Take 417 American partisans — 263 Democrats, 154 Republicans — and ask each of them four things: how passionately they feel about their party, how much they long for its “glory days,” how fused their identity is with the cause, and how willing they’d be to support hostile action against the other side. That’s the design behind Collective Nostalgia Widens the Gap Between Obsessive and Harmonious Passion in Support for Political Violence, published Oct. 1, 2026 in the Journal of Applied Social Psychology by Jocelyn J. Bélanger of Carnegie Mellon University in Qatar, with Daniel Snook, Danaiym Talantbekova, and Jaïs Adam-Troian, and covered Oct. 17, 2026 by PsyPost. The same wistfulness for the party’s past predicted more support for violence against the rival party — but only among partisans whose passion was obsessive. Among partisans whose passion was harmonious, that identical nostalgia predicted less. Baruch Spinoza never heard of a dualistic model of passion. But in 1677, in the part of his Ethics literally titled “Of Human Bondage,” he drew the same line: a feeling that owns you and a feeling you own are not different amounts of the same thing. They’re different in kind, and one of them is dangerous specifically because it’s imposed on you.

The Quick Verdict

Question	What we found
What was studied?	Whether collective nostalgia for a political party’s past predicts support for political violence against the rival party, and whether that depends on the type of passion a partisan feels
Who ran it?	Jocelyn J. Bélanger (Carnegie Mellon University, Qatar), Daniel Snook, Danaiym Talantbekova, Jaïs Adam-Troian
Where and when?	Journal of Applied Social Psychology, Oct. 1, 2026; covered by PsyPost
Sample	417 U.S. partisans (263 Democrats, 154 Republicans), surveyed online in March 2022
Core finding	Obsessive passion for a party predicted greater support for political violence against the rival party, in both subsamples. Harmonious passion did not.
Moderation finding	Collective nostalgia amplified the obsessive-passion-to-violence link, but reversed direction for harmonious passion — more nostalgia, less support for violence
Key quote	Bélanger: “How you care about a cause matters more than how much you care”
What Spinoza already argued	In Ethics Part IV, “Of Human Bondage” (1677), passions become dangerous when they’re passive — produced by an external cause the self doesn’t govern — not when they’re merely strong

The Quick Version

A 2026 study of 417 U.S. partisans found that longing for the party’s past pushed obsessively passionate partisans toward supporting violence against the other side, while the exact same longing pushed harmoniously passionate partisans away from it. Spinoza described this split 350 years earlier: it isn’t the intensity of a feeling that makes it dangerous, it’s whether that feeling is forced on you from outside or integrated with the rest of who you are.

What the Study Actually Found

Most research on political violence asks what predicts it directly — grievance, dehumanization, threat perception. Bélanger’s team asked something a step upstream: does it matter how someone is passionate about their side, independent of how strongly?

They borrowed a framework that’s been running in sport and work psychology for two decades, Robert Vallerand’s Dualistic Model of Passion. Harmonious passion is a strong attachment to an activity or cause that stays under your control — you choose when to engage it, and it coexists with the rest of your life. Obsessive passion is the same intensity of attachment, minus the control. People describe feeling compelled, unable to not think about it, identity-fused to the point that the cause has started running them rather than the other way around.

Applied to partisanship, the team measured both passion types alongside collective nostalgia (longing for the party’s past glory) and ideological commitment, then asked everyone how much they’d support hostile, even violent, action against the rival party. The headline: obsessive passion tracked higher support for political violence in both the Democrat and Republican subsamples. Harmonious passion didn’t carry that risk at all — strength of feeling wasn’t the variable doing the damage.

Then nostalgia entered as a moderator, and this is the part that should make you sit up. Among highly obsessive partisans, more nostalgia for the party’s past went with more support for violence against the rival side. Among highly harmonious partisans, more nostalgia went with less. Same memory, same sense of “it used to be better,” opposite behavioral signature — depending entirely on what kind of passion was holding it. As Bélanger put it to PsyPost: “How you care about a cause matters more than how much you care.”

It’s worth being honest about scale here too. Average support for political violence across the whole sample sat low, around 1.7 to 1.8 on a 7-point scale — most people, obsessive or not, weren’t anywhere near endorsing violence. And nostalgia’s moderating effect, while statistically reliable in both subsamples, explained something on the order of 1 to 2% of additional variance. Small. Real, replicated across party lines, and theoretically pointed — but small.

What Did Spinoza Mean by Human Bondage?

