Perhaps no modern president has highlighted the often irregular and adaptable nature of politicians as much as Donald Trump.

With his dominance of the Republican Party, its politicians and voters have rearranged themselves around Trump — the career of the late Senator Lindsey Graham offered a powerful example of this. How did we get here?

The political scientist Seth Masket, author of “The Elephants in the Room: How Trump Voters Seized the Party From Republican Leaders,” reflected on the twists and turns in a written conversation with John Guida, an editor in Times Opinion. It has been edited for length and clarity.

John Guida: In some ways, Lindsey Graham was not the most natural candidate to become a Trump whisperer (an old-school hawk who once called Trump a “kook”; an institutionalist; a friend to Republicans and Democrats). In other ways, though, perhaps we should not be surprised — Graham was remarkably adaptable, which earned him both praise and harsh criticism. How is Graham a window into this moment, and the Trump G.O.P.?

Seth Masket: Graham is a fascinating case study, in part because he was one of the most incisive and cleareyed critics of Trump in 2015 and 2016, and within a year was refashioning himself as one of Trump’s closest friends and allies.

He remained a stalwart hawk on the use of the U.S. military abroad — even on Ukraine, when Trump clearly had different views — but he compromised a lot to get to that point. Being adaptable to shifting political environments is a hallmark of skilled politicians, and we saw that from a lot of Republicans after 2016 (see Ted Cruz). But few did so quite so visibly as Graham.

Guida: At an event in Greenville, S.C., late in Donald Trump’s first term, at a time when people were asking how Graham could’ve evolved so quickly from Trump skeptic to supporter, he declared: “What happened to me? Not a damn thing.” He appeared to embrace one principle above all others: As he later said, by becoming a Trump ally, “I’m still in the game” — the game presumably being one of power and influence.

What do Graham and Trump reveal about the true nature of politics? For politicians, how should we think about the tension between being adaptable and holding fast to your principles?

Masket: Maybe a useful counterexample would be Mitt Romney. Both he and Graham had very harsh things to say about Trump in 2016 and tried to organize against him, but they took significantly different turns from 2017 on. Romney was in the Senate for part of Trump’s first term and often spoke out against Trump’s most anti-democratic tendencies. He was the only Republican to vote to convict Trump in both of his impeachment trials.

It’s plausible Romney was more principled, or perhaps, given his age, he figured he didn’t have all that many years in the Senate left. Graham, meanwhile, eight years younger, wanted to remain in the Senate and remain influential in national politics for many more years and was willing to do what it took to achieve that.

And in that sense, Graham may have been correct in saying that “not a damn thing” changed. He was always willing to adapt to circumstances to remain a force in government. And on the thing that probably mattered most to him — the use of military force abroad — he arguably retained his principles even as they became less popular. But he was willing to sacrifice many things, including a longstanding reputation as a congenial and witty senator with deep friendships in both parties.

Guida: Could Graham’s death and Mitch McConnell’s extended absence from the Senate be a turning point for the Republican Party? Especially in terms of the Senate as a counterweight to the president.

Masket: It’s not as if Graham’s and McConnell’s successors are likely to be anti-Trump, but there is some growing pushback within the party — and especially among some Republican senators — about Trump’s prosecution of the war with Iran, some of his nominees, the “weaponization” fund for people prosecuted defending Trump, etc. One can see inklings of a bit of institutional pushback, with the Senate increasingly seeking to use its institutional powers as a (modest) check on the White House. That may continue, even without Graham (and perhaps McConnell) on the Hill.

Guida: What about in the context of the party’s posture toward the world? Both those men believed America could be, as the shorthand has it, a “force for good” in the world, however misguided that might be (and again, plenty of people have pointed out that Graham especially pushed the U.S. toward foreign catastrophes in the Middle East). MAGA has a very different global posture.

Masket: Yes, for better or worse, McConnell’s and Graham’s approach to foreign policy is generally consistent with what we thought of as pre-Trump conservatism — a willingness to use military power to advance American values and interests and to sometimes seek to create new democracies or steer other nations in that direction. Trump’s “America first” approach is very much opposed to that. McConnell and Graham have been a check on this, and Trump occasionally listened to Graham.

