Weekly FUD & Labels Alert - 9/11/26
Prepared by Rick Fiorito • September 11, 2026
On this 25th anniversary of 9/11, let us remember this tragedy what we lost and those how stepped up and showed the best of who we are together!
How will labeling and FUD affect midterm election results?
Summary of the Week
Across all the daily reports this week, there were 31 unique flagged statements using Labels and FUD (See list at end of Blog). Split by party: 16 Republican, 14 Democratic, 1 Independent. 20 statements out of the week's coverage, concentrated around the GOP's first-ever “Midterm Convention” in Dallas and same-day Democratic counterstatements. Republican officials leaned on apocalyptic framing (Trump: the country will “spend the next decade just getting back on its feet” and go “through hell”; Cox: “complete failure of our democratic republic”) and ideological branding (“communism,” “radical,” “weird”). Democratic officials leaned on unsupported character claims (Pritzker on Trump's cognitive state) and sweeping corruption superlatives (Crow, Garcia: “most corrupt administration in history”). Both sides also reached for “rigged election” framing over the broader window. The question below is what the research literature says happens when politicians use these tools — not whether any individual claim is true.
Across the political-science literature, fear-and-doubt rhetoric of the kind flagged in this week's reports is best understood as a high-variance, low-persuasion, base-consolidating tool rather than a reliable vote-getter. It rarely converts opposing or undecided voters, it carries a real chance of backfiring on the speaker, and its clearest documented effect is on the speaker's own supporters — raising their engagement and, in the case of election-legitimacy claims, measurably eroding their trust in the system.
Both parties are using structurally different flavors of it this cycle (ideological/civilizational framing from Republicans, specific-actor-corruption framing from Democrats), which the research literature treats as a long-running asymmetry, not a new development.
The audience gap matters most: this messaging is built for — and mainly reaches — people who have already picked a side. The genuinely undecided are more skeptical, more substance-seeking group, and the research below suggests FUD is a poor tool for reaching them.
What the Research Shows
1. Attack rhetoric is close to a coin flip, not a winning formula
The most-cited academic assessment of negative campaigning — Lau, Sigelman & Rovner's meta-analysis of over 100 studies in the Journal of Politics (2007) — concluded there is “simply no evidence that attack [messaging] is a smart move.” Of the studies that directly tested effects on vote choice, only a handful found a clear benefit for the attacker, and roughly as many found a backlash effect that hurt the attacker more than the target. A 2019 review in French Politics reached the same conclusion and added a sharper point: the backlash effect on the sponsor's own image is, on average, slightly stronger and more consistent than the damage done to the target. Attacking is a gamble with the attacker's own numbers, not a guaranteed win against the opponent's.
2. Fear appeals do work — but the evidence for politics specifically is thin
A large meta-analysis (Tannenbaum et al., Psychological Bulletin, 2015, 127 studies) found fear appeals reliably move attitudes and behavior (average effect size d = 0.29) and found no clear conditions under which they backfire. But the finding comes almost entirely from health-communication research — smoking, disease prevention, safety — not political persuasion; only about a sixth of the underlying samples were non-health. Critically, the appeals that work best pair the fear with a specific, actionable response (“here's exactly what to do about it”). Generic political catastrophizing (“we'll go through hell”) without a concrete, credible mechanism is a much weaker match to the conditions under which fear appeals are shown to succeed.
3. It doesn't suppress turnout — but it mostly rallies people who were voting anyway
An older hypothesis held that negative advertising depresses turnout by disgusting voters into staying home. That has not held up: the most recent meta-analyses find no consistent demobilizing effect. What has replaced it in the literature is “negative partisanship” (Abramowitz & Webster, Political Psychology, 2018) — dislike and fear of the opposing party is now a stronger driver of turnout and straight-ticket voting than positive enthusiasm for one's own side. Fear-and-doubt messaging plays directly into that dynamic: it is well-suited to energizing an already-committed base, but the same research finds shrinking evidence that it moves the genuinely persuadable middle, because that group responds less to threat framing from either side.
