1

Why Concepts Creep to the Left

Jonathan Haidt
New York University -- Stern School of Business
October 30, 2015
Comment on Haslam (in press), Concept Creep.
Forthcoming in Psychological Inquiry

Abstract
Nick Haslam's forthcoming paper, titled "Concept Creep" (available on SSRN)
shows that many concepts in psychology have changed over time (e.g., bullying, trauma,
addiction). Meanings shift so that these concepts apply to more phenomena and smaller
phenomena . In this paper I extend Haslam’s analysis to explain why they change in one
direction only: they “creep to the left.” As psychology has become politically purified, its
concepts have morphed to make them more useful to social justice advocates trying to
prosecute and convict their opponents. This political shift poses a grave danger to the
credibility of psychology.

*****

Last year I co-wrote an essay with Greg Lukianoff titled The Coddling of the
American Mind. Lukianoff and I analyzed several new concepts that have been spreading
rapidly around the academy – but almost nowhere else in American society. The two most
colorful are “Trigger warnings” (warnings given to students before professors assign
readings that might reactivate painful memories in survivors of trauma) and
“microaggressions” (words, questions, or even facial expressions that have the effect –
often unintended -- of making another person feel marginalized, different, or excluded). A
search for these terms on Google Trends shows that they were barely mentioned before
2012, but have been rising rapidly in popularity since late 2013.
These terms are part of a new conceptual package that includes all of the older
concepts long referred to as “political correctness” but with greatly expanded notions of
harm, trauma, mental illness, vulnerability, and harassment. These concepts seem to have
expanded in just the way that Haslam (2016) describes -- horizontally, to take in new
kinds of cases (such as adding the reading of novels to the list of traumatizing activities)
and vertically, to take in ever less extreme versions of older cases (as is made explicit by
the prefix “micro” in the word “microaggression”). In this conceptually augmented
political correctness, the central idea seems to be that many college students are so fragile
that institutions and right-thinking people must all work together to protect vulnerable
individuals from exposure to words and ideas that could damage them in a lasting way. If

Electroniccopy
Electronic copy available
available at:
at:https://ssrn.com/abstract=2702523
http://ssrn.com/abstract=2702523
 2

this protection requires banning certain speakers from campus, or punishing student
newspapers that publish opinions that upset the dominant campus sensibility, then so be it.
But the reactions to the Coddling article—in essays, blog posts, and commentary on
the original post--has revealed a large generational gap. As far as I can tell, the vast
majority of Americans over the age of 40, including progressives, and including
progressive professors, dislike the illiberal tendencies of the new political correctness.
They do not share the view that college students must be shielded from words, books, and
visiting speakers. These older progressives value freedom of speech to such an extent that
they oppose efforts to shut down student newspapers or shout down professors or visiting
speakers. President Obama himself recently spoke out against “coddling,” and in favor of
vigorous cross-partisan debate on campus (see quotes in Haidt, 2015). There has been
hardly any published criticism of the Coddling article, but what little pushback there has
been has come almost exclusively from current college students and from humanities
professors under the age of 35 (e.g., Manne, 2015).
Why is this? Why has this new and expansive sense of student fragility spread so
rapidly, but only among millennials who are currently living or working on college
campuses? Lukianoff and I tried to explain the recent spread of trigger warnings and
micro-aggression theory by examining broad historical trends, such as increases in
protective parenting that began in the 1980s, and we examined more recent changes in
federal laws that pressured universities to over-police language use on campus. But
Haslam’s explanation of concept creep provides a large and crucial missing piece of the
story. In this essay I expand upon a point that Haslam (2016) raised only briefly at the end
of his essay: concept creep has happened primarily to concepts related to a left-liberal
moral agenda. As he noted on p. <to come>:

the concept creep phenomenon broadens moral concern in a way that aligns with a
liberal social agenda by defining new kinds of experience as harming and new
classes of people as harmed, and it identifies these people as needful of care and
protection.