Guida: Looking ahead, do you think “America first” will be the dominant force within the G.O.P.?

Masket: It depends how far ahead we’re looking. In the near term, 2028 presidential contenders like JD Vance will proceed with one overriding goal: securing Donald Trump’s endorsement. Vance and other hopefuls will support “America first” priorities to help them get it. Once Trump is no longer part of the political scene, however, “America first” won’t disappear as a force within the G.O.P., but it will have a lot less power. Different versions of conservatism will compete for control within primary elections and within the next G.O.P. presidential campaign. Vance might be deft enough to manage those fights, but it will be a highly uncertain environment.

Guida: Let’s look at how we got to Trump 2.0 in the first place. Political parties generally do not renominate losing candidates. In your research on the 2024 G.O.P. primary, you took a deep dive into how Trump became the nominee. What did you discover from speaking to local party chairs?

Masket: I wanted to understand how they were thinking about the 2024 nomination back in early 2023, and how their views evolved over that year. One thing that surprised me was that, in early 2023, most chairs were not committed to a candidate, and among those who were, Ron DeSantis was more popular than Trump. At least early on, a lot of local Republican leaders were deeply skeptical of another Trump candidacy. And these were people who mostly liked his first term.

They thought he carried too much baggage (scandals, criminal investigations, a belligerent personal style, etc.). They tried to steer Republican primary voters toward the alternatives.

Guida: Because of baggage, policy or both, or something else?

Masket: I spoke to some local party chairs who were fond of Trump’s stances on the economy, taxes, abortion. But they were clearly uncomfortable with his more populist “America first” side, especially with regards to foreign policy, and wanted to serve as a check on that influence on policy.

But other party chairs who were skeptical of Trump took that view largely on “electability” grounds. They were generally fine with Trump’s policies but thought they’d have a better shot at winning with a different presidential candidate. Many of them noted to me that DeSantis had governed Florida very conservatively but managed to turn his narrow 2018 gubernatorial victory into a re-election blowout in 2022. Quite a few of the people I spoke to talked up Nikki Haley on electability grounds as well, and could point to polling that backed that up.

Guida: You describe an intense connection between G.O.P. base voters and Trump. So in one sense, it’s easy to understand how he has maintained control of the party. But it is unusual, right, for presidents to get as involved in the party’s primary contests, as Trump has done? Did you get a sense of why that connection is so strong?

Masket: I was part of a research project with Mike Cowburn and Rachel Blum a few years ago in which we tried to measure the influence of a Trump endorsement in the 2022 Republican primaries. We found that when he endorsed a candidate, it increased that candidate’s primary vote margin by about 13 points. That’s a huge effect, as we saw this year with the primary defeats of Representative Tom Massie, Senators John Cornyn and Bill Cassidy, and others.

This sort of presidential involvement in party primaries is highly unusual. It’s certainly plausible that, say, Barack Obama could have had a similar influence on Democratic primaries, but he didn’t try to do that — most presidents would rather not pick fights within their own party. Franklin Roosevelt attempted this sort of thing in the 1938 midterm elections, seeking to punish those Democrats who had not backed his New Deal policies, but it largely backfired on him, and his party lost ground in the fall elections.

Presidents tend to be very popular within their own party, but this connection may well be stronger for Trump. Part of this may just be a trait of “charismatic” leadership, in which a politician’s supporters feel a personal connection to him or her, and they become sad or angered when that politician loses or gets in trouble. But it is a rare trait.

Guida: You noted above that, in the 2024 primaries, the local county leaders could not persuade G.O.P. voters away from Trump. Is that because the Republican Party is particularly weak, or is Trump, to your point about charisma, a unique political figure in his relationship to his base of support?

Masket: There’s some great recent research by the political scientists James Hollyer, Marko Klasnja and Rocío Titiunik about the effects that charismatic leaders (they use Prime Minister Indira Gandhi in India as another example) have on a political system. Because of the politician’s strong ties to their supporters, normal aspects of political accountability just don’t apply to them. They don’t lose support when they are mired in scandal or lawbreaking. Their party can’t really discipline them. This can lead to further lawlessness and democratic erosion.

