ArticlePDF AvailableLiterature Review

Social Functionalist Frameworks for Judgment and Choice: Intuitive Politicians, Theologians, And Prosecutors

Authors:

Abstract

Research on judgment and choice has been dominated by functionalist assumptions that depict people as either intuitive scientists animated by epistemic goals or intuitive economists animated by utilitarian ones. This article identifies 3 alternative social functionalist starting points for inquiry: people as pragmatic politicians trying to cope with accountability demands from key constituencies in their lives, principled theologians trying to protect sacred values from secular encroachments, and prudent prosecutors trying to enforce social norms. Each functionalist framework stimulates middle-range theories that specify (a) cognitive-affective-behavioral strategies of coping with adaptive challenges and (b) the implications of these coping strategies for identifying empirical and normative boundary conditions on judgmental tendencies classified as errors or biases within the dominant research programs.
Social Functionalist Frameworks for Judgment and Choice:
Intuitive Politicians, Theologians, and Prosecutors
Philip E. Tetlock
The Ohio State University
Research on judgment and choice has been dominated by functionalist assumptions that depict people as
either intuitive scientists animated by epistemic goals or intuitive economists animated by utilitarian
ones. This article identifies 3 alternative social functionalist starting points for inquiry: people as
pragmatic politicians trying to cope with accountability demands from key constituencies in their lives,
principled theologians trying to protect sacred values from secular encroachments, and prudent prose-
cutors trying to enforce social norms. Each functionalist framework stimulates middle-range theories that
specify (a) cognitive–affective–behavioral strategies of coping with adaptive challenges and (b) the
implications of these coping strategies for identifying empirical and normative boundary conditions on
judgmental tendencies classified as errors or biases within the dominant research programs.
Once an esoteric specialty of a small cadre of cognitive psy-
chologists, experimental research on judgment and choice has—to
judge just by citation counts—become psychology’s leading in-
tellectual export to the social sciences as well as to a host of
applied fields. The influence of this research program has spread
(critics might say “metastasized”) into such diverse domains as
public opinion, international relations, finance, organizational be-
havior, marketing, medical diagnosis, and the law (Gilovich, Grif-
fin, & Kahneman, 2002). Scholars with little else in common share
a familiarity with foundational concepts in behavioral decision
theory such as heuristics and biases, framing and choice, and the
psychophysics of gain and loss functions.
The current article is, in one sense, testimony to the profound
success of the heuristics-and-biases research program in extending
cognitivist concepts into a vast array of disciplines. By advancing
a succession of ingenious demonstrations, investigators faithful to
the core tenets of the program have converted many skeptics,
although not all (Gigerenzer, Todd, & the ABC Research Group,
1999), to the view that limitations of humans as information
processors, especially the widespread reliance on inferential short-
cuts, produce systematic and often surprisingly difficult-to-correct
deviations from rationality ranging from base-rate neglect to my-
opic loss aversion (Kahneman & Tversky, 2001). The research
program thus advances a core scientific value of psychology: the
reduction of superficial diversity to unifying principles. However,
parsimony is not a trump value. The current article is, in another
sense, a cautionary comment on the perils of success. It proceeds
from the contextualist premise that the more aggressively one
extends abstract principles into new domains, the greater the risk
of overextension (McGuire, 1983).
This article makes the case that the risk of overextension is
substantial. Once placed in historical and philosophical context,
the heuristics-and-biases program is revealed to be but one (albeit
far and away the most successful) of an array of possible func-
tionalist perspectives on judgment and choice.
Key Arguments
The key arguments—each to be expanded later in the article—
include the following points.
Psychology Is Inherently Functionalist
William James (1890/1983) was the first of a long succession of
theorists to observe that psychology is an inherently functionalist
discipline. Even the most forbiddingly formal psychological ex-
planations ultimately rest on functionalist assumptions about the
goals that people try to achieve by thinking, feeling, and acting as
they do.
Influential Functionalist Starting Points
The traditional functionalist starting points for research on judg-
ment and choice have depicted people either as intuitive scientists
(who seek causal understanding and predictive leverage; Kelley,
1971) or intuitive economists (who strive to maximize subjective
utility; Edwards, 1962). There has also been long-standing interest
in hybrid forms of functionalism that permit the ideal-type intui-
tive scientist’s value-neutral pursuit of truth to be deflected by
various directional goals, such as signal-detection-theory concerns
about avoiding one class of inferential error more than another
(Friedrich, 1993) or the pursuit of intrapsychic objectives such as
protecting self-esteem, restoring cognitive consistency, or affirm-
ing belief in a controllable world (cf. Baumeister, 1998; Kruglan-
ski, 1990; Kunda, 1999; Tesser, 2000; Tetlock & Levi, 1982).
The Pivotal Role of Functionalist Metaphors
Functionalist metaphors play the role of “hard-core” assump-
tions in Lakatosian philosophies of social science: They set theo-
Philip E. Tetlock, Department of Psychology, The Ohio State University.
I gratefully acknowledge financial assistance from National Science
Foundation Grant SBR-9696162 and helpful critiques from Hal Arkes and
the social psychology faculty dinner group at The Ohio State University.
Correspondence concerning this article should be addressed to Philip E.
Tetlock, who is now at Haas School of Business, University of California,
Berkeley, California 94720-1900. E-mail: tetlock@haas.berkeley.edu
Psychological Review Copyright 2002 by the American Psychological Association, Inc.
2002, Vol. 109, No. 3, 451–471 0033-295X/02/$5.00 DOI: 10.1037//0033-295X.109.3.451
451
retical priorities and the stage for subsequent inquiry (Lakatos,
1970; Laudan, 1997). Metaphors are the psycholinguistic methods
of choice for playing this role in psychological science because of
their evocative power to capture superordinate features of similar-
ity that cut across superficially diverse spheres of human activity
(cf. Glucksberg & McGlone, 1999). For example, when speakers
say, all lawyers are sharks, they do not mean that lawyers are
fast swimmers or have fins and sharp teeth. They meanand most
people quickly understand them to meanthat lawyers are vi-
cious, predatory, and tenacious. In a similar vein, when investiga-
tors launch research from the premise that people are intuitive
scientists or economists, they do so because they believe it is
valuable to highlight abstract commonalities underlying these
highly organized, indeed, professionalized, forms of human ac-
tivity and the far more loosely organized thoughts, feelings, and
actions of ordinary people. Even if they do not always live up to
the exacting standards of professional scientists, people try to
make causal sense of events, to test hypotheses against reality, and
to make choices that advance their interests. A natural focus for
research becomes the following question: How well do ordinary
people size up against well-defined normative benchmarks of
rationality such as Bayesian-belief updating and expected-utility
maximization?
Theorists need not, however, agree on the answers. Indeed, there
are strong professional incentives to disagree and to articulate
testable theories that, although faithful to the hard-core tenets of
the program, are as distinctive as empirically plausible from con-
ceptual rivals. Over the last 3 decades, the spectrum of portraits of
intuitive scientists has ranged from rigorous Bayesians to glib
top-of-the-head information processors, with an eclectic assort-
ment of dual-process models between these end points (Chaiken &
Trope, 1999). Although functionalist research programs are not,
strictly speaking, falsifiable (there is, in principle, no end to the
conceptual tinkering in which determined defenders can engage to
preserve core premises), these programs are exhaustiblea point
reached when the patience of investigators and professional gate-
keepers collapses.
Social Functions of Thought
The approach advocated here represents a sharp departure from
psychological theories that have traditionally stressed the intrapsy-
chic functions of judgment and choice and that have placed the
isolated Cartesian thinker at the center of inquiry (Gergen, 1986).
The focus shifts to the social functions of thought and to embed-
dedness of human beings in relations with other people, institu-
tions, and the broader political and cultural environment. Building
on the observations of classical sociologistsDurkheim, Weber,
Mead, Parsonsas well as on more recent advances in cross-
cultural psychology, the analytical starting point becomes the
psychological prerequisites of social order, the properties that
people must possess to cope with the challenges of life within
intricately interdependent collectivities in which they alternate,
often in rapid succession, between being targets of the account-
ability demands from others and being the sources of accountabil-
ity demands on others (cf. Aberle, Cohen, Davis, Levy, & Sutton,
1951).
Working within this overarching framework, the current article
proposes three socialfunctional theories especially relevant to
psychological debates on human rationality and to current classi-
fications of error and bias in judgment and choice. Each theory is
designated by a guiding metaphor that captures the essence of a
particular functional orientation that the vast majority of people
can, under the right activating conditions, adopt toward the
social world. As objects of accountability pressures from others,
people strivelike intuitive politiciansto maintain positive so-
cial identities vis-a`-vis significant constituencies in their lives. A
key function of thought becomes internalized dialogue in which
people try to anticipate objections to possible lines of action and to
craft responses. As transmitters of accountability pressures onto
others, people trylike intuitive prosecutorsto detect cheaters
and free riders who seek the benefits but shirk the responsibilities
of membership in the collective. A key function of thought be-
comes closing loopholes in accountability regimes that unscrupu-
lous intuitive politicians might otherwise exploit. Finally, as be-
ings capable of reflecting on the accountability regimes within
which they live, love, and work, people are posited to be intuitive
theologians who have an existential need to believe that the rules
governing their social world are not just the arbitrary preferences
of currently dominant interest groups but rather are anchored in
sacred values that confer legitimacy on collective practices. A key
function of thought becomes protecting sacred values from secular
encroachments.
Intuitive Politicians, Theologians, and Prosecutors
Just as it is possible to advance an array of psychological models
of how people function as intuitive scientists or economists, the
same can be said for the intuitive politician, theologian, and
prosecutor. It is possible to posit intuitive politicians of varying
skill in anticipating the objections of key constituencies, intuitive
theologians of varying forgiveness toward secular encroachments
on sacred values, and intuitive prosecutors of varying punitiveness
toward norm violators. The mere possibility of doing something
does not, however, demonstrate the scientific desirability of doing
it. The scientific community confronts a tricky effort-allocation
decision in which it must weigh the risks of excessive closed-
mindednessof failing to diversify its portfolio of guiding as-
sumptions and of thereby neglecting key aspects of human nature
that traditional assumptions slightagainst the risks of excessive
open-mindednessof aiding and abetting the proliferation of friv-
olous frameworks that divert scarce scientific talent from proven to
unproven approaches.
