How "whopping" is 78 percent monosyllables?
« previous post | next post »
The other day, someone asked me about the claim that "a whopping 78 percent of the words that Trump uses are monosyllabic".
We've previously debunked the idea that Trump's speeches aim at a fourth-grade reading level ("More Flesch-Kincaid grade level nonsense", 10/23/2015).
And long ago, we took aim at careless assertions about how young people/media/txting/etc. are degrading the language to the point that "the top 20 words used … account for a third of all words": "Britain's scientists risk becoming hypocritical laughing stocks, research suggests", 12/16/2006; "Only 20 words for a third of what they say: A replication", 12/16/2006; "Vicky Pollard's revenge", 1/2/2007.
So here's a quick evaluation of that "78 percent" claim.
Running a quick script over the lexical histograms for the third presidential debate, I get
So Mr. Trump is not quite at 78% — but Senator Clinton is not far behind him.
As another point of comparison, I ran the same script over Terry Gross's 9/29/2014 Fresh Air interview with Lena Dunham, on the theory that Ms. Gross is widely recognized as the epitome of an intelligent conversationalist:
The percentage of monosyllabic words means something, I guess — but surely it doesn't mean what the author of the "78 percent" claim meant it to imply.
[Note: For the Fresh Air interview, I used my own "disedited" transcript, which includes filled pauses (the monosyllables UM and UH) and false starts (many of which are monosyllabic). If all disfluencies were eliminated, the percentage of monosyllables would be a bit lower — but not by much. Thus removing all filled pauses would put Terry Gross at about 78.8% monosyllables.]
Guy said,
November 4, 2016 @ 11:14 am
Numbers in political discussion aren't supposed to be compared to other numbers, they're supposed to sound big (or sometimes small) when presented in isolation.
DWalker07 said,
November 4, 2016 @ 2:22 pm
Guy is right: Don't try to give meaning to claims made in political discussions!
Jonathan said,
November 4, 2016 @ 2:39 pm
Guy is not right: 72.3 percent of the claims should be debunked.
NSBK said,
November 4, 2016 @ 5:19 pm
Any particular reasoning for using the title Senator Clinton vs Secretary Clinton?
Also, I wonder what kind of "syllabic distribution" is found in pre-written remarks/speeches vs off-the-cuff ones, even accounting for filled pauses and false starts.
D.O. said,
November 4, 2016 @ 9:00 pm
Just a technical question: are you counting word stems or complete words? Say, is fishing counts as monosyllabic?
[(myl) Word forms.]
David Marjanović said,
November 5, 2016 @ 12:05 am
"not intended as a factual statement"
Y said,
November 5, 2016 @ 1:17 am
What are the percentages when you take out function words? I imagine in, the, to, and, etc. are comparably abundant in the speech of intellectuals and clods.
Chris said,
November 5, 2016 @ 3:48 am
Shakespeare's use of monosyllables could go way above 78 percent.
"So long as men can breathe or eyes can see, /So long lives this, and this gives life to thee."
Sidney Wood said,
November 5, 2016 @ 6:06 am
Chris is right. Many years ago, I did a similar word count on various genres of English text. They all landed around %70+. One reason could be loss of syllabic inflections in the past, and other reduction processes from what used to be longer words. The morphology of a language is involved, for example Swedish and Finnish words tend to e longer than English.
BackOfBeyond said,
November 5, 2016 @ 9:24 am
I do both Russian and Bulgarian and yesterday, I found some extended Bulgarian blogs from Blog.bg so I prepared both Bulgarian and Russian print text for my wife to see if she could spot the difference. The one thing I noticed is that Bulgarian is full of monosyllables while they are rare in Russian. I have done no analysis but I wondered if anyone could come up with a comparison between those two languages
Coby Lubliner said,
November 5, 2016 @ 11:56 am
BackOfBeyond: One possible reason for the difference between Russian and Bulgarian is that Bulgarian lacks case endings, which usually add a syllable to Russian words.
Guy said,
November 5, 2016 @ 4:38 pm
@Sidney Wood
Also orthographic conventions probably play a role (assuming we're using orthographic words as the standard, which I'm sure we are). Whether something is conventionally written as separate words or one often has little to do with whether the classification as one word or multiple words is better supported from a grammatical perspective.
Neal Goldfarb said,
November 6, 2016 @ 12:04 am
Isn't it the case that common words tend to be shorter than uncommon ones? If that's right, it would go at least part of the way toward explaining the high percentage of monosyllables.
Has anyone ever plotted out the percentages of monosyllables, di-, tri-, etc. among, say the most frequent 50,000 words? And then maybe break the results down by quartile or decile or something? Or does anyone in the audience want to take a stab at it?
Sidney Wood said,
November 6, 2016 @ 8:43 am
@Guy
That's correct, orthography is relevant. I took that into account. For example the bird species "blackbird" and "blue tit" are compounds and each counts as one (fixed accentuation BLACKbird, BLUE tit). These are sometimes, but not always, written as one. But "black bird, blue tit, fish soup" are examples of attribution, usually accented black BIRD, blue TIT, fish SOUP, but the accent can be moved according to rule. These count as two each.
@Neal Goldfarb
True as a generalization perhaps. But there are language specific differences of monosyllable frequency in continuous texts. I just checked a Finnish and a Swahili text and found 10% monosyllables. But English texts are usually around 70% monosyllables.
chris said,
November 6, 2016 @ 4:48 pm
@Guy, Sidney: for example, "setup", the noun, is one disyllabic word, and "set up", the verb, is two monosyllabic ones (and many similar examples, of course). If English didn't have that word-fusing capability, its monosyllable percentage would go even higher. ("Didn't" is an even stronger example…)
Other language specific effects: some languages lack articles, and when they do have them, usage frequency varies; English uses a lot of helping verbs where other languages have more complex verb inflections instead; some pronouns can be dropped when they're implied by a verb form; case languages don't always need a preposition to express a relationship that would require one in less inflected languages; etc.
10% still sounds really low to me, but I don't know either of those languages. Maybe they're compulsive compounders like German?
P.S. Several Trumpisms that have been discussed here consist entirely of monosyllables, e.g. "the blacks" and "big league" (at least, if you're not analyzing it as "bigly"). If he actually does say those things in situations where another speaker would say something else, then that could be reflected in his monosyllable numbers.
Mark Meckes said,
November 7, 2016 @ 9:46 am
How much does Gross's number change if you cut out all the repetitions of "This is Fresh Air"?
Wolfgang Behr said,
November 7, 2016 @ 10:13 am
Claims about monosyllabicity being tantamount to simplicity (of language, thought, culture etc.) have been around since the early missionary accounts of Chinese (see, e.g., Gustav Ineichen, "Historisches Zum Begriff Des Monosyllabismus Im Chinesischen", HIstoriographia Linguistica, Vol. 14:3 (1987), pp. 265–282DOI: 10.1075/hl.14.3.04ine) and they seem to recur every year in Classical Chinese 101 courses. One nice way of making student think about it is to have them read George Boolos (1940-1996) "Gödel's Second Incompleteness Theorem Explained in Words of One Syllable", Mind 103/409, 1994), available as .pdf here: www2.kenyon.edu/Depts/Math/Milnikel/boolos-godel.pdf).
Cervantes said,
November 7, 2016 @ 9:01 pm
Wolfgang Behr said:
That PDF rendition is incomplete: it leaves out George's (non-monosyllabic) explanation of his explanation. Those who want to compare and contrast might take a look at the complete paper.