HOOVER INSTITUTION

Once More Unto the Breach

Is America Polarized?

Morris P. Fiorina

In the wee small hours of November 3, 2004, a new country appeared on the map of the modern
world: The DSA, the Divided States of America. . . . Not since the Civil War has the fault lines
between its two halves been so glaringly clear.
—Simon Schama1

It is time for our society to acknowledge a sad truth: America is currently fghting its second
Civil War. In fact, with the obvious and enormous exception of attitudes toward slavery,
Americans are more divided morally, ideologically, and politically today than they were during
the Civil War.
—Dennis Prager 2

Quotations like the preceding illustrate what academics refer to as “conventional wisdom.”
This instance of conventional wisdom sprouted in the early 2000s, took deep root during
the 2004 George Bush–John Kerry presidential race, and has continued to grow ever since.3
It holds that the United States is a deeply polarized country, probably more so than at any
time since the Civil War (notably overlooking more than a half-century of labor violence from
1870–1930, not to mention the 1960s).4 In the past two decades only a very brief period of
thermidor between Obama’s victory in 2008 and the rise of the Tea Party temporarily called
the conventional wisdom into question.

Culture War: The Myth of a Polarized America (2004) was a frst attempt to counter the bur-
geoning conventional wisdom.5 Putting forth a “man bites dog” argument, the book received
extensive discussion in the media.6 Its central thesis was soon forgotten, however. Disconnect:
The Breakdown of Representation in American Politics (2011) was a second attempt.7 As
an academic book it received much less attention. Unstable Majorities: Polarization, Party
Sorting and Party Stalemate (2017), the predecessor of this essay series, seemingly had no
impact at all.8 Nothing in the data then or now suggests that I should renounce these earlier

An Era of Unstable Majorities Continues: Essays on Contemporary American Politics

Essay 2, October 2024

 arguments, however; so, once more unto the breach, dear friends.9 As Chico Marx might have
said, “Who you gonna believe, the data or your own eyes?”10

WHERE’S THE POLARIZATION?

As elementary statistics courses ofen note, in the natural world many variables, such as
height, weight, or intelligence, follow a bell-shaped or normal distribution as depicted in the
top panel of fgure 1. Most observations fall close to the center and become rarer toward the
extremes. In contrast, the bottom panel illustrates a “polarized” distribution, with most obser-
vations occurring in the tails of the distribution and few in the middle. Many observers claim
that in past decades distributions of American political attitudes looked like the top panel
but now look like the bottom panel. As James Pierson writes, “The number of people and the
percentage of the electorate at the center have gradually diminished over time. Public opinion
now appears to divide us up to the point that we have a couple of lumps—a liberal lump on
one side and a conservative lump on the other.”11 So, if we consider, say, political ideology,
believers in a polarized America would contend that the United States has divided into liber-
als and conservatives with few moderates in between.

Is this true?

For fve decades, academic survey organizations have asked Americans to characterize their
ideologies. Figure 2 summarizes the data from one such organization, the General Social
Survey (GSS). Contrary to claims like Pierson’s quoted earlier, there is little change over the
half-century even as millions of older voters lef the electorate and were replaced by younger

FIGURE 1 Two contrasting political distributions

Not polarized

Polarized

2 MORRIS P. FIORINA U ONCE MORE UNTO THE BREACH

FIGURE 2 No change in ideological self-identifcations

100%

75%

50%

25%

0%

1976 1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 2020
Year
Liberal Conservative Moderate

Source: General Social Survey

ones. The data show a liberal lump and a conservative lump, but an even larger middle-of-the-
road lump. There is essentially no trend except for a small rise in liberal identifcation during
the period of the “great awokening” since 2011.12 Even this apparent movement might be
exaggerated because of pandemic-created survey difculties.13