Payoffs of Diversification
Advocates of diversifying the disciplines portfolio of function-
alist frameworks need to reassure skeptics that diversification
yields tangible empirical and conceptual payoffs. Functionalist
frameworks need to move expeditiously from vague metaphorical
pontificating to testable process models that specify the adaptive
challenges that activate functionalist mind-sets, the goals linked to
these mind-sets, and the coping strategies used to achieve these
goals. The resulting social functionalist theories should identify
empirical boundary conditions on judgmental tendencies classified
as errors or biases within the dominant epistemic or utilitarian
research programsthat is, conditions under which judgmental
shortcomings are attenuated or amplified by social contextual
452
TETLOCK
manipulations that create adaptive challenges that, in turn, activate
sociocognitive processing goals and coping strategies. The result-
ing social functionalist theories should also identify normative
boundary conditions that specify when it is reasonable to challenge
extant classifications of judgmental tendencies as errors or biases.
Social functionalism raises the possibility of normative Gestalt
shifts: Effects that look dysfunctional within an intuitive-scientist
or intuitive-economist framework sometimes look quite adaptive
within a political, prosecutorial, or theological framework.
The remainder of this article develops the preceding arguments.
The next section directly addresses skeptics who worry that the
approach recommended here is too permissive of runaway func-
tionalist speculation. It lays out the overarching themes that unify
and impose common standards on the three social functionalist
theories. The following three sections are organized around the
politician, theologian, and prosecutor programs. Each section be-
gins by characterizing the functionalist premises of the incipient
research program: the core goals, adaptive challenges, and coping
strategies attributed to human beings. Each section then delineates
a conceptual range of testable theories consistent with these pre-
mises, advances a specific theory that falls within the programs
tolerance zone of opinion, derives testable hypotheses, and eval-
uates relevant evidence. The concluding section addresses influ-
ential objections to social functionalism and stresses the need for
a distinctive style of psychological theorizing pitched at a societal
level of analysis that links intrapsychic processes explicitly to the
political and institutional contexts within which people live and
work.
Conceptual Guidelines for Constructing
Social Functionalist Theories
Screening Tests
Skeptics have long suspected that functionalism is inherently
vague and tautological (Boring, 1963). If a thought, feeling, or
action appears maladaptive within one functionalist framework,
one can always concoct an alternative functionalist rationale.
Without denying the kernel of truth to this characterization of
earlier forays into functionalism, the current article implements
several checks against undisciplined functionalist speculation. So-
cial functionalist research programs must initially pass three
screening tests and should eventually be held to account for
satisfying a fourth:
1. The program should be organized around a postulated func-
tion that highlights a fundamental adaptive challenge that arises
whenever human beings are locked together in complex patterns of
interdependence;
2. The program should stimulate testable theories that provide
reasonably specific answers to questions concerning the adaptive
challenges that activate functionalist mind-sets, the goals that
people are trying to achieve, and the coping strategies that people
use in pursuit of the goals;
3. The testable theories should stimulate discoveries of empir-
ical and normative boundary conditions that would have been
unlikely had theorizing been confined to the dominant functional-
ist frameworks;
4. The testable theories should ultimately address the outstand-
ing problem of integrating the psychosocial functions of judgment
and choice by specifying how people resolve cross-functional
conflicts.
Adaptive Challenges
The social functionalist theories advanced here depict human
beings whose thought processes are organized around three adap-
tive imperatives of collective life. First is the necessity of coping
with accountability demands from others with whom one is inter-
twined in reciprocal networks of interdependence. Each person is
embedded in a matrix of accountability relationships that specify
who must answer to whom, for what, and under what ground rules.
To survive and prosper in any social unit, people (in their capacity
as intuitive politicians) must possess a reasonably reliable mental
compass for navigating the self through these sometimes Byzan-
tine rolerule structures (Semin & Manstead, 1983; Stryker &
Stratham, 1985).
The second imperative is social control, the need to place
accountability demands on others who might be tempted to derive
the benefits of collective interdependence without contributing
their fair share or without respecting other aspects of the prevailing
rolerule regime (Cosmides & Tooby, 1994). To survive and
prosper in any social unit, intuitive prosecutors need to protect
themselves from exploitation but must be careful to avoid over-
zealousness in the pursuit of wrongdoers by respecting culture-
specific norms for defining the boundaries of self and the concept
of harm (Scott & Lyman, 1968).
The third imperative is legitimation of the accountability ground
rules. People need moral backstops (M. Lerner & Lerner, 1981).
To stave off anomie, alienation, and even existential despair,
intuitive theologians need to believe that the prevailing account-
ability and social control regime is not arbitrary but rather flows
naturally from an authority that transcends accidents of history or
whims of dominant groups (Durkheim, 1925/1976). Framed in this
fashion, the social functionalist research programs are an integral
part of a societal social psychology that must ultimately be con-
sistent with, but is not reducible to, the intrapsychic levels of
analysis at which most theory is now pitched. The theories can thus
be viewed as distinctive members of an already large family of
identity-maintenance and social control formulations in the liter-
ature (cf. Haidt, 2001; Hamilton, 1980; Schlenker, 1985; Steele,
1988; Tesser, 2000).
Adequate Specification of Hypothesized Functions
The social functionalist theories also fit a common explanatory
template. Each specifies switch-on rulesthe types of adaptive
challenges that activate the metaphorical mind-set as well as the
types of goals and coping strategies that become increasingly
cognitively accessible, and are seen as increasingly applicable,
under conditions of mind-set activation. Each also specifies
switch-off rulesthe types of negative feedback from the envi-
ronment that deactivate the hypothetical mind-sets and their ac-
companying goals and coping strategies.
The self-regulatory innards of social functionalist theories owe
an obvious debt to ideas drawn from the cybernetic control of
complex systems (Higgins, 1989). Each theory presupposes an
autonomous self-regulatory system that is activated by adaptive
challenges and that relies on discrepancy-reducing negative feed-
453
FUNCTIONALIST PLURALISM
back mechanisms designed to minimize the gap between current
condition and desired end state.
The adaptive challenges for activating the intuitive-politician
mind-set take the form of accountability pressures and demands:
the knowledge that one is under the evaluative scrutiny of impor-
tant constituencies in ones life who control valuable resources and
who have some legitimate right to inquire into the reasons behind
ones opinions or decisions. This knowledge activates the goal of
establishing or preserving a desired social identity vis-a`-vis these
constituencies. In pursuit of this goal, people may resort to a wide
range of coping strategiesfrom attitude shifting to preemptive
self-criticism to defensive bolsteringthat shut off only when
people believe that they have achieved the desired end state or that
the costs of pursuing that end state have become prohibitive.
The adaptive challenges for activating the intuitive-theologian
mind-set take the form of perceived threats to sacred values, values
thatby community consensusare deemed beyond quantifica-
tion or fungibility. One common type of threat arises whenever
people believe that they or others have subjected sacred values to
the same crass mental operations routinely performed on secular
values (e.g., computing the dollar value of ones family or friend-
ships or loyalty to a collective). Taboo violations of this sort
activate the goal of reaffirming individual and group commitment
to the sacred value. In pursuit of this goal, people may resort to a
mixture of coping strategiesincluding moral outrage and moral
cleansingthat shut off only when people are reassured that the
secular encroachment has been rebuffed or when people are con-
vinced that the costs of maintaining the social illusion of infinite
importance are prohibitive.
The adaptive challenge for the intuitive prosecutor is the per-
ception that norm violation is both common and commonly goes
unpunished. This perception activates the goal of tightening stan-
dards of accountability as well as the coping strategies of looking
for grounds for rejecting excuses and justifications for norm vio-
lation and for ways of closing loopholes. The process disengages
only when people believe that the collectivity is no longer under
siege or when people sense that the pursuit of social order has gone
too far and has become oppressive.
Discovery of Empirical and Normative Boundary
Conditions on the Error-and-Bias Portrait of Human
Nature
From the standpoint of cognitive theories of judgment and
choice, all this effort to specify the self-regulatory dynamics of
social functions is of little interest if the resulting theories do not
generate testable predictions that shed light on either empirical or
normative boundary conditions on the error-and-bias portrait of
human nature. Accordingly, the social functionalist theories ad-
vanced here are evaluated by a common conceptual standard: their
capacity to shed light on social contextual conditions under which
one should expect to observe the amplification or attenuation of
effects labeled as biases or should question the very classification
of effects as biases. For instance, the social contingency model of
how people function as intuitive politicians identifies not only
distinctive strategies of coping with accountability but also the
implications of particular coping strategies (such as preemptive
self-criticism) for when the amplification of biases (rooted in over-
reliance on simple, easy-to-execute, heuristics) or the attenuation
of biases (rooted in fear of being second guessed) should be
expected. The sacred-value-protection model of how people func-
tion as intuitive theologians not only identifies moral outrage and
cleansing as coping responses to violations of sacred values but
also examines the implications of those coping strategies for
trade-off aversion and base-rate neglect. The fair-but-biased and
occasionally erratic model of the intuitive prosecutor not only
identifies strategies of coping with social deviance but also ex-
plores the implications of these strategies for debates over the
normative status of the fundamental attribution error and the
severity bias.
The Challenges of Modeling Cross-Functional Conflicts
There is already considerable evidence consistent with key
empirical-and-normative-boundary-condition predictions of each
social functionalist theory. There is much less evidence, however,
for guidance in the next phase of theory development: the integra-
tion of the social and cognitive functions of judgment and choice.
Little is known about when the pursuit of one self-regulatory
objective has negative externalities that impede attainment of other
objectives. In political theory, negative externalities offer the clas-
sic justification for empowering government to regulate the con-
duct of its citizens (to paraphrase John Stuart Mill, your freedom
to swing your arm ends where my nose begins). Likewise, in
psychological theory, negative intrapsychic externalities have of-
fered the classic justificationsince Freudfor some system of
mental governance or executive control, to arbitrate conflicts be-
tween modularized functions.
One common form of motivational conflict arises when social
functions circumscribe cognitive and material-utilitarian functions.
The scientists quest for the truth is subordinated to the politicians
need to craft compelling justifications; the economists calculus of
self-interest is checked by the theologians obligations to commu-
nally shared values; the scientists curiosity about the causes of
behavior is qualified by both the prosecutors determination to
hold rule breakers responsible and the theologians desire to block
inquiry that demystifies objects of veneration. The flip side forms
of conflict arise when social functions are circumscribed by cog-
nitive and material-utilitarian functions. The intuitive theologians
effort to treat certain values as beyond compromise runs aground
the intuitive economists recognition of resource constraints. The
intuitive politicians efforts to escape accountability, or the intui-
tive prosecutors efforts to impose accountability, collide with the
intuitive scientists recognition of the implausibility of certain
claims. The closing section of the article returns to these subtle
issues of mental balancing.
