Thread by @KobiHackenburg

🚨 New paper! 🚨 w/ @william__brady & @manos_tsakiris:

Today in @PNASNexus we map the moral language of 39 U.S. presidential candidates to show how they're connected / differentiated by their use of moral rhetoric.
ow.ly/8Zwu50OO5QA

A [visual] thread on findings:

Main takeaways:

1) @TheDemocrats & @GOP
candidates use sharply divergent moral vocabularies

2) Candidates can separate themselves from the rhetorical norms of their party by using unique moral language, but rarely do

Taken together with prior research (@RobbWiller), this work empirically shows that the moral rhetoric employed by recent candidates is probably un-persuasive across party lines & likely exacerbates polarization, particularly in online social networks.

Below, see the moral words we measured & how we built the networks.

By combining network analysis & #NLP to position candidates in moral-rhetorical space, we illustrated not only the rhetorical norms of each party, but also how individual candidates *deviate* from those norms.

We find that partisan affiliation communities can be re-constructed purely based on the moral words used by candidates in their rhetoric.

In other words, you can predict the partisanship of *every* 2016 & 2020 candidate based on their moral language.

We find two dynamics driving the partisan divergence revealed in the networks.

First, @TheDemocrats and @GOP candidates discuss different moral values. @TheDemocrats emphasize care and fairness, while @GOP emphasize loyalty, authority, and sanctity.

Second, within each party, candidates discuss "popular" moral values in the same ways.

E.g., just as D’s talk about care *more*, they also do so in highly similar ways, selecting the same "care" words and using them at similar frequencies.

At the candidate level, we show that Trump separated himself from his competition, partly by using lots of negatively-valanced fairness language.

A small number of candidates deviated in other interesting ways!

For example, candidates like @JoeBiden & @PeteButtigieg were able to use “conservative” moral values (sanctity, loyalty) without being pushed to the periphery of moral-rhetorical space or towards @GOP.

They generated unique "Democratic" loyalty/sanctity vocabularies.

Candidates like @TulsiGabbard and @AndrewYang also used "conservative" moral foundations -- but deviated *towards* the @GOP. They used moral language most similar to 2016 @GOP candidates:

We hope that this work illustrates how patterns of moral expression delineate the partisan factions that define U.S. politics.

We also hope it reveals how & when those boundaries are crossed.

Networks are tough to interpret -- Q's welcome :)

As a last note, I want to thank my coauthors @william__brady & @manos_tsakiris for patiently supporting this project over the past year. Couldn't ask for better mentors!

I also want to thank all of our reviewers, and our editor @jayvanbavel.

This work might be of interest:
@Amit_Goldenb
@RobbWiller
@chris_bail
@mjbsp
@freddy_hopp
@LindaBos80
@JonHaidt
@mollycrockett