Perhaps ideology is not an appropriate measure. According to another major survey orga-
nization, over the same period, between one-ffh to one-third of the electorate report that
they don’t know what they are, ideologically speaking.14 That is because many, if not most,
Americans look at specifc issues without putting them in a broader ideological context.15
The American National Election Study (ANES) includes such issue-specifc questions. The
time series is not as lengthy as the ideology series, but fve issues have a reasonably long
representation in the data. Between 1984 and 2020 survey respondents were invited to posi-
tion themselves on seven-point scales ranging from extremely liberal to extremely conserva-
tive. Although these issue distributions show some increase in extreme positions over the
past thirty-six years, the shape of the contemporary distribution in fgure 3b still resembles
the upper panel of fgure 1 far more closely than the bottom panel. Even on some of the past
decade’s most contentious issues such as healthcare, the proportions who hold extreme
positions like adopting a single-payer model or leaving everything to insurance companies
are far smaller than the proportions who espouse something in between.

Still, a long tradition in political science holds that most people do not have frm positions
on political issues—and even that a signifcant number adopt whatever positions their pre-
ferred candidates advocate. For example, studies found in 2016 that some voters frst decided
whom to vote for and then adopted the economic and cultural positions of that candidate.16

HOOVER INSTITUTION U STANFORD UNIVERSITY 3

 FIGURE 3A Issue positions are centrist (1984)

40%

30%

20%

10%

0%

1 2 3 4 5 6 7
Year
Insurance Aid Jobs Military Services

Source: American National Election Studies

FIGURE 3B Issue positions remain centrist (2020)

40%

30%

20%

10%

0%

1 2 3 4 5 6 7
Year
Insurance Aid Jobs Military Services

Source: American National Election Studies

4 MORRIS P. FIORINA U ONCE MORE UNTO THE BREACH

Other studies similarly report that issue preferences change with candidate preferences.17
But there are a few “easy” issues that are the exception to such arguments.18 Abortion is one
such issue, a long-standing controversy that some consider the single most important con-
tributor to the “culture war.” One recent study found that abortion was the only issue in the
modern period for which people would change their party identifcation to agree with their
issue stance, rather than vice versa.19 The GSS has queried Americans about their views on
abortion for fve decades, with the starting date in 1972, just before the 1973 Roe decision.
Respondents answer the following question battery:

Please tell me whether or not you think it should be possible for a pregnant woman to obtain
a legal abortion . . .

if pregnancy endangers the woman’s health

if the woman has become pregnant as a result of rape

if there is a strong chance of a serious defect in the baby

if she does not want any more children

if her family is very low income and cannot aford more children

if she is unmarried and does not want to marry the man.

Once again, the data surprise. Figure 4 shows that, although there is a clear uptick in pro-
choice views in the three elective circumstances between 2016 and 2022, before that time
there was little movement on the issue for more than four decades while millions of voters
entered and exited the electorate and polarization supposedly surged. Large majorities
favored abortion in the so-called traumatic conditions while splitting nearly evenly about the
more “elective” conditions.20 The average American believed that abortion should be legal
in about four of the six circumstances, with a large plurality adopting positions between the
extreme zero and six poles.

A less specifc Gallup survey item frst asked in 1975 shows similar continuity (fgure 5).
The item reads, “Do you think abortions should be legal under any circumstances, legal
only under certain circumstances, or illegal in all circumstances?” Once again, for about
a half-century “any” and “none” are chosen much less frequently than “only under certain
circumstances,” which half the country supports.

For a slightly shorter period the ANES has posed a diferent question: “By law, when should
abortion be allowed?” And again, fgure 6 shows little change in Americans’ views. Support for
the most liberal position has trended slightly upward over the past two decades, but the item
shows little movement overall.21

HOOVER INSTITUTION U STANFORD UNIVERSITY 5

 FIGURE 4 Pregnant women should be able to obtain an abortion if . . .

100%

75%

50%

25%

0%
1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 2020
Year

Low income˜can't a°ord more children Pregnant as result of rape

Married˜wants no more children Strong chance of serious defect
Not married˜wants no more children Woman’s health seriously endangered

Source: General Social Survey

FIGURE 5 Do you think abortion should be . . . ?