Intuitive-Politician Research Program
People resemble intuitive politicians in that they are accountable
to a variety of constituencies, they suffer consequences when they
fail to create desired impressions on key constituencies, and their
long-term success at managing impressions hinges on their skill at
anticipating objections that others are likely to raise to alternative
courses of action and at crafting accounts that preempt those
objections. Moreover, just as there is room for diversity in how
theorists characterize people as intuitive scientists or economists,
there is room in the intuitive-politician research program for dif-
454
TETLOCK
ferent approaches to developing that metaphor. Some theorists
may opt to portray people as ruthless Machiavellian schemers
(who will do whatever it takes to win), whereas others might
portray people as bumbling impression managers (who frequently
miscalculate social impact because of cognitive biases rooted in
heuristics [Gilovich, Medvec, & Savitsky, 2000] or in transference
[Andersen & Glassman, 1996]).
A Testable Model of How People Function as Intuitive
Politicians
The social contingency model (SCM) of judgment and choice
falls toward the midpoint of the Machiavellianism continuum
(Tetlock, 1992, 1998, 1999; Tetlock & Lerner, 1999). It assumes
that people seek the approval and respect of the key constituencies
in their lives. But the model eschews a purely strategic conception
of human nature. The SCM can be summarized in seven proposi-
tions, described below.
The universality of accountability. People do some things
alone, but it is difficult to escape evaluative scrutiny in an intri-
cately interdependent society. Escape arguably becomes impossi-
ble when self-accountabilitythe obligations that most human
beings, excluding psychopaths, feel to internalized representations
of significant others who keep watch over them when no one else
is lookingis taken into account (Mead, 1934). The SCM adopts
the symbolic-interactionist tenet that accountability is a ubiquitous
feature of everyday life that links individuals to institutions by
reminding people of the need to be cognitively equipped with good
reasons for their opinions in the event that observers request or
demand accounts.
The audience-approval motive. Whether it is a function of
natural and sexual selection or of relentless social reinforcement
from infancy onward, human beings have a deep-rooted need for
social approval (Baumeister & Leary, 1995). Moreover, people are
generally risk averse and more concerned with falling in the
esteem of others than with missing opportunities to rise. Failure to
create desired impressions on a target audience triggers an aversive
state of emotional arousalembarrassment or shamethat people
try to reduce by means of a host of impression-management tactics
(cf. Keltner & Buswell, 1997; Schlenker, 1985).
Motive competition. Although the SCM assigns a central role
to the approval motive, it does not elevate that motive to sovereign
regulator of all conduct. Social psychology has had too many
disappointing flirtations with monistic theories that anointed mas-
ter motives (Allport, 1968/1985), and there is no point in replicat-
ing the mistakes of early role theorists who treated approval
maximization as the sociological equivalent of utility maximiza-
tion. The SCM identifies four classes of motives that circumscribe
the approval motive, including the following:
1. Achieving cognitive mastery of the causal structure of the
environment (emphasized by classic attribution theory, which por-
trayed people as cognitively diligent intuitive scientists; Kelley,
1971);
2. Minimizing mental effort and achieving rapid closure (em-
phasized by later theories of social cognition, which portrayed
people as lazy or cognitively miserly intuitive scientists; Kruglan-
ski & Webster, 1996);
3. Maximizing the benefits and minimizing the costs of rela-
tionships (emphasized by evolutionary and exchange theories;
Kelley & Thibaut, 1978; Ridley, 1994);
4. Asserting personal autonomy and integrity by reaffirming
private convictions (emphasized by theories of ego and moral
development, by self-affirmation variants of dissonance and social
identity theory, and by reactance theory; Schlenker, 1985).
Generative rules for predicting coping strategies. A laundry
list of motives does not a theory make. It is necessary to specify
how motives can be amplified or attenuated by the interpersonal
context. The conceptual formula for generating predictions fits the
classic template for expectancy-value theories in psychology:
Identify situational and dispositional moderators of the perceived
importance of identity objectives and the perceived feasibility of
achieving those objectives. The SCM posits that the likelihood of
coping strategies should wax or wane as a multiplicative function
of independent variables that influence (a) the cognitive accessi-
bility and perceived applicability of dimensions of social identity
and (b) the perceived feasibility, costs, and benefits of claiming a
given identity.
Coping-strategy predictions. Every request for justification
raises the question of how it will be received by either external
constituencies (that one sees as separate from ones self) or inter-
nalized ones (familiar voices inside the head that one accepts as
integral parts of oneself). In each case, the attributions that ob-
servers make define ones reputation: If I do or say X, will others
view me as cooperative or confrontational, craven or candid,
flexible or rigid, principled or opportunistic . . .? The SCM distin-
guishes an array of identity-defining choices that arise in account-
ability predicaments, but two stand out as most relevant to theories
of judgment and choice: preemptive self-criticism and defensive
bolstering. Preemptive self-criticism is the predecisional strategy
of anticipating plausible objections of would-be critics, factoring
those objections into ones mental representation of the problem,
and reaching a complex synthesis that specifies how to deal with
trade-offs. The SCM asserts that people will perform this cognitive
work only when low-effort strategies have been systematically cut
off by the logic of the situation. Thus, preemptive self-criticism
should be especially likely when people are unconstrained by past
commitments, the views of the audience are unknown or known to
conflict, the audience is knowledgeable and powerful, and the
audience possesses a legitimate right to inquire into the reasons
behind opinions or decisions (J. S. Lerner & Tetlock, 1999).
Defensive positioning is the postdecisional mirror image of pre-
emptive self-criticism. Accountability still motivates thought, but
that thought now takes self-justifying rather than self-critical
forms. People devote mental energy to generating reasons why
they are right and would-be critics wrong. Self-justification should
be most pronounced when people are accountable for prior public
commitments, when people fear that previous decisions cast doubt
on their competence or morality, and when people know that they
cannot plausibly deny responsibility but suspect that they can put
a more positive spin on earlier decisions (Tetlock, Skitka, &
Boettger, 1989).
Empirical boundary conditions on the error-and-bias portrait of
human nature. Skeptics could concede the previous predictions
but still insist that none bears on the foveal concerns of the
traditional research programs: Namely, when and why do judg-
ment and choice deviate from economic and scientific standards of
455
FUNCTIONALIST PLURALISM
rationality? Skeptics could just dust off the disciplinary division
of labor between psychology and the social sciences proposed
by N. E. Miller and Dollard (1941): The mission of cognitive
research programs is to shed light on how people think, whereas
the politician research program sheds light merely on what
people think about and on when they will say what is on their
minds.
To be sure, certain strategies of coping with accountability, such
as attitude shifting, can be socially significant but cognitively
trivial. The SCM implies, however, that the time-honored process
content division of labor crumbles in many settings. Self-
presentation triggered by accountability can interact in complex
ways with putatively basic cognitive processes. A central function
of private thought is preparation for public performancesa prop-
osition that echoes early 20th-century theorists such as Vygotsky
(1934/1978) and Mead (1934). Thought frequently takes the form
of internalized dialogues in which people gauge the justifiability of
options by imagining conversations in which accounts are ex-
changed, debated, revised, and evaluated. Whether these imagi-
nary conversations amplify or attenuate judgmental tendencies
deemed by the expert community to be biases depends on both the
substantive content of the internal dialogues and the cognitive
affective sources of the alleged biases.
This, the sixth proposition of SCM, asserts that the effect of
accountability hinges on its power to prime metacognitive ideals
that people in various subcultures have, to varying degrees, inter-
nalizedideals such as be self-critical,”“attend to all conceiv-
ably relevant information,”“stay the course, or dont cry over
spilt milk. Certain prescriptive ideals direct people to think in
ways that, given the problem at hand, correct judgmental tenden-
cies widely viewed in the expert community as biases or errors.
For example, forms of accountability that activate the metacogni-
tive maxim anticipate plausible objections should attenuate bi-
ases that are rooted in overreliance on simple, easy-to-execute
heuristics. Following this maxim should also, however, amplify
biases rooted in indecisiveness, risk aversion, or fear of being
second guessed. Conversely, forms of accountability that activate
the flip-side maximsstay the course and defend yourself
should have mirror-image effects, amplifying biases such as sunk-
cost justification but attenuating biases linked to premature aban-
donment of sound policies (such as myopic loss aversion; Benartzi
& Thaler, 1995). Finally, for certain inferential problems, people
may either have no pertinent prescriptive ideals or contradictory
ones. Here the SCM predicts null effects.
Normative boundary conditions on the error-and-bias portrait.
Inherent in the politician metaphor is the notion of contending
factions with opposing conceptions of rationality and morality.
The SCM is cognitively egalitarian: It treats academic commen-
tators on rationality as but one of a number of constituencies that
have staked out positions on how people should think, feel, and act
in particular settings. Disagreements will arise over the classifica-
tion of judgmental tendencies as errors or biases whenever observ-
ers hold conflicting beliefs about either (a) the degree to which
judgmental tendencies that deviate from scientific or economic
standards of rationality have offsetting social or political benefits
or (b) the usefulness of judgmental tendencies in facilitating at-
tainment of agreed-on goals.
Evidence Bearing on Key Predictions of the Model
When does the politicalprocess of computing the justifiability
of response options begin to impinge on the putatively more basic
cognitive processes of maintaining a stable, internally consistent
world view at acceptable cost in mental effort? Experimental
evidence bears out the fifth proposition of the SCM (see Empirical
boundary conditions on the error-and-bias portrait of human
nature above), that the goals of minimizing cognitive effort, dis-
sonance, and ambiguity clash most directly with the political goals
of social identity protection when decision makers confront ac-
countability demands for which they do not have a ready-at-hand
or dominant responsethat is, when decision makers have not
made any prior commitments to a position and are accountable
either to a single audience with unknown views (Tetlock, 1983a)
or to multiple audiences who hold contradictory views and who
possess strong arguments in defense of their positions (Green,
Visser, & Tetlock, 2000). In such cases, people respond by engag-
ing in more reflective and self-critical patterns of thinking in which
they attempt to anticipate potential objections of reasonable critics
and to construct defensible integrative positions (Tetlock, 1992).
Content analysis of thought protocols reveals not only that people
generate more thoughts but also that thinking tends to take a
dialectical on the one hand and on the other hand rhythm, as
though people were sending the implicit message: I may believe
X, but Im no fool, and I recognize counterarguments Y and Z.