4. It spreads fast online — that's an attention effect, not a persuasion effect
Brady et al. (PNAS, 2017) found that each additional moral-emotional word in a political message increases its sharing by roughly 20%, and that attacking the opposing side drives far more engagement than praising one's own. This maps closely onto the sweeping, unfalsifiable superlatives in this week's reports (“most corrupt,” “most dangerous”) — they are well-built for virality. But the underlying research measures reach and engagement within already-sympathetic networks, not documented changes in the views of people outside them. Visibility and persuasion are not the same thing, and the literature is careful to keep them separate.
5. It measurably erodes trust — mainly among the speaker's own side
A 2021 PNAS panel study exposed Trump supporters to his election-legitimacy rhetoric and found no broad increase in support for political violence or a general collapse in belief in democracy. What it did find was a specific, targeted effect: increased belief that “the election is rigged” and reduced trust and confidence in elections, concentrated among the speaker's own supporters. Applied to this week's “rigged election” and “constitutional crisis” framing from both parties, the research suggests the most likely real-world effect isn't persuading the other side — it's a slow, one-sided erosion of the speaker's own voters' confidence in the electoral system itself, which is a cost with no obvious offsetting electoral benefit.
6. Unsupported factual and character claims are the most exposed
Wood & Porter's widely replicated study (“The Elusive Backfire Effect,” 2019, 8,100 subjects across 36 issues) found that factual corrections mostly work — voters generally update toward the truth even when it cuts against their side, and the once-popular idea that corrections entrench false beliefs (the “backfire effect”) turns out to be rare. That cuts a specific way against the more checkable claims in this week's set — a medical diagnosis asserted with no evidence, a specific policy “architect” claim, a specific accusation about a candidate's intentions toward children. These are the statements most vulnerable to a credible correction actually landing, compared with vaguer, unfalsifiable rhetoric that has nothing concrete to correct.
7. The two parties are running structurally different playbooks
Grossmann & Hopkins' Asymmetric Politics (2016) documents a long-running, non-partisan-in-origin pattern: Republican political rhetoric organizes more around ideological and civilizational stakes (“communism,” existential-threat framing), while Democratic rhetoric organizes more around specific-actor wrongdoing and group harm (“corrupt,” “dangerous,” named individual accusations). This week's reports track that pattern almost exactly — GOP convention speakers reached for “communism” and civilizational framing, Democratic officials reached for corruption superlatives and character claims about specific individuals. The research treats this as a structural feature of how the two coalitions are built, not evidence that one side is uniquely prone to exaggeration.
The Undecided Voter: A Different Audience
Everything above describes what FUD does for a party's own coalition. Undecided and persuadable voters are a separate question, and the research treats them as a genuinely different audience — smaller than either party's messaging usually assumes, and less receptive to this style of rhetoric specifically.
It isn't built to move them — and mostly doesn't
The most direct evidence comes from Kalla & Broockman's meta-analysis of 49 field experiments on general-election campaign contact (American Political Science Review, 2018): conventional messaging — ads, mail, calls, the vehicles FUD travels through — has an average persuasive effect on vote choice indistinguishable from zero. That doesn't mean nothing works; it means the standard toolkit, attack rhetoric very much included, mostly reinforces existing preferences rather than changing minds. Separately, Nai & Maier's research on individual differences (Political Communication-adjacent, 2021) found that tolerance for negative and uncivil campaigning tracks specific personality traits — high conscientiousness, high agreeableness, low tolerance for conflict — rather than partisanship as such. Because persuadable voters skew more moderate and conflict-averse than committed partisans on average, they sit closer to the profile the research shows is least receptive to attacks, not most.
The evidence points toward substance, not more heat
Where research does find durable persuasion, it looks nothing like FUD. Broockman & Kalla's “deep canvassing” studies (Science, 2016; extended to immigration attitudes in later work) found that structured, non-judgmental, two-way conversations — listening, personal stories, no attack framing — produced attitude shifts larger than 14 years of background social change on one issue, with effects still measurable three-plus months later. The mechanism researchers point to is straightforward: attack-style messaging triggers identity-protective defensiveness, while substantive, respectful engagement doesn't. Consistent with this, Franklin & Marshall's polling analysis found still-deciding voters report more negative views of both candidates/parties than committed voters do — a group already fatigued by conflict, not one a sharper attack is likely to win over. Diana Mutz's research on uncivil political media (In-Your-Face Politics, Princeton, 2015) adds a related, broader point: incivility grabs attention through a physiological alarm response, but it also measurably lowers trust in politicians and government generally — a cost that falls on persuadable, lower-trust voters disproportionately, since they have the least partisan loyalty to cushion it.