100%

75%

50%

25%

0%

1976 1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 2020
Year
Legal only under certain circumstances I˜˜egal under all circumstances

Legal under any circumstances

Source: Gallup

To some extent polarization is in the eye of the beholder; yet, whatever one’s view, and con-
trary to what some commentators assert, polarization on abortion has changed little over the
past half-century.22 For decades surveys nearly always measured support for Roe to fall in the
mid-60s percentage range. Americans have long difered on exactly where to draw the line,
but large majorities have always drawn it far from the “always” or “never” points where abor-
tion issue activists would draw it.

6 MORRIS P. FIORINA U ONCE MORE UNTO THE BREACH

FIGURE 6 By law, when should abortion be allowed?

100%

75%

50%

25%

0%

1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 2020
Year
Always as personal choice Don't know/other
Clear need Never permitted
Rape, incest, danger to health

Source: American National Election Studies

Surprisingly, the much-publicized Dobbs decision in June 2022 that overturned Roe did little
to change public opinion on the issue. Consistent with past support for Roe, recent polls
report that more than 60 percent of Americans disapprove of the decision, and their specifc
views on abortion did not show any notable movements afer the Court handed down the
Dobbs decision.23 The Pew Research Center asks a question that is similar to Gallup’s in their
trends panel and reports little change. Table 1 compares the distributions almost two years
afer Dobbs with the distribution two years before.

The failure of restrictive abortion laws and initiatives to gain popular support in state elec-
tions afer Dobbs illustrates the electorate’s general acceptance of the framework estab-
lished by Roe.

If general ideology or specifc policy issues do not show the of-claimed polarization of
Americans, maybe polarization can be found in something more primordial: our identities—us
versus them, my team versus your team, Republican versus Democrat. Perhaps polarization
is not something that occurs on the cognitive level but instead on an emotional or afective
level. This brings up the topic of “afective polarization,” one of the more active political sci-
ence research programs of the past decade or so. A subsequent critical essay will deal at
length with this topic, but for now, consider this observation.

The view that Americans have split down the middle into two warring partisan camps has one
obvious problem: far more Americans belong to neither camp than they did a generation ago
in the prepolarization era. For nearly three-quarters of a century, the ANES has presented
Americans with the following survey battery (fgure 7): “Generally speaking, do you usually
think of yourself as a Republican, a Democrat, an Independent, or what?” If the response

HOOVER INSTITUTION U STANFORD UNIVERSITY 7

 TABLE 1 PUBLIC OPINION ON ABORTION BEFORE AND AFTER DOBBS

Jan 22–Feb 2, 2020 (%) Apr 8–14, 2024 (%)

Legal in all cases 25 25

Legal in most cases 35 38

Illegal in most cases 27 28

Illegal in all cases 11 8

Source: “Broad Public Support for Legal Abortion Persists 2 Years Afer Dobbs,” https://www
.pewresearch.org/politics/2024/05/13/broad-public-support-for-legal-abortion-persists-2
-years-afer-dobbs/.

FIGURE 7 Not a 50/50 partisan nation—closer to a 33/33/33 nation

100%

75%

50%

25%

0%

1952 1956 1960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 2020

Year
Strong and weak Democrats Strong and weak Republicans
Independents including leaners

Source: American National Election Studies

is Democrat or Republican, the respondent is classifed as a partisan, and the interviewer

proceeds to a strength of identifcation probe: “Would you call yourself a strong Republican
[Democrat] or a not very strong Republican [Democrat]? If the answer to the stem question
is Independent, however, the interviewer next asks, “Do you think of yourself as closer to the
Republican Party or the Democratic Party?”

In the 1950s about three-quarters of the electorate reported Democratic or Republican identif-
cation, with only a small minority in the independent category. In the ensuing years Democrats
lost adherents, but rather than the Republicans gaining them to produce a more polarized
half-Republican and half-Democratic electorate, the general trend was for fewer and fewer

8 MORRIS P. FIORINA U ONCE MORE UNTO THE BREACH

Americans to admit identifying with either party. In the spring of 2023 Gallup reported that the
proportion of independents in its surveys had reached an all-time high.24 There are numerous
claims in the academic and popular literature that most of the independents are “closet parti-
sans,” a generalization that rests on weak supporting evidence and ignores contrary evidence.25
But even if the claim were completely true, how widespread or strong could partisan identities
be if less than 60 percent of the electorate will even admit that they have one?