Cognitive effort directed to protecting ones social identity as
rational need not, however, always promote open-mindedness.
Postdecisional accountability often reminds people of metacogni-
tive maxims such as stay the course and encourages self-
justifying patterns of thinking that have been linked to amplifica-
tion of dissonance effects such as escalating commitment to sunk
costs and trade-off avoidance (Staw, 1980; Tetlock et al., 1989).
The sixth proposition of the SCM posits that whether academic
observers applaud or deplore the impact of accountability will
hinge on the correspondence between the informal theories that
ordinary people hold about good decision making (the metacog-
nitive maxims primed by accountability and problem context) and
the more formal theories that academic observers hold about
rationality. The SCM predicts that accountability of the right sort
(which motivates people to make judgments they deem most
justifiable on careful reflection) will attenuate bias when there is
convergence between formal and informal theories. The best can-
didates for de-biasing, therefore, should be response tendencies
that satisfy two criteria. First, unaccountable decision makers are
making suboptimal judgments because they have not been moti-
vated to think sufficiently carefully and self-critically about their
own mental processes. Second, decision makers intuit that the
normative expectation is to engage in sober second thought when
a neutral authoritative figure in an academic setting requests a
justification before they have taken any position. Under such
conditions, people should bring their judgments into closer corre-
spondence with those of the expert community, the academic
arbiters of normative standards. Examining the evidence, J. S.
Lerner and Tetlock (1999) find considerable support for this hy-
pothesis. Accountability that encourages preemptive self-criticism
tends to have the following effects.
1. It reduces the overattribution effect (Tetlock, 1985). Self-
critical thinkers are more reticent about drawing dispositional
456
TETLOCK
attributions from conduct constrained by context in the low-
choice conditions of essay-attribution experiments, but they are
no more reticent about drawing such conclusions in the high-
choice conditions.
2. It reduces primacy and recency effects (Kennedy, 1993;
Tetlock, 1983b). Self-critical thinkers are more cautious about
drawing conclusions from incomplete evidence and more willing
to change their minds in response to evidence.
3. It improves the calibration of the subjective probabilities that
people attach to their predictions (Siegel-Jacobs & Yates, 1996;
Tetlock and Kim, 1987). Moreover, these improvements in cali-
bration can be achieved at minimal cost in resolution (the
variance of proportion of correct predictions across confidence
categories)evidence that people are not indiscriminately scaling
down all confidence estimates but rather are carefully weighing
evidence.
4. It reduces the incompatibility biasthe tendency for negoti-
ators to assume incorrectly that relationships are zero sum and that
the other partys interests are completely opposed to their own (L.
Thompson, 1995).
5. It reduces numerical-anchoring effects by encouraging sub-
jects to consider additional evidence and to revise estimates. This
effect holds up as long as subjects are not under time pressure or
cognitive load (Kruglanski & Freund, 1983).
6. It decreases reliance on noncompensatory decision rules in
favor of compensatory ones that explicitly acknowledge trade-offs
(Kahn & Baron, 1995; Tetlock, 1983a).
7. It decreases the influence of incidental affect evoked in one
situation on judgments made in a completely unrelated situation
(J. S. Lerner, Goldberg, & Tetlock, 1998).
8. It decreases the influence of sunk costs on future investment
decisions by increasing adherence to self-imposed earlier limits on
amounts to be invested and by increasing the salience of decision-
making procedures (as opposed to outcomes) in evaluation (Si-
monson & Staw, 1992).
9. It reduces illusory correlations and improves covariation
assessment, in part, by motivating more complex inferential strat-
egies (Murphy, 1994).
10. It increases correspondence between the judgment strategies
that people claim to be using and the strategies that statistical
models impute to them (Hagafors & Brehmer, 1983).
The SCM predicts bias amplification whenever the implicit
theories of good judgment held by research participants diverge
from those held by the researchers. Here, efforts to put on the
cognitive equivalent of ones Sunday best”—to follow the salient
metacognitive maximslead to judgments that depart ever more
markedly from those of the expert community. The following lines
of evidence are consistent with this hypothesis.
Ambiguity aversion refers to the tendency of decision makers to
prefer less ambiguous alternatives when they are given a choice
between options that differ only in uncertainty about the probabil-
ities of relevant outcomes (Curley, Yates, & Abrams, 1986).
Expected value is held constant, so this preference is, strictly
speaking, a bias from a microeconomic point of view. Thought
protocols suggest that accountability increases preferences for
well-defined probabilities, in part because participants feel that, in
the event of failure, it would be more difficult to justify having
selected the ambiguous options (people are supposed to look
before they leap).
In the compromise effect, a product gains attractiveness simply
by virtue of becoming the middle option in a choice set (Simonson,
1989). Accountable participants were more likely to select the
compromise product because they believed that it was more de-
fensible than options that were clearly superior on one dimension
but clearly inferior on another.
In the attraction effect, a relatively inferior alternative (Brand X)
is added to a set of closely competing options (Brands A and B),
thereby increasing the attractiveness of the preexisting option that
happens to be superior to Brand X on all evaluative dimensions
(Simonson, 1989). Again, accountable subjects were drawn to
dominating options, apparently because they thought that the dom-
inating options were less vulnerable to criticism.
The dilution effect refers to the tendency of observers to lose
confidence in the predictive power of diagnostic evidence when
that evidence is mixed with irrelevant or nondiagnostic evidence
(Tetlock & Boettger, 1989). Motivating self-critical thought can
induce people to try too hard to discern relevance amidst irrele-
vance. Rather than zeroing in on the one relevant cue (e.g., number
of hours studied as a predictor of grade point averages), account-
able subjects often struggle to integrate irrelevant cues into their
schematic representations of the target individuals (vain searches
for indicators of academic success, such as. If hes never dated
anyone longer than 2 months, he may be emotionally unstable,or,
Tennis players are usually physically and mentally alert). In
environments with unfavorable signal-to-noise ratios, accountabil-
ity can send people off on inferential wild-goose chases.
The status quo occupies a privileged position in decision mak-
ing. It is possible to reverse preferences for alternatives by holding
expected value constant and arbitrarily designating one option as
the status quo and the other as change (Samuelson & Zeckhauser,
1988). The changeoption is held to a higher standard of proof
and accountable subjects are particularly prone to do so when
abandoning the status quo creates identifiable losers who are likely
to complain (Tetlock & Boettger, 1994).
In cases where accountability has had no effect, there is pre-
sumptive evidence either that people lack intuitions about how to
solve the problem (the statistically untutored do not know Bayess
theorem) or have conflicting normative intuitions (some of which
coincide with formal models of procedural rationality and some of
which diverge). The proverbial file drawer of null-hypothesis
results contains a sizable set of studies in which accountability has
had no effect on judgmental biases, including insensitivity to
base-rate information, giving more weight to causal as opposed to
merely statistical relationships between variables, preference
reversals as a function of choice-versus-matching elicitation pro-
cedures, insensitivity to sample size, and the conjunction fallacy
(cf. Simonson & Nye, 1992).
Normative Boundary Conditions
Turning to the normative-pluralism postulate of the SCM, there
are both good arguments and suggestive evidence that what looks
like an error or bias from an intuitive-scientist or intuitive-
economist perspective will often look quite defensible to observers
working within an intuitive-politician framework:
Ambiguity aversion. Decision makers may be right that they
will incur greater blame for choosing options with ambiguous
probabilities than those with well-defined probabilities (holding
457
FUNCTIONALIST PLURALISM
expected values constant). They just need to imagine the recrim-
inations from constituents if they choose an ambiguous gamble
(with a probability of winning between 10% and 90%) over a
gamble with a well-defined probability (50/50) and if the ambig-
uous gamble turns out to have had only a 10% chance of positive
pay-off (Taylor, 1995). Unaccountable decision makers will feel
suckered; accountable ones will feel the same way but, in
addition, will feel labeled as suckers by others.
Attraction effect. Simonson (1989) argued that people believe
they have a more persuasive justification for choosing A over B
when A but not B dominates the irrelevant alternative C. Although
introducing C does not alter the offsetting strengths of Options A
and B (and being swayed by the introduction of C puts one in
violation of the rational-choice axiom of the independence of
irrelevant alternatives), invoking the argument A dominates C but
B does not may be a quite compelling justification for many
audiences.
Dilution. Decision makers infer that they will get along more
smoothly with their prospective conversational partners if they
make special efforts to detect relevance in the seemingly irrelevant
information that the partners have bestowed on them. This inter-
pretation (which differs from the representativeness-heuristic ac-
count that portrays people as flawed statisticians) gains credibility
from evidence that accountability amplifies the dilution effect only
when participants believe that the conversational norm of rele-
vance holds and the interaction partner views the information
provided as germane (Tetlock, Peterson, & Lerner, 1996). Indeed,
dilution may be one of several biases to which conversational
norms contribute. Others include underutilization of base rates and
the conjunction fallacy (Schwarz, 1994).
Disjunction effect. Shafir, Simonson, and Tversky (1993)
demonstrated in a series of experiments the sensitivity of decision
makers to the ready availability of reasons for their choices. In the
disjunction effect, people flout a basic economic axiom of rational
choice, the sure-thing principle, and sacrifice resources to delay
making a choice until an uncertainty is removed, notwithstanding
that they believe they would choose the same option no matter
which way the uncertainty was resolved. This reasoning takes the
paradoxical form Ill go on a Hawaiian vacation if I fail the exam
(to console myself) and if I pass (to celebrate), but wont decide
until I know the outcome of the exam. Reason-based choice
reaffirms the core SCM tenet that the mind is configured for
coping with accountability challenges. The guiding maxim seems
to be: Dont do anything consequential unless you have a sensible-
sounding reason for doing it.
Although the previous examples were not politically controver-
sial, the politician metaphor suggests that competing ideological
factions will often have different views of the benefits and costs of
judgmental tendencies. Drawing on the literature on political ide-
ology (Sniderman & Tetlock, 1986), the SCM posits that egalitar-
ians and authoritarians will attach systematically different nor-
mative spinsto an array of decision-making practices. In a survey
of managers, Tetlock (2000a) has documented that these disagree-
ments are closely coupled to assumptions about human nature
(authoritarians were more suspicious of human nature and thought
it prudent to hold others tightly accountable), about the nature of
the decision-making environment (authoritarians had more faith in
the efficacy of fast-and-frugal heuristics), and about effective
leadership (authoritarians applauded simple managerial styles that
projected can-do confidence but were dubious about both the
cognitive and impression-management value of preemptive self-
criticism). Effects such as overattribution and overconfidence can
polarize along parallel political and psychological lines.