I position concept creep within the recent historical trend of rising political
polarization, particularly “affective partisan polarization,” which refers to the increasing
hostility felt by partisans toward people on the other side. I tell this story in three graphs.
Together, the trends in these graphs can explain why concepts of trauma and victimhood
have undergone such rapid expansion on university campuses and among psychologists.
In brief, the loss of political diversity in many universities--and in psychology in
particular--at a time of rising cross-partisan hostility has amplified the already powerful
process of motivated reasoning. Concepts are morphing to become ever more useful to
“intuitive prosecutors” (Tetlock, 2002) who are prosecuting their enemies in the culture
war.

Electroniccopy
Electronic copy available
available at:
at:https://ssrn.com/abstract=2702523
http://ssrn.com/abstract=2702523
 3

Graph #1: Political Polarization is Rising in Congress

The US Congress went through a period of historically low polarization from the
1920s through the 1970s, as you can see in Figure 1, which plots the degree to which a
legislator’s votes on major issues can be predicted if all you know about that legislator is
his or her placement on the left-right dimension. Ideology has become steadily more
powerful as a predictor of voting, particularly since the 1980s. Before the 80s, politics was
more flexible; the existence of conservative Democrats (mostly in the South) and liberal
Republicans (mostly in the North East and North West) made it much easier than it is
today to create bipartisan coalitions in support of major projects and reforms. But for a
variety of reasons (described in Haidt & Abrams, 2015), the parties began to shuffle and
purify themselves in the 1970s so that by the 1990s, the Democrats had become the liberal
party and Republicans were the conservative party. Crossovers, and even moderates, were
less and less welcome, particularly in the Republican Party.

Figure 1. Political polarization in the U.S. Congress has risen since the 1980s.
Source: McCarty, Poole, & Rosenthal (2006), updated graph retrieved from
http://voteview.com/Polarized_America.htm on October 24, 2015

Graph #2: Affective Partisan Polarization is Increasing Among Americans

As the two parties were sorting themselves along the left-right axis, what it meant to
be a Republican or a Democrat changed, and hostility to the increasingly purified and
increasingly extreme “other side” intensified. Figure 2 shows data from the American
National Election Survey, from the “feeling thermometer” section in which a
representative sample of Americans are asked to rate how warm or cold they feel toward

Electronic copy available at: https://ssrn.com/abstract=2702523

 4

various groups and institutions in American society. The top two lines show that
Americans have held steady in their generally positive ratings of their own party. The
bottom two lines show that there has been a steady drop in how they feel about the other
side. It is interesting to note that in the 1980s, cross-partisan ratings were only a little bit
below 50 – not very “cold.” But the decline began to accelerate after 2000 – particularly
for Democrats responding to the Republican party of George W. Bush and the war in Iraq.

Figure 2. Affective partisan polarization: Americans increasingly dislike the other

side, compared to their own side. Data from American National Election Survey,
graphed by Sam Abrams.

The net effect has been a sharp rise in public expressions of anger and incivility, at
the elite level (e.g., a Republican congressman shouting “you lie!” to President Obama) as
well as at the mass level (as seen in populist movements such as The Tea Party and
Occupy Wall Street). What effect might this national trend of increasing cross-partisan
hostility have had on University life? How might it have changed psychology in
particular?

Graph #3: Psychology is Becoming Politically Purified

In 2011, I gave a talk at the annual meeting of the Society for Personality and Social
Psychology in which I argued that social psychology was becoming a “tribal moral
community” bound together by moral commitments to social justice and progressive
ideals. I described ways in which the absence of conservatives in the field made it harder
to do good research. I was later joined by four other social psychologists (Jose Duarte,
Jarett Crawford, Lee Jussim, and Phil Tetlock) as well as one sociologist (Charlotta Stern)
in laying out a more formal and complete argument. Our paper (Duarte et al., 2015) was
titled “Political diversity will improve social psychological science.” In that paper,

Electronic copy available at: https://ssrn.com/abstract=2702523

 5

Charlotta Stern pulled together all extant published information on the political identities
of academic or research-oriented psychologists, and she plotted all available data points on
a single graph. The graph, reprinted as figure 3 below, stunned us. We had not realized
just how rapidly psychology had changed in the 1990s when the Greatest Generation
(which had some political diversity in it) retired just as the Baby Boomers were coming to
dominate the field.