In sum, claims that the United States of America has divided into the Divided States of America
fnd little support in the data. Yet, American politics over the course of the past few decades
indisputably seems nastier and less productive than in earlier periods. One explanation for
this apparent contradiction is simple: as discussed in Essay 1, “Historical Context: An Era of
Tenuous Majorities Continues,” the American electorate in the aggregate has not polarized,
but the two parties have.

WHAT DID HAPPEN: NOT POLARIZATION, BUT PARTY SORTING

Let us dig deeper into the data underlying fgures 2 and 6, which both show little change
in the American electorate over the past half-century. Figure 8 divides the data in fgure 2
into Democrats and Republicans. The proportion of Democrats who classifed themselves
as liberals more than doubled between the 1970s and today. In 1980 there were as many
Democratic conservatives as there were liberals, whereas the ratio is now 5:1 in favor of liber-
als. The picture is similar for Republicans. With the exception of the George H. W. Bush years,
there has been a fairly steady rise from 50 percent or so conservative to nearly 75 percent of
Republicans now. Like conservative Democrats, liberal Republicans now register in the single
digits. Whereas both parties historically contained both liberals and conservatives, today’s
Democrats are a liberal party, and today’s Republicans are a conservative party.

FIGURE 8 Liberal Democrats and conservative Republicans

100%

75%

50%

25%

0%

1976 1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 2020
Year
Democrat and liberal Republican and conservative

Source: General Social Survey

HOOVER INSTITUTION U STANFORD UNIVERSITY 9

 Now consider figure 9, which divides Americans’ views on abortion by party. Until the early
1990s when the Webster and Casey decisions were handed down, the average Republican
and Democrat did not differ much on abortion. But then the process of sorting began and has
accelerated since.

What the preceding figures show is that the parties have sorted: Democrats have moved in
one direction and Republicans the opposite. These changes largely offset so the aggregate
distributions of opinion shown earlier do not change very much. Analyses such as these
have been replicated across other issues and political beliefs.26 Although the political views
of the electorate at large have changed little in recent decades, the distributions of those
views across the two parties have changed dramatically. We could call this development
party polarization, but I use the term “party sorting” for two reasons. First, political views
have not become more extreme, but have become more consistent, as explained later.
Second, commentators too often omit the modifier “party” when discussing party polariza-
tion; some younger scholars even define polarization as party sorting.27 According to that
definition however, 1860 would not qualify as a polarized period because that election did
not pit one party against the other. Lest readers think this is an exaggerated claim, a recent
history of party polarization, in fact, does omit the Civil War because it was not a case of
party polarization.28

Even in the modern era there are examples of polarization in the electorate where deep politi-
cal divisions do not pit one party against the other. Take race in the late 1950s to the early
1970s when the Democratic Party contained both voters most opposed to further civil rights
advances and those most supportive, while Republicans were in between. This was a time of
considerable popular polarization on civil rights but not party polarization. Another example
from a slightly later era would be public opinion on the continuation of the Vietnam War in the

FIGURE 9 Partisan sorting on abortion: abortion should always be permitted

100%

75%

50%

25%

0%
1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 2020
Year
Democrats Republicans
Independents including leaners

Source: American National Election Studies

10   MORRIS P. FIORINA U ONCE MORE UNTO THE BREACH

late 1960s. The Democrats were the party of both antiwar protestors and police and blue-
collar “hardhats” who brutalized the protestors. In both cases popular polarization was sig-
nificant and led to violence far more often than in the current era, but the parties were not yet
sorted. Polarization over the war took place largely within the Democratic Party.

SOME ADDITIONAL OBSERVATIONS

The sorting process that has occurred in recent decades has three features that should be noted
before moving on. First, the process began earlier and has proceeded more completely among
the most active participants in politics than among the less active and involved. Candidates
and elected officials are very well sorted, as much literature reports.29 So are donors and
assorted activists. Strong partisans are not as sorted as activists but are better sorted than
weaker partisans. Figure 10 illustrates the gradations of sorting in the case of abortion.