Intuitive Theologians
From Aristotle to Nietzsche, philosophers have posited a deep-
rooted human need to believe that the moral codes that regulate our
lives are not arbitrary social constructions but rather are endowed
with transcendental significance. These bedrock values provide
reassuringly absolute answers to unsettling questions about the
meaning of existence and the ends to which we should devote our
lives. Moral communities strive to insulate these sacred values
from secular contamination. What looks like rigidity within other
functionalist frameworks suddenly becomes commendable: the
principled defense of the sacred from encroachments by powerful
societal trends toward science, technology, and the calculus of
capitalism (and attendant pressures to pursue inquiry wherever it
leads and to translate all values into a utility or monetary metric).
This emphasis on unswerving principles and faith distinguishes the
theologian research program from the economist and scientist
programs (which place a premium on cognitive agility in pursuit of
epistemic or market goals) as well as from pragmatic variants of
the politician program (which value agility in self-presentation).
A Testable Model of How People Function as Intuitive
Theologians
Whereas theories of the intuitive scientist and economist vary on
a rationality dimension and theories of the intuitive politician on a
social effectiveness dimension, theories within the intuitive-
theologian program vary on a ferocity continuum anchored at one
end by brutal theocrats fanatically devoted to sacred values and at
the other end by flexible pragmatists reconciled to the necessity of
compromise and perhaps even the hypocritical necessity of
concealing those compromises. The middle-range option adopted
here falls near the middle of this ferocity continuum. The sacred-
value-protection model (SVPM) can be captured in four sets of
propositions.
Observer-based predictions. The SVPM defines a sacred
value as any value toward which a moral community proclaims, at
least in rhetoric, an unbounded or infinite commitment. Building
on Durkheims (1925/1976) observations of how people respond
to affronts to the collective conscience that disrupt the normative
equilibrium, the SVPM predicts that discovery that members of
ones community have compromised sacred values triggers an
aversive arousal state, moral outrage, which has cognitive, affec-
tive, and behavioral components: harsh trait attributions to norm
violators, anger and contempt, and enthusiastic support for both
norm and meta-norm enforcement (punishing both violators and
those who shirk the sometimes burdensome task of punishing
violators).
The SVPM also predicts derogation as a function of mere
contemplation. Traditional cognitive accounts trace the difficulty
people have in making trade-offs between secular values such as
money and convenience, and sacred values, such as love and
loyalty, to the incommensurability problemthe absence of a
common metric for comparing secular and sacred values. The
458
TETLOCK
SVPM insists, however, that people find such trade-offs not only
cognitively confusing but morally disturbing and traces this reac-
tion to a deeper or constitutive form of incommensurability
(Durkheim, 1925/1976). Peoples commitments to other people
require them to deny that certain things are comparable. To attach
a finite monetary value to ones friendships, children, or loyalty to
ones country is to disqualify oneself from membership in the
associated moral community. Constitutive incommensurability
arises whenever treating values as commensurable subverts one of
the values in the trade-off calculus. Taboo trade-offs are, in this
sense, morally corrosive. The longer observers believe that a
decision maker has contemplated an indecent proposal, the more
negative their assessments of that persons character, even if that
person ultimately makes the right choice and affirms the sacred
value.
Actor-based predictions. Resource constraints sometimes bring
people into threateningly close psychological contact with propos-
als that compromise sacred values. The SVPM predicts that deci-
sion makers will feel tainted by merely contemplating scenarios
that breach the psychic wall between secular and sacred and will
engage in symbolic acts of moral cleansing that reaffirm their
solidarity with their moral community. This SVPM prediction
should not, however, be confused with the self-affirmation hypoth-
eses that can be derived from variants of dissonance and social
identity theory (Schlenker, 1982, 1985; Steele, 1988). The SVPM
differs from these mostly complementary formulations in two key
ways. First, the SVPM predicts a mere contemplation effect:It
is not necessary to commit a counternormative act; it is sufficient
for counternormative thoughts to flicker briefly through conscious-
ness prior to rejecting them. That prerejection interval, during
which ones natural first reaction to propositions is to consent
(Gilbert, 1991), can produce a subjective sensehowever unjus-
tifiedthat one has been contaminated and fallen from moral
grace. Second, the logic of constitutive incommensurability dic-
tates that the longer one contemplates taboo-breaching proposals,
the greater the subjective contamination and estrangement from
the collective. Unlike dissonance theory, which, in its original
form, focused solely on the intrapsychic function of maintaining
mental equilibrium or which, in later revisionist forms, focused on
protecting the self, the SVPM assigns double-barreled roles to
outrage and cleansing: the expressive function of convincing one-
self and others of ones worthiness and the instrumental function
of shoring up the moral order by punishing violators and by
personifying compliance.
Ideational-content predictions. The SVPM identifies three
categories of proscribed social cognitiontaboo trade-offs, for-
bidden base rates, and heretical counterfactualseach a potential
content boundary condition on strictly cognitive theories of judg-
ment and choice. To predict exactly where people draw the line
between the thinkable and unthinkable, it is, however, necessary to
link the process assumptions of the SVPM with auxiliary con-
tent assumptions drawn from Fiskes (1991) cross-cultural tax-
onomy of relational schemata that specifies how people compart-
mentalize their social worlds. Drawing on Fiske and Tetlock
(1997), the SVPM defines a taboo trade-off as any proposal that
inappropriately extends a market-pricing schema (which presup-
poses ratio-scale fungibility of all values into a single utility
metric) to spheres of activity regulated by three less metrically
demanding relational schemata: equality matching (e.g., offering
to pay a friend for dinner at her house instead of reciprocating the
invitation), authority ranking (e.g., attempting to bribe legitimate
authority figures rather than deferring to their judgment), and
communal sharing (e.g., treating loved ones as objects of monetary
calculation rather than displaying unconditional commitment to
their well-being).
The SVPM maintains that categorical proscriptions on cognition
can also be triggered by blocking the implementation of relational
schemata in sensitive domains. For example, forbidden base rates
can be defined as any statistical generalization that devout
Bayesians would not hesitate to insert into their likelihood com-
putations but that deeply offends a moral community. In late
20th-century America, egalitarian movements struggled to purge
racial discrimination and its residual effects from society (Snider-
man & Tetlock, 1986). This goal was justified in communal-
sharing terms (we all belong to the same national family) and in
equality-matching terms (lets rectify an inequitable relation-
ship). Either way, individual or corporate actors who use statis-
tical generalizations (about crime, academic achievement, etc.) to
justify disadvantaging already disadvantaged populations are less
likely to be lauded as savvy intuitive statisticians than they are to
be condemned for their moral insensitivity.
Heretical counterfactuals can be defined as what-if assertions
about historical causality (framed as subjunctive conditionals with
false antecedents) that pass conventional tests of plausibility but
that subvert core religious or political beliefs. Cultures that em-
phasize authority ranking are likely to treat counterfactuals as
heretical that reduce the conduct of higher-spiritual-status be-
ingsleaders, saints, deitiesto explanatory generalizations that
are routinely applied to lower-spiritual-status beings. For example,
Christian fundamentalists see a direct authority-ranking relation-
ship between God and humanity. Believers are supposed to defer
to the Scriptures, the literal word of God. Counterfactuals that
depict the life of Christ as subject to the same vagaries of chance
as the lives of ordinary mortals mock Gods message. Heretical
counterfactuals are deeply disrespectful and, in earlier times,
would have merited painful punishment. In modern societies,
dissenters can escape such Draconian sanctions, but they still must
endure the scorn of the faithful.
Reality-constraint predictions. The SVPM portrays people
(intuitive theologians) engaged in a delicate mental balancing act.
On the one hand, people are posited to be sincere in their protes-
tations that certain values are sacred. On the other hand, people do
run into decision problems in which the costs of treating sacred
values as infinitely important become prohibitive. The model
predicts that, absent social pressure to confront the contradiction,
people will be motivated to look away and will be easily distracted
by rhetorical diversions or smoke screens. However, when the
gaze-aversion coping response is not an option, people will search
for ways of redefining the situation that transform taboo trade-offs
into either routine trade-offs (one secular value against another, the
sort of mental operation one performs every time one buys or sells
goods and services in competitive markets) or tragic trade-offs
(one sacred value against another, such as honor vs. life, the sort
of comparison featured prominently in classic Greek tragedies).
The latter types of trade-offs do not have the morally corrosive
effects that taboo trade-offs have on both personal and social
identity.
459
FUNCTIONALIST PLURALISM
Evidence Bearing on Key Predictions of the Model
Tests of the SVPM presuppose a great deal of culture-specific
knowledge of what people hold sacred. The focal applications have
been to taboo trade-offs, forbidden base rates, and heretical
counterfactuals.
What counts as a taboo trade-off hinges on ones political
reference group. Tetlock et al. (1996) found considerable agree-
ment among liberal Democrats and conservative Republicans on
the boundaries of the fungible. Widely agreed-on taboo trade-offs
included body organs, adoption rights for children, and basic rights
and responsibilities of citizenship. Disagreement arose mostly on
the ideological fringes: among libertarians who thought it a good
idea to extend market-pricing norms into taboo territory and so-
cialists who thought it a bad idea to permit market-pricing norms
in currently permissible domains such as medical care, legal coun-
sel, and housing.
Building on this earlier work reconnoitering moral boundaries,
Tetlock, Kristel, Elson, Green, and Lerner (2000) explored the
SVPM notion that taboo trade-offs (which pit nonfungible sacred
values against fungible secular ones) possessed distinctive prop-
erties relative to routine trade-offs (which pit secular values
against each other) and tragic trade-offs (which pit sacred values
against each other). Comparisons of reactions to routine and taboo
trade-offs yielded results consistent with the more intuitively ob-
vious implications of the SVPM. Taboo but not routine trade-offs
provoked participants. People censured, directed anger toward, and
ostracized those who made taboo trade-offs. They also engaged in
meta-norm enforcement, imposing sanctions on observers who fail
to condemn taboo trade-offs.
Comparisons of taboo and tragic trade-offs yielded results con-
sistent with the less intuitively obvious predictions of the SVPM.
