the economy is depressed: the Fed bought $3 trillion in bonds after the 2008 financial crisis, and inflation stayed low. But assertions that Fed policy was dangerously inflationary became, in effect, the official Republican view, so simply recognizing reality became seen as a liberal position.

Indeed, in some cases even asking certain questions is seen as a partisan act. If you ask what is happening to income inequality, quite a few conservatives will denounce you as un-American. As they see it, even bringing up the distribution of income, or comparing the growth in middle-class incomes with those of the rich, is “Marxist talk.”

I think McGuffey would approve, at least of Krugman's prosody.

But Krugman's spontaneous conversational speech is very different — here's the start of his 8/23/2025 interview with Phillips O'Brien:

Okay hi everyone. Paul Krugman again.
I'm
um
speaking
uh for a second time with Phillips O'Brien,
uh ((a)) military historian.
's had a lot of influence on how I think-
had a lot of influence on how I think about-
about history even before Ukraine.
uh ((and)) I think his book, How- How the War Was Won, completely changed
how I thought about World War II.
And he's been one of the most uh
uh ((you know)) often contrarian but almost always right when he is commentators since this whole Ukraine thing started.
And um
we're- I thought that after this past week with
uh Alaska and then uh uh the- the gang of Europeans
coming- coming to- to D.C.
would be a good time to check in again.
So hi Phillips.
O’Brien: How ya doin Paul?
um okay um so ((there's)) several things I want to cover but
I- I want to get to your- to your new book
towards the end and I want to talk about the diplomacy or whatever it was that we just saw
uh shortly,
but I'd
like to talk a little bit about the war.
Um
and uh
um
you um
you- you wrote recently uh that- uh
that
uh
uh
we have the worst military analysis community in the history of
military analysis communities and-
wanna talk about that yet?
cause I- I- I think I have some notion of what's- what-
what you mean and uh

And it's not just Paul Krugman. Elon Musk's "elocution", in spontaneous speech, is equally problematic, although this hasn't prevented him from persuading many people of many things.

Here's a short passage from my 3/24/2024 post "'Gentle onsets' are everywhere":

if-
if- if- if given a choice
where
an advertiser is saying like you have to censor all this content ((that-))
on the platform irrespective of where their advertising appears
uh then o- our answer will be like you-
you- you can choose where you want your advertising-
what you want your advertising to appear next to
you can't insist on censorship of the entire platform
if you insist on censorship of the 
entire platform
even where your advertising doesn't appear
uh then uh
obviously we won't- we will not uh
want them as an advertiser

See (and hear) also here.

It should be clear that the ways that spontaneous speech differ from reading are not the same as the differences that bad reading differs from good reading — but I don't think we have good methods for identifying and quantifying the differences, especially automatically.

This doesn't really matter for reading instruction, since good readers don't do any of the things that spontaneous speakers do. But still…

---

A (too large) sample of relevant past posts:

"And uh — then what?" 1/5/2004
"Reanalysis — and not" 2/8/2004
"Status and fluency" 5/11/2004
"Clarifying status in Wolof by fake disfluency" 5/20/2004
"The meaning of eh", 5/1/2005
"um, em, uh, ah, aah, er, eh" 5/2/2005
"Canadian 'eh' and Japanese 'ne'", 5/2/2005
"The the the and the thee the" 7/26/2005
"Young men talk like old women" 11/6/2005
"Trends in presidential disfluency" 11/26/2005
"Trembling to be wrong" 12/20/2005
"I mean, you know" 8/19/2007
"The phonetics of flop sweat" 9/26/2008
"'Babbling points' from all over" 9/30/2008
"Bebop language?", 11/16/2008
"Speaking (in)coherently" 11/20/2008
"More (dis)fluency and (in)coherence" 12/31/2008
"Who knows?" 1/7/2009
"Uh accommodation" 4/2/2010
"If you will" 7/29/2011
"Repetition disfluency" 8/15/2011
"Fillers: Autism, gender, and age" 7/30/2014
"More on UM and UH" 8/3/2014
"UM UH 3" 8/14/2014
"Male and female word usage" 8/7/2014
"Educational UM / UH" 8/13/2014
"UM / UH geography" 8/13/2014
"UM / UH: Life-cycle effects vs. language change" 8/15/2014
"Filled pauses in Glasgow" 8/17/2014
"ER and ERM in the spoken BNC" 8/18/2014
"Um and Uh in Dutch" 9/16/2014
"UM / UH map in the media" 9/17/2014
"UM / UH in German" 9/29/2014
"Um, there's timing information in Switchboard?" 10/5/2014
"Trending in the Media: Um, not exactly…" 10/7/2014
"UH / UM in Norwegian" 10/8/2014
"On thee-yuh fillers uh and um" 11/11/2014
"UM / UH update" 12/13/2014
"Trump's eloquence" 8/15/2015
"Donald Trump's repetitive rhetoric" 12/5/2015
"Trump's rhetorical style" 12/26/2015
"More about UM/UH on the Autism Spectrum", 4/17/2016
"Some phonetic dimensions of speech style", 4/9/2016
"Gertrude Trump" 6/19/2016
"Some speech style dimensions", 6/27/2016
"The em-dash candidate" 8/15/2016
"The rhetorical style of spontaneous speech" 8/16/2016
"The narrow end of the funnel" 8/16/2016
"Disfluencies and smiles" 9/30/2016
"Uh" 10/12/2016
"Mistakes" 3/8/2017
"Fluent disfluency" 3/12/2017
"Presidential fluency" 10/31/2017
"SOTU interpolations", 2/6/2018
"World disfluencies" 5/16/2018
"'Um, tapes?'", 1/20/2019
"Qualifying fluency" 6/4/2019
"Dysfluency considered harmful" 5/19/2019
"Communicative disfluencies interpolations", 12/14/2019
"Disfluency stylings: On beyond hesitation", 7/20/2020
"Donald Trump, now with more filled pauses", 1/3/2021
"The meaning of filled pauses", 2/5/2022
"Read vs,. spontaneous speech", 10/16/2023
"Spontaneous SCOTUS", 3/2/2024
"'Our digital god is a CSV file?'", 3/7/2024
"'Gentle onsets' are everywhere", 3/24/2024
"The future of AI", 6/1/2024
"The evolution of verbal interpolations", 7/6/2024
"More (dis?)fluent interpolations", 8/16/2024

 

 

August 27, 2025 @ 8:14 am · Filed by Mark Liberman under Prosody

Permalink