Second, the sorting is by no means perfect. Joe Manchin may be a unicorn in the Senate,
but more Democrats like him exist in the larger electorate, albeit they are much less numer-
ous than they used to be. Candidates bundle issue positions, and voters gradually adopt
those bundles (or sometimes not).30 Figure 11 may surprise many readers. Although minorities
in their parties, about one-quarter of self-identified Republicans are out of synch with their
party on abortion, believing that it always is a matter of a woman’s choice. A similar propor-
tion of Democrats are out of synch with their national party platform in opposing proposals to
strengthen gun laws. And independents, of course, are even less well sorted. The electorate
as a whole is not nearly as well sorted as their elected representatives.

FIGURE 10 Partisan sorting on when abortion should be permitted increases with

political involvement

100%

75%

50%

25%

0%

1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 2020
Year
Strong partisans Weak partisans
Donated to or worked for a political campaign

Source: American National Election Studies

HOOVER INSTITUTION U STANFORD UNIVERSITY   11

 FIGURE 11 Some partisans remain out of sync with their parties

100%

75%

50%

25%

0%

2000 2004 2008 2012 2016 2020

Year
Pro-gun Democrats Pro-choice Republicans

Notes: Percentage of Republicans who say abortion should always be legal as a personal choice versus percent-
age of Democrats who want to keep gun laws the same or want to make it easier to purchase a gun.

Source: American National Election Studies

Third—and this is an important point, although somewhat more technical—Americans’ views

are no more extreme than they used to be. Party sorting means that political opinions have
become more ideologically consistent. A pro-choice voter today is also more likely to be a
racial and economic liberal than previously. This is why when issues such as those graphed in
fgure 3 are combined into an index, public opinion appears to have become more polarized.
But changes in such indexes refect increased consistency of the underlying issue positions,
not greater extremity.31

CONCLUSION

The political changes that have occurred during the past few decades have lef us with two
parties that are more internally homogeneous than in times past and more distinct from each
other. To put a bit of qualitative imagery on the data presented in this essay, consider the fol-
lowing timeline of events in national politics:

In 1964 southern Democrats flibustered the civil rights bill for ffy-seven days. Twenty-seven
Republican Senators joined northern Democrats to break the flibuster.

In 1972 environmental activists targeted the “Dirty Dozen” antienvironmental members of

Congress for defeat. Five of the dozen were Democrats, including the chairman of the House
Interior Committee who was defeated in the primary.

12 MORRIS P. FIORINA U ONCE MORE UNTO THE BREACH

 In 1981 the House of Representatives passed Republican President Reagan’s tax cuts.
Phil Gramm, a Democrat from Texas, managed the bill in the House, and forty-eight
Democrats joined the minority Republicans to pass it.

In the 1990s, gun control legislation was regularly slowed, weakened, or both in the House of
Representatives by Energy and Commerce Chair John Dingell, a Democrat who served on the
board of the NRA.

Fast forward to the current century:

In 2001 Congress passed Republican President Bush’s tax cuts. Only 13 Democratic
representatives (of 209) voted for it. Only 10 (of 50) Democratic senators voted for it.

In 2010 Congress passed Democratic President Obama’s Afordable Care Act. No (zero)
Republican senators or representatives voted for it.

In 2022 Congress passed Democratic President Biden’s Infation Reduction Act. No (zero)
Republican senators or representatives voted for it.

The diference between the cross-party coalitions that supported major legislation before
2000 and the partisan coalitions in the Congresses since then are the refections of party
sorting, aka partisan polarization.

NEXT: WHAT ABOUT AFFECTIVE POLARIZATION?