Figure 3. The rising ratio of Left to Right in psychology. Circles show ratios of
self-reports of liberal vs. conservative. Diamonds show ratios of self-reports of
party preference or voting (Democrat vs. Republican). See Duarte et al. (2015) for
further explanation.

The diamonds on the left side of the graph represent psychologists’ recollections of
which party they voted for in prior presidential elections, as assessed in the early 1960s
(McClintock et al., 1965). The ratio of voting for Democrats versus Republicans mostly
ranges from 2:1 to 4:1, and a similar ratio was found for self-reports of being liberal
versus conservative -- 4:1 -- as late as 1990. But after 1990, whether the question is party
preference or ideology (liberal-conservative), the ratio skyrockets. Psychology is rapidly
purging itself of all political diversity.
I do not believe it is problematic when an academic field leans left, as psychology
did before the 1990s. In a free society few fields will end up with perfectly proportional
representation by politics, gender, race, or other criteria. As long as there are sure to be
some conservatives (or women, or African Americans) to review papers, speak at

Electronic copy available at: https://ssrn.com/abstract=2702523

 6

symposia, and otherwise challenge the biases and prejudices of the dominant group, the
scientific process of institutionalized questioning can function.
But when the ratio of liberals to conservatives rises above a certain point (five-to-
one? Ten-to-one?) we get a phase change. People start to assume that everyone in the
room shares their politics. They start making jokes, from the lectern, about conservatives.
They create a hostile climate, and the few remaining non-liberals begin to hide their
views. Non-liberal graduate students and assistant professors are particularly vulnerable to
discrimination, as many have told me (see their stories at Haidt, 2011).
In the process of writing our political diversity paper, we learned that psychology is
not unique in undergoing a political purification process with rising hostility toward
political minorities. Most fields in the social sciences and humanities seem to be
experiencing these trends (Klein & Stern, 2009). This makes sense if the rising cross-party
hostility shown in Figure 2 is a national trend.

Creeping Concepts for an Intensifying Culture War

Looking at the three graphs above and then returning to Haslams’ essay, we can now
see the underlying processes that might be driving concepts to the left. I don’t need to
review for this readership the power and ubiquity of motivated reasoning and the
confirmation bias. I will just add that these processes--which are usually studied in
individuals acting on self-serving motives--often become much stronger when individuals
are embedded in groups that are locked in combat with other groups. (For a review, see
Haidt, 2012, ch. 4.) Intense tribalism is fundamentally incompatible with open mindedness
and the search for truth because changing your mind -- or merely acknowledging nuance -
- becomes treason.
So let us imagine ourselves hovering over any great research university at any time in
the last fifteen years, looking down and observing as a few hundred faculty members in
the humanities and social sciences go about their work. Their academic fields are
becoming more politically homogeneous and the attitudes of everyone around them is
becoming gradually more hostile toward conservatives and Republicans. A large minority
of these professors directly study politically charged topics such as race, gender, power,
and marginalization. How might their scholarship be influenced by the trends shown in
figures 2 and 3?
Tetlock (2002) summarized a great deal of research on social cognition by saying
that people sometimes become “intuitive prosecutors,” with a “prosecutorial mindset” that
can get switched on and off as needed. He said that we often strive to be fair, as
prosecutors, but he noted how “blatantly biased ingroups are toward outgroups” (p. 461).
If we allow that many academics studying politically charged topics might, at least
sometimes, be in “intuitive prosecutor” mode, then we can expect them to make three
kinds of conceptual moves that would be helpful in prosecuting the perceived enemies of
social justice (i.e., conservatives, and members of “privileged” groups).

