Philip Converse
Based on Wikipedia: Philip Converse
The Uncomfortable Truth About How Americans Think About Politics
Here's a finding that might unsettle you: most voters don't actually think about politics the way politicians, journalists, and political scientists assume they do. They don't hold coherent ideologies. They can't reliably explain what "liberal" and "conservative" mean. And when researchers ask them the same political questions two years apart, many respond so inconsistently that their answers look almost random.
This wasn't a fringe theory from a crank. It came from Philip Converse, one of the most influential political scientists of the twentieth century, and his research has shaped how scholars understand democratic citizenship ever since.
The Man Who Measured Belief
Philip Ernest Converse wasn't originally headed for political science. Born in 1928 in Concord, New Hampshire, he earned his bachelor's degree in English from Denison University and then a master's in English literature from the University of Iowa. His path might have led to academia in the humanities, but the Korean War intervened. Drafted into military service, Converse found himself working as a newspaper editor on a base in Battle Creek, Michigan.
Something about that experience, or perhaps the broader intellectual currents of the postwar period, pulled him toward the social sciences. After studying for a time in France, he returned to the University of Michigan, where he earned a master's in sociology in 1956 and a doctorate in social psychology just two years later.
Michigan in the 1950s was becoming the epicenter of a revolution in how scholars studied politics. The Survey Research Center, where Converse worked as an assistant study director even as he pursued his graduate degrees, was pioneering the use of large-scale surveys to understand what ordinary citizens actually thought and did. This wasn't armchair philosophizing about democracy. It was empirical investigation, asking thousands of Americans detailed questions and then rigorously analyzing their answers.
Converse joined forces with three other researchers: Angus Campbell, Warren Miller, and Donald Stokes. Together, they fielded the 1956-1960 National Election Study panel survey, tracking the same respondents across multiple elections. The result was "The American Voter," published in 1960, a book that fundamentally changed political science.
What "The American Voter" Revealed
Before Converse and his colleagues did their work, many political observers assumed that American voters operated more or less like political elites—that they held coherent political philosophies, evaluated candidates against those philosophies, and made rational choices accordingly. Maybe they weren't as informed as senators or newspaper editors, the thinking went, but they used the same basic mental frameworks.
"The American Voter" demolished this assumption.
One of the book's most important contributions was introducing the concept of partisan identity as something distinct from ideology. Partisanship, the authors argued, functions more like an attachment to a social group—like being a Cubs fan or a Methodist—than like a carefully reasoned summary of political values. You don't become a Democrat because you've surveyed all the policy positions and concluded they match your preferences. You become a Democrat because your family was Democratic, or your community is, or because something about the party's image resonates with who you feel yourself to be.
This attachment, what became known as the Michigan Model, is the fundamental driver of how people vote. Everything else—the candidates, the issues, the campaign ads—gets filtered through this pre-existing loyalty.
The book also punctured the idealized image of the independent voter. Conventional wisdom celebrated independents as thoughtful citizens who rose above partisan tribalism to evaluate each election on its merits. The data told a different story. Citizens who didn't identify with either party tended to be the least engaged with politics and the least informed about it. Independence wasn't a marker of sophistication. It was often a marker of disconnection.
The Nature of Belief Systems
If "The American Voter" was influential, Converse's 1964 essay "The Nature of Belief Systems in Mass Publics" was explosive. It remains one of the most cited and debated works in political science.
Converse's central concept was "constraint." Imagine a belief system as a network of connected ideas. In a highly constrained system, if you change your view on one issue, that change ripples through and shifts your positions on related issues. If you become convinced that government spending is generally harmful, a constrained belief system would lead you to oppose not just welfare programs but also military spending, infrastructure projects, and agricultural subsidies. Everything hangs together.
Political elites, Converse found, have highly constrained belief systems. Tell a politician that someone is a "liberal," and they can predict with reasonable accuracy that person's views on taxation, foreign policy, civil rights, and environmental regulation. The positions cluster together in recognizable packages.
But ordinary citizens? Their belief systems showed remarkably little constraint.
Converse demonstrated this through several devastating pieces of evidence. First, he analyzed how Americans actually talked about politics when asked open-ended questions about the parties and candidates. Very few—what he called "ideologues"—used abstract principles like liberalism or conservatism to structure their thinking. A slightly larger group, "near-ideologues," gestured vaguely toward such principles but often misunderstood or misapplied them.
The largest category thought about politics in terms of "group benefits." Which social groups—unions, business owners, farmers, Black Americans—would be helped or hurt by each party? This wasn't ideological thinking. It was a kind of tribal calculation: are the Democrats or Republicans better for people like me?
Others fell into even less sophisticated categories: "nature of the times" (things are good or bad, and I'll vote accordingly) or "no issue content" at all.
The Liberal-Conservative Puzzle
Perhaps even more striking was what happened when Converse's team asked Americans to explain what "liberal" and "conservative" actually meant. Many couldn't do it. Many others gave garbled or contradictory definitions. And crucially, many struggled to connect these terms to the political parties in any coherent way.
Think about what this means. Political commentary is saturated with references to liberals and conservatives, to the left and the right, to ideological battles and philosophical disagreements. But a substantial portion of the public for whom these battles are supposedly being fought doesn't actually understand the terms.
Converse offered an illuminating example. Consider the statement: "Even though it may hurt the position of the Negro in the South, state governments should be able to decide who can vote and who cannot." For most Americans, this statement seemed to be primarily about race relations. But an ideologically sophisticated observer would recognize it as fundamentally about federalism—the balance of power between state and federal government. That deeper ideological dimension was invisible to most respondents.