How many people that actually describes
Depending on how tightly “undecided” is defined, current estimates range from a headline ~5% to a broader ~12–15% of the electorate, concentrated heavily among self-described independents and moderates rather than spread evenly across the population. Here is what that looks like:
Headline “undecided/other” on the national generic congressional ballot; Figure - ~5%
Emerson College's July 2026 national poll put Democrats at 53%, Republicans at 42%, undecided/other at 5%. Most recent trackers show a similarly small undecided slice at the topline level.
Americans self-identifying as political independents; Figure -~45% (record high)
Gallup put independent self-identification at a new high of 45% in 2026. Most “independents” still lean consistently toward one party, so this overstates the truly persuadable pool.
Independents still undecided on their House vote, weeks out from the election; Figure -~28%
Within that record-large independent bloc, roughly a quarter to a third remain genuinely undecided this close to the election — and pure independents view both parties unfavorably (24% favorable to Democrats, 15% to Republicans).
“True” persuadable share of the electorate, broadly defined; Figure -~12–15%
Franklin & Marshall's analysis found only ~3% say “don't know,” but roughly another 12% have a preference they say could still change — heavily concentrated among self-described moderates (~25% still fluid, vs. ~2% of strong partisans).
Two things are worth holding onto:
Midterms aren't one national election — they're 435 House races plus Senate and governor contests — so the “undecided” share in a specific competitive district or state race is often considerably higher than the generic-ballot topline nationally.
The independent/persuadable bloc is unusually large and unusually unhappy with both parties right now (record-high independent self-identification, both parties underwater with pure independents), which raises the stakes on the finding above: if that group is genuinely more turned off by FUD and more responsive to substantive engagement, it is also the largest such bloc in recent memory.
Applying This to the Week's Reports
The table below maps the specific tactics flagged in the daily reports to the research findings above.
Catastrophic predictions (“spend the next decade... hell,” “complete failure of our democratic republic,” “city will die”)
Lau, Sigelman & Rovner (2007); Haselmayer, French Politics review (2019)
No consistent electoral payoff. Roughly as many studies find backlash against the speaker as find benefit; effect on the sponsor's own image is slightly worse, on average, than the effect on the target. Ideological branding (“communism,” “radical,” “weird,” “hate-America extremists”)
Grossmann & Hopkins, Asymmetric Politics (2016)
Consistent with a long-documented structural pattern: GOP messaging skews toward ideological/civilizational stakes, Democratic messaging toward specific-actor wrongdoing. Neither is new or unique to this cycle. Unsupported personal/character claims (“wants to transition your kids,” “suffering from dementia,” “architect of… war”)
Wood & Porter, “The Elusive Backfire Effect” (2019)
These are the most exposed claims in the set. Research finds voters generally do accept factual corrections — backfire (a correction hardening the false belief) is rare — so specific, checkable accusations are the ones most likely to erode the speaker's credibility if debunked. Sweeping, unfalsifiable superlatives (“most corrupt… in history,” “most dangerous… in my lifetime”)
Brady et al., PNAS (2017); moral-outrage/virality literature
This is the flavor most rewarded by social platforms — moral-emotional language increases sharing — but the effect is on reach and attention within sympathetic audiences, not documented persuasion of the other side. “Rigged election” / election-legitimacy framing (both parties)
Clayton, Davis, Nyhan, Porter, Ryan & Wood, PNAS (2021, on election-legitimacy rhetoric)
What the evidence suggests - The one tactic with a documented, if narrow, effect: it doesn't move persuadable voters much, but it measurably erodes trust and confidence in elections among the speaker's own supporters — a real but one-sided cost.