Electronic copy available at: https://ssrn.com/abstract=2702523

 7

1) Maximize the victim class. A good prosecutor will strive to recruit ever more
groups to register complaints against the accused. This might explain the ever-lengthening
list of groups and identities that fall under the protection of diversity and inclusion
policies. This is a form of horizontal creep in the concept of victimhood.
2) Maximize the damages. A prosecutor has a stronger case if she can show that the
damages done to victims are far graver than they appear at first sight. Thus scholars from
across the academy have an incentive to find new ways in which members of allegedly
victimized groups are harmed by current practices. This prosecutorial imperative might
explain the creeping concepts of “marginalization” and “exclusion” (horizontal creep), as
well as the constant lowering of the criteria (vertical creep) for harm in general and trauma
in particular that Haslam described in his essay.
3) Minimize the defendant’s defenses. In criminal cases, most serious charges require
mens rea -- a “guilty mind.” You can’t convict someone of murder or assault if the harm
was entirely unintentional. But if you develop a new legal theory that removes the need
for mens rea, you can vastly increase your conviction rate. This is one of the central
innovations of microaggression theory. Microaggressions are defined as “brief and
commonplace daily verbal, behavioral, or environmental indignities, whether intentional
or unintentional, that communicate hostile, derogatory, or negative racial slights and
insults toward people of color. (Sue et al., 2007, p. 271, emphasis added). Indeed, Haslam
specifically notes that many creeping concepts, including abuse, bullying, and
discrimination, have shed their older requirements for mens rea, requiring only a
subjective assessment by the apparent victim. Unlike courts of law, in academic settings a
lack of intent -- or even the presence of good intentions -- is no longer a valid defense
against charges of racism, sexism, or other crimes. All that matters is that a member of a
protected group felt marginalized. This is why we are increasingly hearing from left-
leaning professors who write essays with titles such as “I'm a liberal professor, and my
liberal students terrify me” (Schlosser, 2015)
To sum up, if an increasingly left-leaning academy is staffed by people who are
increasingly hostile to conservatives, then we can expect that their concepts will shift, via
motivated scholarship, in ways that will help them and their allies (e.g., university
administrators) to prosecute and condemn conservatives. We can expect academic
concepts to “creep” in ways that increase the number of victims and the damages those
victims suffer, and in ways that make it ever harder for anyone to defend themselves
against ugly moral charges. Such politically motivated scholarship may sometimes
originate in humanities departments rather than in psychology, but it draws heavily on
psychological concepts and research, and it feeds back into the six streams of creeping
psychological research that Haslam (2016) reviewed.

Rising Danger for Psychology

Electronic copy available at: https://ssrn.com/abstract=2702523

 8

Haslam lists a number of dangers to psychology from left-ward concept creep. I’ll
just try to make Haslam’s warning more vivid by asking readers to imagine a situation in
which the politics is reversed (see Tetlock, 2005, on turnabout tests).
Suppose that the Federal Bureau of Investigation (FBI) was traditionally a right-
leaning organization, like most law enforcement organizations. Suppose conservatives
outnumbered liberals by about three to one from its founding in 1908 through the 1990s.
But suppose that during the administration of George W. Bush the agency began to lean
much further to the right. After the 9/11 attacks, the agency’s culture became extremely
hostile to liberals and Democrats, who were widely associated with the gravest threats to
the nation. By 2012 the ratio of conservatives to liberals was fourteen to one (as in Figure
3). Do you suppose this transformation might affect the way the FBI did its job, or would
you trust the agency’s professionalism to keep politics out of law enforcement? Might the
agency shift its resources toward conservative priorities, such as fighting terrorism and
moral decay, while ignoring liberal priorities such as abortion clinic bombings, civil rights
infringements, and environmental crimes? And might we begin to see law enforcement
concepts creeping to the right, such that more and more citizens fall under suspicion of
entitlement cheating, abetting illegal immigration, or subverting American values?
Perhaps we’d even see the creation of brand new legal concepts such as “micro-treasons,”
defined as “brief and commonplace daily verbal, behavioral, or environmental indignities,
whether intentional or unintentional, that communicate hostile, derogatory, or negative
attitudes toward the United States of America.”
Most importantly: how would you feel about this new FBI? Would you trust it and
believe its pronouncements? Would you support its continued operational independence,
or would you want Congress to demand reform?
Psychology (like almost all of the other social sciences and humanities) has a serious
problem. Haslam (2016) has shown us that our concepts are creeping to the left in ways
that make psychology ever more appealing to the left and ever less appealing to the right. I
have shown that our membership is creeping to the left as well, as part of a broad national
trend of rising affective polarization. Will we just sit back and let our field drift, or will
we take charge and try to reverse these trends?
In Duarte et al. (2015) we offer a variety of steps that psychology can take. The
essential first step is to examine why psychology is so committed to diversity. We have a
highly refined set of concepts for describing the benefits of diversity. We have argued
since the 1980s that all institutions, organizations, and industries should embrace
diversity, not in order to right past wrongs, but to make themselves better. We should
embrace our own teachings. It would take only small changes to our policies and our
thinking to recognize that the kind of diversity psychology (and the rest of the academy)
most desperately needs is diversity of viewpoints.
(For more on the benefits of viewpoint diversity, please visit
www.HeterodoxAcademy.org)

