Language Log

In "Political sound and silence II" I noted a large difference in measures of speaking rate across the Weekly Addresses of the past three American presidents:

And in "Trends in presidential speaking rate", I showed that

… across the first few months of Weekly Addresses, President Trump is tending to talk somewhat faster as well as using a higher range of pitches.

I wondered about the causes:

This might simply reflect a choice to use more words per pause group — pre-pausal lengthening would thus be amortized over a large number of words, and similarly for phrase-final lowering. Or perhaps the distribution of words per pause group is stable, with the speaking rate and pitch range increasing for phrases of a given length. Or both. I’ll test these hypothesis another morning.

In case you missed it — Ben Zimmer recently turned his meticulous scholarly attention to the lexicographical and metrical analysis of shit-gibbon: "The Surprising Rise of the 'S—gibbon'", Slate 2/9/2017.

The metrical part:

Shitgibbon has a lot going for it, with the same punchy meter as other Trumpian epithets popularized last summer like cockwomble, fucknugget, and jizztrumpet. (Metrically speaking, these words are compounds consisting of one element with a single stressed syllable and a second disyllabic element with a trochaic pattern, i.e., stressed-unstressed. As a metrical foot in poetry, the whole stressed-stressed-unstressed pattern is known as antibacchius.)

Over the past decade, I've noticed that Barack Obama's speaking style often involves short, definite-sounding phrases with steeply falling pitch. For an example, take this clip from his 2009 inaugural address:

I tried to quantify — or at least visualize — some of the temporal aspects of this pattern in "Political sound and silence", 2/8/2016, comparing Obama with G.W. Bush in terms of the distribution of speech segment and silence segment durations.

We can visualize (some aspects of) the associated pitch patterns by looking at dipole difference statistics of f0 estimates, as discussed e.g. in "More on pitch and time intervals in speech", 10/15/2016; "Carl Kasell: diabolus in musica?", 11/5/2016; "Some visualizations of prosody", 10/23/2016. This analysis yields a two-dimensional density plot, where one axis represents time differences and the other axis represents f0 differences. And a syllable-scale plot of f0 dipole difference statistics, from the whole of Obama's 2009 inaugural, does support the intuition about the preponderance of rapid local f0 falls:

Inspired by "Trumpchant in B flat", Joel Roston sent me a link to his 1/22/2014 post "How's Carl this time?", where he proposes that

As the excitement builds over the course of each hour-long Wait Wait…Don’t Tell Me! episode, Carl Kasell’s exclamation of the last two syllables of the word “Chicago,” commensurately, rises in pitch.

This is an example of Carl Kasel's performance of  the word "Chicago", in the context of the obligatory periodic station identification in the cited show:

And here are the 30 instances of "…cago" that Joel investigated — six station breaks from each of five shows:

During the last presidential debate Donald Trump started off speaking in a deliberate and controlled manner that soon gave way to his usual animated style. In an informal exchange, Cynthia McLemore observed that he was manipulating his pitch range, as he did in the second debate as well — beginning with a narrow range that dramatically widened as the debate progressed. This post is an attempt to quantify that pitch range variation with a metric Neville Ryant and I have found useful in other work.

This post presents some stuff I did last March — I thought I had blogged about it but apparently I only put it into these lecture notes. It came up in some discussions today in Shanghai, because I thought that maybe similar visualizations might help explore prosodic differences between the speech of English native speakers and Chinese learners of English. This is going to get a little wonkish, so let's start with a picture:

The opening phrase of Donald Trump's speech in Mannheim PA, 10/1/2016, was sung on a single well-controlled pitch:

The fundamental frequency of this monotone chant is about 238 Hz, to which the closest tempered pitch class, at concert A=440, would be the B flat below middle C at 233 Hz. And the next phrase is about a semitone lower, at about 218 Hz, pretty close to A 220:

I haven't heard this type of chanting before from Mr. Trump, or indeed from any other political figure. (But see "Trump's prosody", 8/8/2016, for a different sort of sing-song delivery…)

This is a guest post by Cynthia McLemore, following up on Ben Zimmer's post on "'Uptalk' in the OED", 9/12/2016.

Twenty three years after James Gorman coined a word for “those rises” in the New York Times and unleashed a viral phenomenon associated with my name, and on the occasion of the OED's latest entries, Language Log has invited me to take stock of my experiences and offer some comments.