Tetlock et al. (2000) documented a mere contemplation effect
for moral outrage that took opposite functional forms for taboo and
tragic trade-offs. The longer observers believed that decision mak-
ers pondered a taboo trade-off (e.g., the dollar value of lives), the
more damage done to their reputations, even if they ultimately
made the right choice of lives over money. By contrast, the
longer observers believed that decision makers thought about a
tragic trade-off (e.g., which life to save?), the wiser and more
judicious they were deemed to be. Dwelling on a sleazy taboo
trade-off increases its moral corrosiveness. Dwelling on an en-
nobling tragic trade-off reinforces the impression that one has
displayed due respect for the profound importance of each of the
clashing sacred values. In this same vein, Tetlock et al. (2000) also
documented a mere-contemplation effect on moral cleansing.
Merely contemplating taboo (but not tragic) trade-offs motivated
decision makers to reaffirm their moral character by expressing
stronger intentions to perform good deeds.
Tetlock et al. (2000) documented similar outrage and cleansing
responses to forbidden base rates and heretical counterfactuals.
One study manipulated observers beliefs about the correlation
between a base-rate classification (the distribution of fires across
neighborhoods) and a forbidden predictor such as race. Observers
with egalitarian views reacted with outrage to executives who used
race-correlated base rates in setting premiums but not to executives
who used race-neutral base rates. In a follow-up experiment,
participants role-played executives who discovered that they had
inadvertently used a forbidden base rate. Thrust into this moral
predicament, racial egalitarians were more likely to morally
cleanse themselves by volunteering for good causes.
Another study tested the SVPM prediction that moral commu-
nities will erect emotionally charged boundaries against counter-
factual speculation that applies secular or scientific standards of
evidence and proof to the founders of sacred movements. Tetlock
et al. (2000) showed that fundamentalists were outraged by heret-
ical counterfactuals that undid key decisions by Jesus Christ but
reacted with equanimity to everyday counterfactuals derived from
the same schemata. Fundamentalists also felt contaminated by
merely contemplating such counterfactuals and reported renewed
dedication to serving their church.
As noted earlier, the SVPM is but one of a range of testable
theories of how people function as intuitive theologians. It is
possible in any given culturalhistorical context that the model
either underestimates or overestimates how dogmatic, intolerant,
or punitive people are prepared to be. For instance, advocates of
more cognitively flexible portraits of the intuitive theologian sus-
pect that sacred values may merely be pseudosacred (L. Thomp-
son, 2001). Insofar as there is disagreement here, it is over how
much emphasis to place on the reality-constraint postulate of the
SVPM. How quickly do people start looking for good pretexts for
abandoning their supposedly bedrock commitments to sacred val-
ues? In our laboratory, we do indeed find some evidence that
people can be inducedthrough the give-and-take dynamics of
political debateto redefine taboo trade-offs as either routine or
tragic in character. Tetlock (2000b) found that small but significant
fractions of samples will qualify their opposition to the buying and
selling of body organs for medical transplants if convinced that (a)
such transactions are the only way to save lives that otherwise
would have been lost and (b) steps have been taken to assist the
poor in purchasing organs and to prevent the poor from selling
their organs in deals of desperation. The first type of argument
effectively sacralizes the secular side of the trade-off (recasting
the issue as one sacred value against another: lives vs. moral
objections to commodifying organs); the second type of argument
secularizes the sacred side of the trade-off (now recasting the
issue as one in which it may be possible to use transfer payments
to neutralize at least egalitarian objections to organ markets). A
once clear-cut taboo trade-off thus blurs into either a tragic trade-
off (in the first case) or a routine trade-off (in the second). Either
way, the SVPM needs to acknowledge that the primal Polynesian
meaning of tabooabsolute, automatic, unreasoned aversion to
any breach of the psychic barriers separating profane from sacred
(Radcliffe-Brown, 1952)is occasionally misleading. Anathema
though the idea is to hardliners, the permeability of these barriers
(to use a Lewinian metaphor) may be a negotiable item for cog-
nitively flexible theologians.
Advocates of more cognitively flexible portraits of intuitive
theologians can also point to historical evidence of how once-
taboo trade-offs or exchanges have become permissible (nouveau
riche merchants buying the sacred land of financially strapped
feudal lords) and of how routine trade-offs and exchanges have
become taboo (between the U.S. Civil War and World War I, it
ceased to be acceptable to pay others to perform military-service
obligations). These theorists can even argue that if the SVPM is
right that people have a deep-rooted need to believe that they are
faithfully upholding sacred values, and if microeconomists are
right that resource constraints regularly compel taboo trade-offs
460
TETLOCK
anyway, then surely moral outrage should be far more pervasive
than it is. Why arent policy makers constantly vilified? One
possibility is that policy makers out-fox the public by cloaking
taboo trade-offs in artful rhetoric. Tetlocks (2000b) obfuscation
experiments suggest that these cloaking exercises need not even be
all that artful. Policy makers who traded off lives and money in
cleaning up toxic waste sites were sharply censured when they
honestly acknowledged what they had done but they escaped
censure when they offered vacuous utilitarian or deontic justifica-
tionssuch as on balance or in principle Policy X is the right
thing to dofor the same decision. This result may resolve the
paradox of how moral communities can sustain the illusion of an
unbounded commitment to sacred values when econometric (re-
vealed preference) studies indicate the community regularly trades
those values for secular ones. Moral communities may be instruc-
tively modeled as systems of organized hypocrisy (Brunsson,
1989), whose members either cannot see or do not try to see
through rhetorical smoke screens for taboo trade-offs and who
sometimes even collude in perpetuating the deception.
It is also possible, however, to criticize the SVPM for underes-
timating the moral resolution of defenders of sacred values. Green-
berg, Pyszcynski, Solomon, Simon, and Breuss (1994) terror
management theory, for example, suggests conditions when intu-
itive theologians may become especially punitive. Reminding in-
tuitive theologians of their mortality should motivate them to seek
out the existential comfort of a shared worldview that transcends
their life spans. Linking this inventive work on mortality salience
to the SVPM leads to the hypothesis that, agnostic Bayesian
libertarians excepted, people reminded of their mortality should
become especially hard-line theologians who are motivated to
defend their cultural worldview and are easily roused to wrath by
challenges to the collective moral order.
Connecting the SVPM to work on self-presentation also sug-
gests conditions under which intuitive theologians may become
sterner defenders of sacred values. One hypothesis is rooted in the
plausible assumption that people derive social identity benefits
from ostentatious displays of virtue. Moral outrage and cleansing
should be amplified when people feel monitored by their commu-
nity of cobelieverswhich may explain the extremity of punitive-
damage awards levied by juries when corporations have been
caught placing explicit dollar values on human lives (Sunstein,
Kahneman, & Schkade, 1998). A related hypothesis is that, when
people are prevented from distancing themselves from proscribed
cognitions and this failure to condemn is observed by cobelievers,
people will become self-dramatizing theologians who seek reinte-
gration into the community by publicly cleansing themselves and
denouncing norm violators.
Normative Challenges to Prevailing Taxonomies of Error
and Bias
True believers dismiss the opportunity costs of shunning taboo
trade-offs or the distortions in market pricing from ignoring for-
bidden base ratesalthough intuitive economists and statisticians
are vexed by these deviations from rationality narrowly construed.
True believers halt attributional inquiry as soon as it trespasses into
the domain of the sacredalthough intuitive scientists are dis-
turbed by the abrupt curtailment of curiosity about heretical coun-
terfactuals. True believers deplore the belief-bending ways of
pragmatic intuitive politicians. When sacred values are at stake,
attitude shifting smacks of sellout; self-criticism blurs into genuine
doubt on articles of faith, and decision avoidance looks like
cowardice.
Intuitive theologians tend to be suspicious of the classic En-
lightenment values of open-minded inquiry, free markets, and of
tolerant pluralism. Some questions should never be asked, some
transactions never proposed, and some compromises never struck.
How can it be determined, though, whether the coping strategies of
intuitive theologians serve their intended function? The new nor-
mative benchmarks are mostly subjective: Do people who aim
moral outrage at norm violators feel vindicated by the opportunity
to do so (and frustrated when thwarted)? Do moral cleansers
follow through on their announced intentions to do good deeds that
then, by a dissonance or self-perception process, further solidify
their faith? In addition, however, there may be communitarian
benchmarks of effectiveness. Durkheim (1925/1976) may have
been right. Group expressions of outrage and cleansing may in-
crease the cohesiveness of communities of cobelievers. Commu-
nities that censure together may stay together.
Intuitive Prosecutors
The core functionalist premise is that people seek to defend
rules and regimes that they endow with legitimacy. Again, as with
any functionalist metaphor, there is a conceptual range of permis-
sible theories, ranging from the conscientious jurist (who draws
conclusions about culpability only after dispassionately weighing
the evidence) to the capricious, vengeful, or opportunistic prose-
cutor (who permits mean-spirited or self-interested motives to taint
evaluations). But there are empirical constraints. Theorists will
ultimately be compelled to incorporate personality variation (some
people are more punitive than others), situational variation (certain
contexts promote leniency or punitiveness) and personality-by-
situation interactions (certain contexts move some but not others to
be punitive). These concessions to complexity are required by
salient features of the social landscape: (a) by the bitter partisan
divisions between those who seek procedural safeguards that
shield individuals from arbitrary exercises of state power and those
who seek to empower society to deter misconducteven if at the
expense of individual rights (cf. Sniderman, Fletcher, Russell, &
Tetlock, 1996); (b) by the deep philosophical divisions over the
appropriate goals of punishment, ranging from rehabilitation to
specific and general deterrence to retribution (D. T. Miller &
Vidmar, 1981); (c) by the vast variation in how blatantly biased
in-groups are toward out-groupsan historical continuum an-
chored at one end by state-sponsored genocide and at the other by
systematic exercises in constitutional engineering to promote tol-
erance and pluralism (Horowitz, 1985).
A Testable Model of How People Function as Intuitive
Prosecutors
The theory advanced here strikes a compromise among these
competing positions: the fair-but-biased-and-occasionally-erratic
model of the intuitive prosecutor. This fair-biased-erratic (FBE)
model can be captured in five sets of propositions.
Commitment, albeit imperfect, to procedural fairness. People
think of themselves as fair and pay homage to basic norms of
461
FUNCTIONALIST PLURALISM
procedural justice that stipulate they should weigh evidence im-
partially, accord equal respect to each side, and respect fundamen-
tal human rights, even if doing so impedes apprehending violators
(Tyler & Smith, 1998). The FBE does not challenge the sincerity
of this commitment to procedural norms, but it does challenge the
consistency with which people apply abstract norms to specific
cases and it predicts considerable principle-policy slippage for
reasons that follow.