This Week’s Citations of Labels and FUD
Trump, President: “The city will die” (Chicago); Chicago Sun-Times, Mediaite
Mike Johnson, Speaker: “Political palooza” (Court-packing amendment); The Hill
Mike Johnson, Speaker: “Communist playbook” (press release); Speaker.gov
JD Vance, Vice President: “Rig the game” (WH press briefing); Twitchy
Denny Hoskins, MO Sec. of State: “War to preserve our constitutional republic” (redistricting); Missouri Independent
Catherine Hanaway, MO AG: “Full-fledged constitutional crisis” (redistricting); Missouri Independent
Trump, President: Fed rate ultimatum; Mediaite, CNBC
Ted Cruz, Senator (TX): “Marxist/Islamist” Senate Democrats; LifeZette
Spencer Cox, Governor (UT): “Complete failure of our democratic republic”; ABC News
John Kennedy, Senator (LA): “Gutting the middle class like a fish” (inflation); NBC News
Trump, President: “Hell” / “tragedy” (GOP Midterm Convention); Washington Times
Ken Paxton, TX AG/Senate candidate: “Lose Texas, lose America” / “transition your kids”; KWTX
Steve Scalise, House Majority Leader: “Impeach President Trump and his cabinet”; Just the News
Mike Johnson, Speaker: “Competition against communism”: Fox News
Tim Scott, Senator (SC): “Most dangerous, radical… weird”; Washington Times
Tom Emmer, House Majority Whip: “Common sense” vs. “communism”; American Greatness
JB Pritzker, Governor (IL): “Murdered” / “cover it up” (Border Patrol shooting); NBC Chicago
Hakeem Jeffries, House Minority Leader: “Rig the midterm elections” (voter-ID bill); Jeffries.house.gov
Kathy Hochul, Governor (NY); “Betraying every family… 9/11” (counterterrorism funds); Governor's Office
Zohran Mamdani, Mayor (NYC): “Terrorizing people” (ICE, The View); Yahoo/ABC
Jamie Raskin, Rep. (MD): “MAGA built” (Court-packing floor remarks); The Hill
Steven Raga, NY State Assembly member: “Campaign of fear and intimidation” (ICE, Woodside); politicsny.com
Chuck Schumer, Senate Minority Leader: “Destroy the precious integrity” (SCOTUS mail-voting); Senate Democrats
Bernie Sanders, Senator (I-VT): Unsourced “rogue AI” quote (AI ban bill); X / Sen. Sanders
Gavin Newsom, Governor (CA): “Going to try to steal this election”; Fortune
Jason Crow, Rep. (CO): “Most corrupt administration in history”; CBS News
Jamie Raskin, Rep. (MD): “Undoubtedly” / “he believes he is a king” (impeachment); NBC News
Chellie Pingree, Rep. (ME): “Unhinged” / fitness for office (re: Trump AI posts); X / Rep. Pingree
JB Pritzker, Governor (IL): “Dementia” / Vance “architect of… war”; Daily Kos
James Talarico, TX State Rep./Senate candidate: “Toasting” / “corrupt tariffs” / “forever war”; PBS NewsHour (convention-day coverage; no direct URL in source report)
Robert Garcia, Rep. (CA): “Most corrupt administration in… United States”; PBS NewsHour (convention-day coverage; no direct URL in source reports)
Report Caveats
This is a synthesis of the peer-reviewed and widely replicated literature on negative campaigning, fear appeals, and elite rhetoric, not a study of this specific week's statements. Three limitations are worth keeping in mind. First, most of the negative-campaigning literature comes from paid advertising and lab experiments, which is a different delivery mechanism than live convention speeches and cable-news soundbites amplified through partisan and social media — the underlying incentives (attention, base energy) are the same, but effect sizes may not transfer exactly. Second, effects are heterogeneous: what demobilizes or backfires with one audience (e.g., persuadable independents) can simultaneously mobilize another (the base), so “does it work” depends heavily on which audience a campaign is actually trying to move. Third, 2026 is a highly polarized, high-information environment relative to when much of this literature was produced, which plausibly strengthens the negative-partisanship dynamics described above and weakens the persuasion effects, though that specific interaction hasn't been directly tested.