The Instability Problem
Converse's most controversial finding involved what he called "non-attitudes." When his team asked the same people the same political questions two or four years apart, they found startling instability. On some issues, the correlation between a person's response in 1956 and their response in 1960 was so low that it looked almost like random noise.
This wasn't true for political elites, whose views remained stable over time. And it wasn't true for all citizens on all issues. But for substantial portions of the public on many political questions, Converse concluded that people weren't expressing genuine attitudes at all. They were essentially making up answers on the spot, offering what he later termed "non-attitudes" dressed up as opinions.
The implications were profound. If many citizens don't hold stable political views, what does it mean to say that elections reflect "the will of the people"? What does public opinion polling actually measure? How should we think about democratic representation?
A Broader Pattern
Converse didn't think this was uniquely American pathology. His later work with Roy Pierce, published as "Political Representation in France," found similar patterns in French public opinion. The conclusion seemed to be about mass publics in general, not about any particular national failing.
Political elites construct ideologies. They decide which issue positions go together, which values connect to which policies, what it means to be on the "left" or the "right." Then they broadcast these packages to the public. But most citizens, lacking the time, interest, or cognitive resources to fully absorb these frameworks, receive only fragments. They pick up bits and pieces, often inconsistently, and fill in the gaps with group loyalties, personal circumstances, and responses to whatever's happening in the moment.
The Controversy and the Legacy
Converse's work provoked decades of scholarly debate. Some researchers argued he was too pessimistic, that his measures missed forms of political thinking that didn't map onto elite categories. Others found that apparent instability in survey responses might reflect measurement error as much as genuine attitude change. Still others noted that constraint might vary across issues and over time—the public might think more ideologically about some questions than others, and the level of ideological thinking might rise during polarized periods.
But the core of Converse's insight has proven durable. Most people, most of the time, don't think about politics the way political professionals do. They don't organize their views into coherent philosophies. They don't update their entire belief system when they change their mind about one issue. They often don't have firm opinions on policy questions that elites consider fundamental.
This isn't necessarily a criticism. Converse wasn't saying ordinary citizens are stupid. He was saying they're busy. Politics is one domain among many in a person's life, and most people—reasonably enough—don't invest the time to develop the kind of systematic political thinking that is the daily bread of politicians, journalists, and professors.
A Note on Gender
Reading Converse's work today, some passages land with a thud. His 1964 essay includes the observation that "the wife is very likely to follow her husband's opinions, however imperfectly she may have absorbed their justifications at a more complex level." "The American Voter" suggested that wives who voted but otherwise paid little attention to politics tended to defer to their husbands' judgment.
These weren't unusual sentiments for the 1950s and 1960s. But as later scholars like Christina Wolbrecht and Kevin Corder have noted, the correspondence in voting between married couples admits of multiple explanations. Perhaps women do influence their partners, not just the reverse. Perhaps both spouses are influenced by the same external factors. Perhaps people choose politically like-minded partners, or choose partners based on characteristics that happen to correlate with political views.
The assumption that wives simply follow husbands reflected the gender attitudes of the era more than rigorous analysis of causal mechanisms.
The Personal Story
Converse's career traced an arc through the heights of American social science. He rose from assistant professor to full professor at Michigan with unusual speed, earning tenure and then promotion within five years of joining the faculty. He directed the Center for Political Studies and then the entire Institute for Social Research, the umbrella organization housing Michigan's survey research operations.
In 1989, he left Michigan to direct the Center for Advanced Study in Behavioral Sciences at Stanford University, an elite institution that brings scholars together for intensive interdisciplinary exchange. He returned to Michigan as an emeritus professor in 1994.
The honors accumulated: election to the American Academy of Arts and Sciences, membership in the American Philosophical Society and the National Academy of Sciences, honorary doctorates from the University of Chicago and his undergraduate alma mater Denison, the presidency of the American Political Science Association, Guggenheim and Fulbright and Russell Sage fellowships.
He married Jean McDonnell in 1961, herself a social scientist who directed the Detroit Area Study and was an expert in interviewing techniques. They had two sons.
One curious footnote: Philip's sister, Connie Converse, was a singer-songwriter who recorded music in the 1950s—folk songs of striking originality that languished in obscurity during her lifetime. In the 1970s, she disappeared, driving away from her home and never seen again. Her music was rediscovered decades later, and she has since been recognized as a lost pioneer of the singer-songwriter genre. The Converse family, it seems, produced two remarkable figures: one who revealed uncomfortable truths about public opinion, and one whose private truths about her music only emerged long after she was gone.
Philip Converse died on December 30, 2014, in Ann Arbor, Michigan, at the age of 86.
Why This Matters Now
Converse's research feels especially relevant in an era when politicians and commentators constantly invoke "what the American people want" and when polls measuring public opinion are released daily. His work is a reminder that "public opinion" is a more complicated, more unstable, and frankly weirder thing than we usually acknowledge.
When someone tells you that the public supports or opposes some policy, Converse would want you to ask: How many of those respondents have thought carefully about this issue? How many would give the same answer if asked next month? How many are expressing a genuine attitude, and how many are constructing an answer on the spot because someone asked them a question?
Democracy may not work quite the way civics textbooks suggest. But understanding how it actually works—how real people really think about politics—is the first step toward making it work better. That's the uncomfortable gift Philip Converse gave us.