Mechanisms for switching the prosecutorial mind-set “on” or
“off.” The prosecutorial mind-set can be activated by experi-
mental manipulations of the perceived fragility of the social order
as well as by chronic personality variations in these perceptions. It
makes sense to become warier of situational justifications and
excuses for violating societys rules to the degree one has been
induced to believe, or is already predisposed to believe, that
misconduct is on the rise, that growing percentages of misconduct
are going unpunished, and that growing numbers of citizens are
tempted, by the collapse of general deterrence, to behave badly.
Once this prosecutorial mind-set is activated, it seems only prudent
to place greater weight on punishing norm violators (minimizing
Type 2 errors of acquitting the guilty) and less weight on protect-
ing the innocent from the wrath of the collective (minimizing
Type 1 errors of convicting the innocent). This policy preference
is rarely stated so baldly (bordering as it does on a taboo trade-off).
Intuitive prosecutors should, however, feel more justified in adopt-
ing an across-the-board hard-line stance to the degree they believe
that Type II errors are both more likely and more serious than
Type 1 errorsbeliefs that, in turn, hinge on assumptions about
human nature as well as the benevolence and competence of
authority figures.
Of course, even the most judgmental judges do not relish living
under an oppressive regime that renders them constantly vulnera-
ble to false accusations. Intuitive prosecutors must strike a balance
between upholding the social order and neutralizing defensive
strategies that they themselves might want to deploy when they are
in the metaphorical docket. The FBE therefore posits an exculpa-
tion gradient for the accounts offered by those accused of wrong-
doing. This gradient is anchored at one end by diminished-capacity
excuses that invoke well-established biological causes outside
volitional control (e.g., epilepsy) that virtually no one challenges.
At the other end, it is anchored by frivolous justifications or
excuses that virtually no one accepts and that may well backfire
(e.g., I was too drunk to see the pedestrian). Toward the middle
of the gradient are such defenses as addiction, past victimization,
and peer pressure that possess some plausibility but that intuitive
prosecutors fear can readily degenerate into abuse excuses. Effects
of prosecutorialmind-set manipulations should be most pro-
nounced in this zone of intermediate exculpatory power. This zone
is not, however, fixed forever. The relative openness of intuitive
prosecutors to situational defenses is a quasi-stationary equilib-
rium (to use a Lewinian formalism) that shifts with the relative
force of the conflicting pressures to avoid Type I and II errors.
The FBE also posits that, even in the absence of pressing threats,
the prosecutorial mind-set should be more chronically accessible
for those who are ideologically predisposed to believe that the
normative order is at risk and that it is therefore prudent to set low
thresholds for attributing intentionality to norm violators. This
prediction does need, however, to be qualified by ideology-by-
context interactions. Intuitive prosecutors do not dispense blind
justice. The predisposition to blame should be moderated by the
affect people harbor toward the act (e.g., certain actsflag burning
vs. cross burningare more vexing to certain groups), toward the
actor (e.g., is there evidence of repentance or extenuating circum-
stances?) and the group affiliation of the actor (e.g., does the group
uphold or subvert existing social structures?). Conservatives
should be maximally punitive when the consequences are severe
and when the perpetrator cannot invoke extenuating circum-
stances, is unrepentant, and is affiliated with groups flagrantly
contemptuous of the social order. Egalitarians should be maxi-
mally punitive under similar circumstances, except that the act
should target a traditionally disadvantaged group and the perpe-
trator should hold socioeconomic underdogs in disdain (cf. Skitka
& Tetlock, 1993).
There is a palpable tension between the fair and biased
components of the FBE. The FBE resolves this tension by positing
that intuitive prosecutors will play favorites only when they can
generate convincing justifications that they have not done so. In
the language of impression management, biased prosecutors need
the cover of attributional ambiguity to deflect accusations of bias.
Naked discrimination is unacceptable in most cultures wealthy
enough to fund psychological research programs.
Intuitive prosecutors: Forward- and backward-looking mind-
sets. Intuitive prosecutors can be either forward looking or back-
ward looking when they translate their attributions of responsibil-
ity and emotional reactions into recommendations of punishment.
The forward-looking mind-set should be activated in settings that
legitimize the efficiency logic of Homo economicus: Given press-
ing resource constraints, the groups goal is to identify cost-
effective forms of punishment that promote specific and general
deterrence (minimize the likelihood that the offender or others will
act in the proscribed manner). The backward-looking mind-set
should be activated in settings that legitimize the retributive quest
to restore the ante of the moral status quo, settings such as cultures
of honor in which agents of social control had better move quickly
to reaffirm the dignity of victims lest the victims take justice into
their own hands (Nisbett & Cohen, 1996).
Although retributive justice is often portrayed as a mindlessly
vengeful and limbic reaction to transgressions, the FBE traces the
desire for retribution to a more reasoned calculus in which people
implicitly think along the following lines: The transgressors
treated the victims contemptuously; the damage to the victims
dignity is real; the transgressors see themselves as above the
victims and the laws of society; the punishment should humiliate
the transgressors to a degree commensurate with the humiliation
inflicted on the victims; anything less fails to reaffirm the dignity
of the victims and indeed adds to the original insult. The FBE
predicts that retributive prosecutors will impose punishment be-
yond that required by deterrence whenever the deterrence prescrip-
tion for punishment falls short of reaffirming the identity damage
to victims. The acrimonious debate over hate crimes illustrates the
point. Advocates of traditionally victimized groups often insist
that, even if the conventional penalties for assault and battery are
adequate for deterrence, additional penalties for hate-motivated
crimes affirm the value that society now places on protecting
groups wronged in the past (cf. Jacobs & Potter, 1998). There is,
however, an ironic twist. Advocates of laws against hate crimes are
mostly on the left, which has traditionally denied the legitimacy of
retribution, whereas opponents of such laws are mostly on the
462
TETLOCK
right, which has traditionally upheld retribution as a legitimate
goal.
Irrelevant affective states. Intuitive prosecutors can be fickle
and inconsistent, allowing irrelevant moods and emotions to taint
attributions of responsibility. These irrelevant affective states fall
into two categories: spillover from classic Durkheimian outrage
triggered by violation of societys standards and spillover from
disruptions to the classic Freudian equilibrium between unaccept-
able impulses and the internalized voice of conscience (disruptions
triggered by violations of norms that one is tempted to violate
oneself).
Flexible self-correction. When motivated and able, intuitive
prosecutors do engage in flexible self-correction of their judgment
processes, checking irrelevant influences and difficult-to-justify
double standards (cf. Wegener & Petty, 1995). Flexible self-
correction should work best when intuitive prosecutors are them-
selves under tight accountability scrutiny, possess a good intuitive
sense of the direction and magnitude of the potential bias, and are
not under cognitive load.
Evidence Bearing on Key Predictions of the Model
Tetlock et al. (2002) reported a series of experimental tests of
FBE predictions. These studies manipulated independent variables
designed to activate the prosecutorial mind-set, especially beliefs
about the direction and rate of change of certain offenses and the
proportion of offenses going unpunished. By the logic of the FBE,
these manipulations should promote punitive dispositional attribu-
tions that advance the goals of specific and general deterrence and
of retribution. Results indicated that people in the prosecutorial
mind-set were (a) less moved to leniency by extenuating circum-
stances but more moved to punitiveness by exacerbating circum-
stances (evidence that intuitive prosecutors were not oblivious to
context, just selective in which cues they used); (b) more punitive
as the severity of the consequences of an act mounted, holding
intentionality and foreseeability constant (extending past work on
the severity effectBurger, 1981); and (c) less interested in over-
turning verdicts when there had been a violation of procedural
justice that could be dismissed as a technicality. It was difficult,
however, to activate the prosecutorial mind-set when people sym-
pathized with the violator or were reminded of the dangers of false
conviction.
Tetlock et al. (2002) also found support for personality-by-
context predictions of the FBE. In studies of managers, authori-
tarian conservatives favored a managerial style that gave little
leeway for justifications or excuses for failing to achieve organi-
zational objectives, whereas liberals preferred an empathic style
that gave floundering employees the benefit of the doubt. A certain
insensitivity to mitigating situational factors is, from a conserva-
tive standpoint, a useful message to send to subordinates tempted
to test the tolerance zone of acceptable performance by offering
exculpatory accounts. Conservatives were not always, however,
the most vigilant defenders of the normative order. Conservatives
were more forgiving of an overzealous police officer accused of
beating a suspected drug pusher. And liberals were more punitive
toward racist vandals and toward corporations that befouled the
natural environment. It is also worth stressing that the prosecuto-
rial mind-set was not inherently simplistic. Intuitive prosecutors
engaged in relatively complex attributional analysis that required
greater sensitivity to contextual cues when those charged with
enforcing norms were overzealous in their duties. People in the
prosecutorial mind-set were more likely to let off a police officer
who beat a suspected drug dealer when there was an extenuating
circumstance (emotional stress). Moreover, intuitive prosecutors
were not indiscriminately soft on the police. The police got no
benefit from mitigating circumstances when they stood accused of
corruption.
Affect played a negligible role in mediating the effects obtained
in the low-emotional-involvement vignette studies of Tetlock et al.
(2002). Investigators using more vivid stimulus materials have,
however, found evidence for more emotion-driven portraits of the
intuitive prosecutor. Goldberg, Lerner, and Tetlock (1999) exam-
ined anger carryover effects by persuading subjects that they were
participating in two unrelated experiments. In the first study,
people watched a videotape of a brutal assault on a young man and
subsequently learned that the aggressor was or was not punished.
Later, participants judged culpability in tort cases in which there
was presumptive evidence of negligence. Although subjects in the
punished and unpunished conditions were equally angered by
the assault, only participants who believed that the perpetrator had
gone unpunished translated their anger over the original offense
into punitiveness toward perpetrators of torts. It was as though
people needed a Durkheimian warrant for translating anger over
unpunished breaches of norms into punitiveness in later unrelated
cases. The warrant effectively tagged the anger as righteous, and
subjects in that state drastically simplified their attributional anal-
ysis, moving directly from acts to dispositions, with only a per-
functory pause for extenuating circumstances.