NOTES
1. Simon Schama, “Onward Christian Soldiers,” November 5, 2004, http://rense.com/general59/cche.htm.
2. Dennis Prager. 2017. “America’s Second Civil War,” January 24, 2017, https://dennisprager.com/column
/americas-second-civil-war.
3. Morris Fiorina, “A Divider, Not a Uniter—Did It Have to Be?” in The George W. Bush Legacy, edited by
Colin Campbell, Bert A. Rockman, and Andrew Rudalevige (Washington, DC: CQ Press, 2007), 92–111.
4. To be discussed in Essay 10 of this series.
5. Morris Fiorina, with Samuel Abrams and Jeremy Pope, Culture War? The Myth of a Polarized America
(New York: Pearson Longman, 2004).
6. Wayne Baker, “Social Science in the Public Interest: To What Extent Did the Media Cover ‘Culture War?’
The Myth of a Polarized America?” The Forum 3, no. 2, January 2005, https://www.researchgate.net
/publication/40823276_Social_Science_in_the_Public_Interest_To_What_Extent_Did_the_Media_Cover
_Culture_War_The_Myth_of_a_Polarized_America.
7. Morris Fiorina, with Samuel Abrams, Disconnect: The Breakdown of Representation in American Politics
(Norman: University of Oklahoma Press, 2011).
8. Morris Fiorina, Unstable Majorities: Polarization, Party Sorting and Political Stalemate (Stanford:
Hoover Institution Press, 2017).

HOOVER INSTITUTION U STANFORD UNIVERSITY 13

 9. From William Shakespeare’s Henry V, https://www.poetryfoundation.org/poems/56972/speech-once
-more-unto-the-breach-dear-friends-once-more.
10. From the classic movie Duck Soup, https://www.youtube.com/watch?v=cHxGUe1cjzM.
11. Quoted in Andrew Soergel, “Divided We Stand,” US News and World Report, July 19, 2016, https://www
.usnews.com/news/articles/2016-07-19/political-polarization-drives-presidential-race-to-the-bottom.
12. Musa al-Gharbi, We Have Never Been Woke (Princeton, NJ: Princeton University Press, 2024).
13. The pandemic prevented face-to-face interviewing that these survey organizations had always done
in the past. Studies fnd that “web respondents appear to be more politically knowledgeable, ideologically
constrained, and have greater correspondence between issue and vote preferences compared to face-to-
face or phone respondents.” Thus, other things being equal, we would have expected to see more liberals and
conservatives in 2020. See Enrijeta Shino and Michael D. Martinez, “The Diferent Faces of Public Opinion: Is
the American Voter Tinted by Mode?” International Journal of Public Opinion Research 33 (2021): 756–78.
14. “ANES Guide to Public Opinion and Electoral Behavior,” American National Election Study, 2023,
https://electionstudies.org/data-tools/anes-guide/anes-guide.html?chart=lib_con_identifcation_7_pt.
15. See, e.g., Donald R. Kinder and Nathan P. Kalmoe, Neither Liberal nor Conservative (Chicago: University
of Chicago Press, 2017).
16. Peter K. Enns and Ashley Jardina, “Complicating the Role of White Racial Attitudes and Immigrant
Sentiment in the 2016 U.S. Presidential Election,” Public Opinion Quarterly 85 (2021): 539–70.
17. See, e.g., Andres M. Engelhardt, “Racial Attitudes Through a Partisan Lens,” British Journal of Political
Science 51, no. 33 (2020): 1062–79.
18. Edward G. Carmines and James A. Stimson, “Two Faces of Issue Voting,” American Political Science
Review 74 (1980): 78–91.
19. Robert S. Erikson, “Abortion Opinion and Partisan Choice: Untangling the Causal Dynamics,” Political
Science Quarterly,” May 3, 2024, https://doi.org/10.1093/psquar/qqae041.
20. These are terms commonly used by researchers in this area.
21. Responses to “Always as personal choice” (and to a similar item in the GSS) should be treated with caution
because a signifcant fraction of respondents who ofer that response rethink it when queried more closely. In
an ANES pilot study about 30 percent of respondents who chose the “always” option in one survey rejected
gender selection when subsequently ofered that justifcation for abortion in a subsequent survey. Similarly,
GSS researchers report that 20 percent of respondents who answer that abortion should NOT be legal in
one of the six circumstances chose “always legal,” seemingly contradicting themselves. Sarah K. Cowan,
Michael Hout, and Stuart Perrett, “Updating a Time-Series of Survey Questions: The Case of Abortion
Attitudes in the General Social Survey,” Sociological Methods & Research 53, no. 1 (2022): 193–234.
22. In striking contrast, support for gay marriage moved rapidly from not polarized (large majorities against)
to polarized, to not polarized (large majorities in favor). Pew Research Center, “Attitudes on Same-Sex
Marriage,” 2019, https://www.pewresearch.org/religion/fact-sheet/changing-attitudes-on-gay-marriage/.
23. “Abortion and Birth Control,” Polling Report, 2022, https://www.pollingreport.com/abortion.htm.
24. Jefrey M. Jones, “Independent Party ID Tied for High; Democratic ID at New Low,” Gallup, January 12,
2024, https://news.gallup.com/poll/548459/independent-party-tied-high-democratic-new-low.aspx.
25. See Fiorina, Unstable Majorities, chap. 6.
26. Matthew Levendusky, The Partisan Sort (Chicago: University of Chicago Press, 2009).
27. See, e.g., Daniel J. Hopkins, The Increasingly United States (Chicago: University of Chicago Press,
2018), 133.
28. B. Dan Wood, with Soren Jordan, Party Polarization in America: The War Over Two Social Contracts
(New York: Cambridge University Press, 2017).
29. See, e.g., Nolan McCarty, Polarization: What Everyone Needs to Know (New York: Oxford University
Press, 2019).
30. O’Brian shows that positions on various cultural issues were correlated with race in the population
long before they were bundled by candidates; hence, when such issues became salient afer the racial