Electronic copy available at: https://ssrn.com/abstract=2702523

 9

References
Duarte, J. L., Crawford, J. T., Stern, S., Haidt, J., Jussim, L., & Tetlock, P. E. (2015).
Political diversity will improve social psychological science. Behavioral and Brain
Sciences, 38, 1-13.
Haidt, J., & Abrams, S. (2015). The top 10 reasons American politics are so broken.
Washington Post Wonk Blog. Retrieved from
http://www.washingtonpost.com/news/wonkblog/wp/2015/01/07/the-top-10-
reasons-american-politics-are-worse-than-ever/
Haidt, J. (2011). Discrimination hurts real people. YourMorals Blog. Retrieved from
http://www.yourmorals.org/blog/2011/02/discrimination-hurts-real-people/
Haidt, Jonathan. (2012). The righteous mind : why good people are divided by politics and
religion (1st ed.). New York: Pantheon Books.
Haidt, J. (2015). President Obama endorses heterodox academy. Blog post retrieved
10/30/15 from http://heterodoxacademy.org/2015/09/18/obama-endorses-
heterodox-academy/
Haslam, N. (2016). Concept creep: Psychology's expanding concepts of harm and
pathology. Psychological Inquiry.
Inbar, Y., & Lammers, J. (2012). Political diversity in social and personality psychology.
Perspectives on Psychological Science, 7, 496-503.
Klein, D. B. & Stern, C. (2009) By the numbers: The ideological profile of professors. In:
The politically correct university: Problems, scope, and reforms, ed. R. Maranto, R.
E. Redding & F. M. Hess, pp. 15–33. AEI Press.
Lukianoff, G., & Haidt, J. (2015). The coddling of the American Mind. The Atlantic.
Manne, K. (2015) Why I use trigger warnings. New York Times, 9/19/15, Op-ed, p. SR5.
McCarty, N., Poole, K. T., & Rosenthal, H. (2006). Polarized America. Cambridge, MA:
MIT Press.
McClintock, C. G., Spaulding, C. B. & Turner, H. A. (1965) Political orientation of
academically affiliated psychologists. American Psychologist 20:211–21.
Schlosser, E. (2015) I’m a liberal professor, and my liberal students terrify me. Vox.com,
retrieved on 10/30/15 from http://www.vox.com/2015/6/3/8706323/college-
professor-afraid
Tetlock, P. E. (2002). Social functionalist frameworks for judgment and choice: Intuitive
politicians, theologians, and prosecutors. Psychological Review, 109, 451-457.
Tetlock, P. E. (2005). Expert political judgment. Princeton NJ: Princeton University Press.

Electronic copy available at: https://ssrn.com/abstract=2702523