First, some background. In the late 1980s I started working to construct a theory of intonational meaning in English from the ground up. My aim was to gather facts about the intonational system as they occurred in natural settings in order to understand the role of culture and context in meaning-forming processes. I chose a sorority as the community to study because it had features of a natural speech “lab”: a social hierarchy, age stratification, recurrent contexts with consistent roles and expectations, homogeneity in ethnicity, gender, age, social class, religion, and regional affiliation, and pressure on speakers to conform to norms. In other words, identifiable socio-cultural parameters and reduced sources of variation.

One of the recurrent intonational forms I recorded and analyzed was a phrase-final rise used to introduce certain types of monologues in meetings and structure certain narratives. My Linguistics 101 students told me they heard it around campus and associated it with sororities. But while I was holed up in the lab scrutinizing pitchtracks of the sorority speech data, a broader use of those same phrase-final rises was spreading through American culture more generally. By 1991, when I started presenting my research on the more particular uses I’d found in the sorority — and the more abstract meanings I proposed for the intonational forms themselves — I was overwhelmed with invitations from various academic departments around the country, in addition to conferences, and gave over forty talks in little more than a year. Wherever I went, cab drivers, colleagues, friends and fellow travelers gave me their observations and opinions about “those rises.” Media interest was gaining in 1992 and 1993, but went right off the charts in August 1993 when the NYT published Gorman’s piece.

The latest quarterly update to the online Oxford English Dictionary includes a metalinguistic term all too familiar to Language Log readers: uptalk, defined as "a manner of speaking in which declarative sentences are uttered with rising intonation at the end, a type of intonation more typically associated with questions." It's high time that the OED created an entry for the word, given that it has had a significant media presence (for better or for worse) ever since it burst on the scene in 1993.

The big political story of the past 24 hours: Stephen K. Bannon, formerly the Executive Chairman of Breitbart News, has taken over as "chief executive" of Donald Trump's presidential campaign.

The big linguistic story of the past 24 hours, at least here at Language Log: an exchange between Mark Liberman and Geoff Pullum about the rhetorical style of spontaneous speech, as it applies to the linguistic analysis of Donald Trump's rallies.

Are we seeing the first signs of discord at Language Log Plaza? Mark Liberman seems to be flatly rebutting Geoff Pullum's "no structure at all" remark about what he calls "Trump's aphasia." Mark maintains that Trump's speaking style is no different in kind from any other human's spontaneous speech, even crediting him with "eloquence." Geoff, by contrast, seems to regard Trump as barely capable of expressing himself in human language. This looks like the beginnings of a proper scholarly punch-up. Is Liberman pro-Trump and Pullum anti? Have Mark and Geoff fallen out?

I want to follow up on my post about Daniel Libit's presentation of reporters' and transcriptionists' complaints about Donald Trump's speaking style ("The em-dash candidate", 8/15/2016). Libit uses words like "unintelligible", "jumble", "inarticulate", and he is far from the first person to offer a characterization along these lines. A year ago, Geoff Pullum used words and phrases like "aphasia", "no structure at all", "barely a coherent thought" ("Trump's aphasia", 8/5/2015).

I've argued in response that these observers have been misled by "the apparent incoherence of much transcribed extemporized speech, even when the same material is completely comprehensible and even eloquent in audio or audio-visual form" ("Trump's eloquence", 8/5/2015).

In order to underline this point, I thought I'd exhibit a randomly-selected passage from another unscripted spoken-word performance by someone who doesn't speak in paragraphs. That description generally applies to the rhetorical tradition of stand-up comedy, I think.

So I picked a comedian and a performance at random — the start of a performances at Beacon Theater in 2013 (?) by the comedian Louis CK. And you'll see that the clip has many of the same characteristics as Trump's spontaneous performances — repetitions, incomplete phrases, digressions, …

In fact Louis CK is much less fluent than Trump in certain ways, such as a much higher rate of UM/UH usage and of fluent-self-corrections. But otherwise, the similarity is remarkable.

In the context of the current political season, I've started taking a look at rhetorical styles, including the aspects of rhythm, pitch, and voice quality for which linguists generally use the cover term "prosody". Our enormously over-long list of topic categories didn't include "prosody", so I've added it — and in the process of labeling relevant posts, I made a (probably still incomplete) list of linked titles and dates, which is reproduced below.