14 MORRIS P. FIORINA U ONCE MORE UNTO THE BREACH

realignment, it was more natural for candidates to construct some issue packages than others. This is a
helpful fnding, but it still took activists and candidates to notice the potential market and act to stimulate
and organize the latent demand. Neil O’Brian, The Roots of Polarization: From the Racial Realignment to the
Culture Wars (Chicago: University of Chicago Press, 2024).
31. For a detailed discussion see Morris Fiorina and Matthew Levendusky, “The Political Class vs. the
People,” in Red and Blue Nation, vol. 1, edited by Pietro S. Nivola and David W. Brady (Washington, DC:
Brookings Institution Press, 2006), 96–99.

The publisher has made this work available under a Creative Commons Attribution-NoDerivs
license 4.0. To view a copy of this license, visit https://creativecommons.org/licenses/by-nd/4.0.

Copyright © 2024 by the Board of Trustees of the Leland Stanford Junior University

The views expressed in this essay are entirely those of the author and do not necessarily refect the
views of the staf, ofcers, or Board of Overseers of the Hoover Institution.

30 29 28 27 26 25 24 7 6 5 4 3 2 1

Author photo credit: Hoover Institution

HOOVER INSTITUTION U STANFORD UNIVERSITY 15

 ABOUT THE AUTHOR

MORRIS P. FIORINA
Morris P. Fiorina is a professor of political science at Stanford University and a
senior fellow of the Hoover Institution. He has written or edited fourteen books,
most recently Who Governs? Emergency Powers in the Time of COVID. An
elected member of the National Academy of Sciences, Fiorina has received
career achievement awards from two sections of the American Political
Science Association.

An Era of Unstable Majorities Continues

A continuation of the Hoover Institution’s Unstable Majorities series from the 2016 election season, the frst
half of this essay series leads up to the November 2024 elections with general discussions of the past and
present political situation, of particular interest to students and professionals in the felds of political sci-
ence and political journalism. The second half continues post-election with analyses focused specifcally on
the 2024 elections, addressed to a wider audience. The series begins by looking back at the issues raised
in 2016 that continue today.

Hoover Institution, Stanford University Hoover Institution in Washington

434 Galvez Mall 1399 New York Avenue NW, Suite 500
Stanford, CA 94305-6003 Washington, DC 20005
650-723-1754 202-760-3200