Human bondage (servitus humana), the subject of Part IV of Spinoza’s Ethics (1677), is his term for being ruled by passions whose cause lies outside you rather than in your own nature — “passions” in his technical sense, meaning affects for which you are only the partial, not the adequate, cause. Spinoza opens the part defining this bondage as “infirmity in moderating and checking the emotions,” the condition of someone who “is not his own master, but lies at the mercy of fortune.”

That technical distinction — adequate cause versus partial cause — isn’t academic hair-splitting. Spinoza laid the groundwork two parts earlier: we act when something happens through our nature alone, clearly and fully explicable by who we are. We’re passive — in bondage to a passion — when something happens in us that can only be explained by an external cause working on us, something we didn’t originate and don’t control. An emotion caused that way, he says, is confused: you feel it, but you can’t fully account for it from your own nature, because it isn’t coming from your own nature. It’s coming from outside and running through you.

Spinoza’s bleaker proposition in Part IV is that this isn’t a minor risk. “The force whereby a man persists in existing is limited, and is infinitely surpassed by the power of external causes” — meaning nobody talks themselves out of a passion through willpower or correct belief alone. An emotion, he argues, can only be checked by a stronger, contrary emotion, not by truth. If nothing inside you is strong enough to counter the externally-driven passion, it runs unchecked. That’s bondage. Freedom, in Spinoza’s system, isn’t the absence of strong feeling. It’s feeling that’s been brought inside the self — understood, integrated, weighed against your other commitments — rather than imposed from without and left to run alone.

Why Did the Same Nostalgia Predict Opposite Outcomes?

1. Obsessive passion fuses identity to the cause, which removes any counterweight. When the party is who you are, nostalgia for its past glory isn’t one input among several — it’s a direct hit on the self. There’s no competing commitment left to check it, which is exactly the structural failure Spinoza describes: nothing strong enough inside the person to counter an externally-amplified feeling.
2. Harmonious passion keeps the cause one commitment among several, so nostalgia gets processed rather than detonated. A partisan who also has a job, a family, hobbies, and other loyalties runs the party’s remembered glory through all of that before it becomes a verdict about what to do. The feeling arrives, but it isn’t the only feeling in the room.
3. Nostalgia works like an amplifier, not a signal. It doesn’t introduce a new belief. It intensifies whatever structure the passion already has. Fed into an obsessive, externally-driven attachment, it amplifies the bondage. Fed into a harmonious, self-integrated one, it amplifies the stability — which is why it moved support for violence in opposite directions from the identical memory.
4. This is Vallerand’s model doing, with a questionnaire, what Spinoza’s distinction between active and passive affects did with a geometric proof. Neither obsessive passion nor human bondage are about the topic someone cares about. Both are about whether the caring is still answerable to the rest of the person, or has stopped taking instructions from anywhere but itself.

Why the Study and the Philosophy Rhyme

The honest version of this comparison isn’t “a 17th-century rationalist predicted a 2026 partisanship survey.” It’s narrower, and it holds up better for being narrower: Spinoza’s specific claim was that an affect’s danger comes from its source — external and uncontrolled, versus internal and reasoned — not from its size. Bélanger’s team measured a modern instance of exactly that structural split, in a domain (partisan identity) Spinoza never addressed directly, and found the same signature: the version of the feeling with an external, uncontrolled source predicted violence; the version integrated with the rest of a person’s life didn’t, and nostalgia made each version more fully itself.

It connects to the research on meaningless-world belief and extremism covered here before — a study co-authored by the same Daniel Snook, as it happens. That work found the belief that things matter functioned as a brake on violence, regardless of ideology. This study locates a second, related brake: not whether the world matters, but whether your attachment to your side of it is still something you hold, rather than something holding you.

It also sits close to Girard’s account of how imitated desire escalates into rivalry and scapegoating. Girard’s mimetic rivalry needs a self porous enough to let a borrowed desire take over and crowd out its own judgment — which is a social, imitative route to roughly the state Spinoza calls bondage. And it rhymes with Tocqueville’s finding that loving democracy in the abstract barely predicts resistance to antidemocratic action: abstract commitment didn’t distinguish the groups in Claassen’s data either. What held the line there was specific, concrete attachment to institutions — a harmonious kind of commitment, in this study’s vocabulary, rather than a fused, obsessive one.