I’ve struggled a lot with just why Trump has that relationship with his supporters. Part of it is just that he was a household name for decades before entering politics, as a businessman and then a TV star. A lot of voters in 2016, both his supporters and detractors, felt they knew him personally.

Trump has also been very deft at using actual moments of accountability — his two impeachment trials, his criminal indictments in 2023 and his actual felony convictions in 2024 — as evidence that the “deep state” is out to get him. This clearly resonated among many of the party chairs I spoke to.

That said, if another candidate tried this sort of thing, would it work? If JD Vance got in trouble for bank fraud or a book royalties scandal, would he be able to convert that to political support? Or would Republicans just ditch him for another candidate? I tend to think the latter.

Guida: Do you think that connection to Trump extends to the ballot box? In 2024, he built an unusual coalition for a Republican (and brought in a lot of infrequent voters), but he isn’t on the ballot this fall. Is his coalition different from a generic G.O.P. one?

Masket: One of the chairs I spoke to described Trump’s support as “shallow” — not as a criticism, but as an admission that a lot of those voters are there specifically for Trump. They generally don’t show up when he’s not on the ballot.

We saw evidence of this in both 2018 and 2022, in which Republicans underperformed. That’s partly because some of his supporters didn’t show up. But there are a lot of Democratic-leaning voters who are increasingly incensed about Trump, and they will show up in a midterm year.

Guida: What does a post-Trump G.O.P. look like, both in terms of its posture to norms and institutions and its policy priorities?

Masket: I asked several party chairs about this following the 2024 election. They generally saw their party as fundamentally changed — it’s not about to snap back to a Reagan-Bush coalition. They figured there’d be more candidacies from unconventional backgrounds — influencers, celebrities, podcasters, news personalities, etc.

All that said, to me, it was notable that in early 2024, when the G.O.P. presidential primaries and caucuses came down to Trump and Haley, she was pulling around a third of the vote in many of those contests. These were overwhelmingly votes from Republicans who knew by that point that Trump was going to be their nominee but still wanted to cast a vote for something different. Those people are still in the party, for the most part. They won’t be in charge after Trump is gone, but they’ll have a louder voice than they do now.

Guida: What other lessons for politics in general do you see shaping up from the Trump years?

Masket: We will continue to struggle with just how unique Trump is or if there are lessons other politicians could learn from him. It turns out a lot of what we think of as “iron laws” of politics are self-imposed. That is, Obama had to have a “beer summit” with Henry Louis Gates Jr. and the police officer who’d arrested him for breaking into his own home, in part to defang concerns that Obama had “racialized” the event too much. Gary Hart had to withdraw from the presidential race in 1987 because a sex scandal had made him unelectable.

There are lots of examples, but they’re almost never actually put to the test. What if it turns out you don’t actually pay much of an electoral price for things the media focuses on? Trump has definitely pushed the limits on this, but it’s not clear how much these lessons apply to others.

Guida: In terms of the laws of politics, we just saw an instance, in the Maine Democratic Senate primary, of a line being crossed and consequences. Is there a difference between the Democratic and Republican Parties in this sense? Do they operate by different laws?

Masket: It’s reasonable to say that if Graham Platner had been the Republican nominee for Senate, and The Times ran an interview describing him committing sexual assault, he probably wouldn’t have withdrawn from that contest. He wouldn’t have gotten nearly the pressure from his party to do so that Democrats applied to Platner. We know this from plenty of examples, including Roy Moore, Matt Gaetz, Pete Hegseth and Donald Trump himself in 2016.

Seth Masket, a professor of political science at the University of Denver, is the author of “The Elephants in the Room: How Trump Voters Seized the Party From Republican Leaders.” John Guida is an editor in Times Opinion.

The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: letters@nytimes.com.

Follow the New York Times Opinion section on Facebook, Instagram, TikTok, Bluesky, WhatsApp and Threads.