Spinoza’s own account of the difference between whole-person joy and the grasping, localized kind maps onto this almost exactly. Hilaritas is joy that moves the whole person and stays in proportion. Titillatio is joy that seizes one part and can grow to dominate everything else. Obsessive partisan passion, running on amplified nostalgia, looks like titillatio wearing a political uniform — a part of the self that’s stopped answering to the whole of it.

The Gap Spinoza Didn’t See Coming

Here’s where the comparison gets more interesting than a clean match.

Spinoza’s framework treats bondage and freedom as a property of an individual mind working out its relationship to its own affects — a solitary, almost geometric problem. Bélanger’s data is collective from the start: it’s nostalgia for a group’s shared past, measured across hundreds of partisans sorted by party, not one person’s private struggle with a feeling. Spinoza has relatively little to say about what happens when the “external cause” producing bondage isn’t a private appetite but a shared, socially reinforced memory that an entire party apparatus has an interest in keeping alive. The passion in this study isn’t just imposed on an individual by circumstance. It’s actively cultivated, in both parties, by messaging that trades on exactly the nostalgic glory-days narrative the researchers measured.

That’s a real gap. But it sharpens rather than breaks the parallel: if an externally-sourced passion is this dangerous when it arrives through ordinary individual experience, a version of it manufactured and distributed at scale, aimed at people already fused to the identity, is the same mechanism with a louder external cause behind it. Spinoza’s solitary thinker becomes, in 2026, a person whose “external cause” has a communications strategy.

What This Study Doesn’t Show

Be precise about the size of this. The framing invites more confidence than the data earns.

Attitudes, Not Acts

Every outcome here is self-reported willingness to support hostile or violent action, not a measured violent act. That’s standard in this research area and ethically necessary, but it’s a real distance between what was asked and what a headline about “predicting violence” implies.

Cross-Sectional and Four Years Old at Publication

The data was collected online in March 2022 and published in October 2026. It captures one moment, not a trend, and that moment predates a great deal of subsequent political change in both U.S. parties. Correlation at a single time point also can’t establish that obsessive passion or nostalgia caused anything; a third factor, like generalized grievance, could be driving both.

A Small Moderation Effect

Collective nostalgia’s amplifying and dampening effects, while statistically significant in both subsamples, accounted for roughly 1 to 2% of additional variance in violence support. The directional split is the finding worth taking seriously. The size of its practical impact on any one person’s behavior is modest.

Two Parties, One Country

The sample is U.S. Democrats and Republicans exclusively. Whether the same obsessive-versus-harmonious split governs nostalgia’s effect on political violence support in multiparty systems, or in countries without America’s specific two-party structure, is untested here.

What Actually Follows From This

1. “They’re passionate about politics” tells you almost nothing useful on its own. This study is a specific, measured reason to ask the next question — passionate in a way that’s still answerable to the rest of their life, or passionate in a way that’s stopped taking input from anywhere else.
2. Nostalgia isn’t neutral, and it isn’t uniformly dangerous either. The same wistful story about a party’s past can function as a stabilizer or an accelerant, and the deciding variable isn’t the story. It’s the structure of the attachment it’s landing in.
3. Spinoza’s older question still does useful work here: is this feeling something you hold, or something holding you? That’s not a question with a quick fix, and philosophy was never going to disarm a nostalgic political narrative on its own. But it’s a sharper diagnostic than “how strongly do you feel about this” — which is the question most of us actually ask, and the one this study’s data says is pointed at the wrong target.

Spinoza never ran a regression and never heard the word “partisan” used the way we use it now. He had a geometric proof about the difference between a mind that generates its own affects and one that’s run by them, and a historical moment — Dutch religious and political factionalism, an excommunication, a stabbing attempt on his own coat — that gave him every reason to think hard about what makes political feeling turn violent. Three hundred and forty-nine years later, a team surveying 417 American partisans found the same fault line running through a Democrat’s and a Republican’s answers to a questionnaire about their party’s glory days: not how much they loved it, but whether that love still belonged to them.

---

This post discusses a 2026 study published in the Journal of Applied Social Psychology (Bélanger, Snook, Talantbekova & Adam-Troian, DOI: 10.1111/jasp.70100), reported by PsyPost, and Spinoza’s Ethics (1677), Parts III and IV. The study is correlational, based on a single cross-sectional U.S. sample surveyed in March 2022, and measures self-reported attitudes toward political violence, not actual violent behavior. If you or someone you know is expressing intent to harm themselves or others, that’s a situation for a crisis line or mental health professional, not a philosophy blog.