J. S. Lerner et al. (1998) deployed the same anger-priming tape
to assess the impact of case-irrelevant emotions, authoritarianism,
and accountability on punitiveness in fictional tort cases. As pre-
dicted, anger simplified processing of tort information (reducing
the complexity of cues utilization) and encouraged punitive attri-
butions to defendants (negligence began to shade into malev-
olence). But when intuitive prosecutors were held accountable for
their tort judgments, the effects of irrelevant anger on complexity
of thought and punitiveness disappeared.
Normative Challenges to Conventional Taxonomies of
Error and Bias
The distinction between the politician and prosecutor metaphors
can be as murky as the distinction between politics and the law
(witness the endless disputes over politicized judges). In the
prosecutorial arena, as in the political arena, what constitutes an
error or bias is often bitterly contested (Sniderman et al., 1996). In
studies of managerial and political elites, Tetlock (2000a) has
documented ideological fissures in judgments of the rationality of
individual, small group, and institutional decision processes. Au-
thoritarian conservatives believe that most people look for and
exploit loopholes in social control systems, whereas egalitarian
liberals believe that most people will refrain from exploiting
loopholes as long as they feel fairly treated. These discrepant
views of human trustworthiness lead to diverging assessments of
the fundamental attribution error. Authoritarian managers think it
prudent to communicate a low tolerance for justifications and
excuses for conduct that falls short of organizational expectations
and that people will try harder if they believe that failure will
463
FUNCTIONALIST PLURALISM
automatically tarnish their reputationsa social variant of the
legal doctrine of strict liability. Failing to hold people responsible
for outcomes that they could have controlled is every bit as serious
an error as holding people responsible for outcomes outside their
control. By contrast, many liberals view the fundamental attribu-
tion error as punitive, not prudent. They argue that conservatives
exaggerate how frequently people will invent specious justifica-
tions and excuses for substandard performance.
Prosecutorial priorities also collide with epistemic or economic
ones in other arenas. These arenas include the following.
Severity effects. A drunk driver speeds through a red light and,
in one scenario, there is a minor fender bender and, in the other, a
child is killed. From a prosecutorial perspective, people should
deem it justifiable to mete out harsher punishment as a function of
outcome severity, holding volition, foreseeability, and intentional-
ity constant (Tetlock 2000a). The severity effect is a sound method
of sending the message to miscreants that the risks they recklessly
inflict on others can randomly rebound back on them. But from an
intuitive-scientist perspective, the same effect looks like an irra-
tional intrusion of the certainty-of-hindsight bias into the judgment
process. Once observers learn of the outcome, they erroneously
infer that they knew what was coming all along and that the actor
therefore should also have foreseen the outcome.
The ethical impasse cannot, of course, be broken empirically,
but the degree to which ordinary people resonate to the function-
alist defense of the severity effect by the intuitive-prosecutor camp
and the functionalist critique from the intuitive-scientist camp can
be gauged. Tetlock et al. (2002) assessed the degree to which
liberal and conservative managers found it embarrassing to ac-
knowledge that they impose different penalties on careless em-
ployees as a function of the severity of the consequences. Using a
repeated-measures design that made it transparent which cues were
being manipulated, they found that (a) liberals were more lenient
than conservatives and (b) both liberals and conservatives dis-
played a severity effect and were harsher when they first learned
that the consequence of carelessness was serious than when they
first learned the consequence was trivial. However, in response to
later, what-if questions, conservatives saw nothing wrong with
linking penalties to severity of consequences. The severity effect
was good policy. By contrast, liberals were wary of judging people
on the basis of outcomes that were so under the sway of chance.
Liberals were thus more reluctant to modify their initial attribu-
tions of responsibility in response to a cue that they thought should
be given little weight. As a result, liberals who first received the
severe-consequence and then the mild-consequence scenario were
actually more punitive in reactions to the second scenario than
conservatives were.
Proceduralist and retributivist challenges to Homo economicus.
Economic models of legal reasoning assume that intuitive prose-
cutors are continually seeking the most cost-effective societal
strategies to deter wrongdoing. This doctrine of optimal deter-
rence leads to a host of predictions that clash with more psycho-
logical conceptions of intuitive prosecutors. For example, Polinsky
and Shavells (1999) analysis of punitive damages asserted that the
principal function of such awards is to make up for the shortfall in
compensatory damages produced by the failure of many potential
plaintiffs either to detect their injuries or to secure compensation.
This leads to the counterintuitive prediction that Homo economicus
in the legal arena should multiply compensatory awards by the
inverse of the probability that any given person from the popula-
tion of injured persons would receive compensation for the injury.
The fewer perpetrators caught, the more harshly those who are
caught should be punished. Sunstein et al. (1998) documented that
people are understandably reluctant to do this (by this reasoning,
why not save money by laying off a large percentage of the
highway patrol and beheading the hapless few speeders who are
apprehended?). From a narrowly economic standpoint, this recal-
citrance should count as a judgmental defect. The FBE differs from
the Polinsky and Shavell analysis in two key respects. First, it
posits that people care about equitable treatment and this concern
should animate opposition to concentrating punishment on the
hapless few to compensate for the many scofflaws. Hence the FBE
predicts less punitiveness than does the microeconomic model.
Second, the FBE posits that people should punish more severely
than efficiency considerations would dictate when the cost-
effective punishment fails to fulfill the retributive function of
humbling arrogant violators to compensate for the humiliations
that they inflicted on the victims.
Available evidence supports these hypotheses. Baron and Ritov
(1993) found that penalties imposed on corporations for producing
defective birth control pills and vaccines were unswayed by
whether the penalties had beneficial consequences (encourage the
manufacture of safer products) or harmful ones (cause companies
to withdraw from the market, making products less available, more
expensive, and less safe). Wrongdoers must be punishedeven if
it is painful to society as a whole. Baron and Ritov also found that
the penalties were unaffected by removing all utilitarian rationales
for punishment: the penalty was secret, the company had insur-
ance, and, in any case, the company was going out of business. In
punishing companies for dumping hazardous wastes, people
wanted companies to clean up their own waste, even if the waste
did not threaten anyone and even if it were possible to redirect the
same money to clean up the waste of a now-defunct company that
posed a far greater hazard (Baron, Gowda, & Kunreuther, 1993).
From a microeconomic view, these choices are perversely Pareto
suboptimal. But these choices are easily reconciled with a portrait
of the intuitive prosecutor as guided by the primitive Durk-
heimian intuitions that those who harm others, including nature,
must restore the status quo ante.
Collective punishment. Psychologists often dismiss collective
punishment as crude guilt by association (Heider, 1958). The
policy can, however, be defended as a rational prosecutorial strat-
egy to stimulate mutual accountability among group members and
to forge a communitarian consciousness in which people rarely
feel justified in replying none of your business to demands for
accounts from family, friends, and coworkers (Edgerton, 1985).
People who invoke the none-of-your-business defense will receive
the reply. Oh, yes, it isfor we too must answer for what you
do. This functionalist analysis bears on recent claims that the
fundamental attribution error is peculiar to individualistic Western
societies in which folk psychologies stress individual agency and
vanishes in the Confucianist societies of east Asia in which folk
psychologies stress dialectical interdependence and in which
morallegal codes have traditionally endorsed collective punish-
ment (Peng & Nisbett, 1999). Isolating individual perpetrators as
the locus of causal and moral responsibility may be a distinctively
Western method of enforcing order. However, holding significant
others of the perpetrator accountable as well, and yoking their
464
TETLOCK
reputations to the perpetrator, appears to be equally effective, even
if the locus of responsibility transcends the internalexternal
dichotomy.
What Then Is to Be Done?
The approach taken here is open to critique from at least three
conceptual camps: relativism, reductionism, and falsificationism.
This section considers these objections and sketches a metatheo-
retical stance toward functionalism that strikes a principled com-
promise among the complex scientific values at stake. This com-
promise position tries to avoid the excesses of relativism (collapse
of all standards of rationality), the excesses of reductionism (the
distortions that result from squeezing social functionalist explana-
tions into exclusively intrapsychic templates), and the excesses of
falsificationism (the arbitrariness of drawing sharp distinctions
between overlapping fuzzy sets of rival functionalist theories).
Relativism
This camp argues that there is no point of diminishing returns in
diversifying the portfolio of functionalist premises. The more
conceptual vantage points for viewing reality, and the more faith-
ful those vantage points are to the phenomenology of participants,
the better. It is, moreover, insufferably elitist to suppose that the
research community knows any better than the people they study
the diverse ends that judgment and choice serve in cultural or
historical context (Geertz, 2000). Imposing a priori categories of
goals and functions blinds researchers to the transitory and idio-
syncratic webs of meaning that people are constantly spinning
around their decisions and interactions with each other.
This stance is tantamount to abandoning the nomothetic goals of
scienceexplanation and predictionin favor of the ideographic
and hermeneutic goals of history and ethnography. It also amounts
to postmodernist renunciation of rationality itself. Unconstrained
shifting of functionalist premises makes it possible to concoct post
hoc rationalizations for virtually anything people do, no matter
how incoherent, a risk that haunts even the more restrained form of
functionalist pluralism adopted here. When people appear to be
defective statisticians, the functionalist can propose that they are
attentive conversationalists on the lookout for predictive cues in
even the most vacuous utterances (Schwarz, 1994). When people
appear to be flawed psychologists, the functionalist can caution
that people are shrewd prosecutors who know how to keep others
in line (Hamilton, 1980). When people appear to be myopic
economists, the functionalist can cast them in the role of devout
theologians protecting sacred values from taboo trade-offs (Tet-
lock et al., 2000). The conceptual guidelines proposed here for
constructing social functionalist theories do, however, place clear
limits on functionalist forgiveness. The epistemological stopping
rule observed in this article has been to consider only those
functionalist reinterpretations of putative biases that lead to test-
able hypotheses and to corroborative evidence that would other-
wise have gone undiscovered.
Reductionism
This camp seeks to assimilate social functionalist theories into a
common conceptual language and explanatory template. The ideal
type of successful reductionism in science is thermodynamics to
statistical mechanics. In Nagels (1961) lucid exposition, Theory X
is reduced to, and becomes identical to, Theory Y when the bridge
laws that translate X into Y can be specified, as when the temper-
ature of a gas became synonymous with its mean molecular kinetic
energy. In an analogous vein, reductionists would insist that social
functionalism is a placeholder for ignorance of underlying cogni-
tive and affective mechanisms and that as soon as these mecha-
nisms are understood, functionalist explanations can be replaced just
as surely as statistical mechanics replaced classical thermodynamics.
There are, at present, several potentially promising conceptual
strategies for pursuing the reductionist objective. In the spirit of
signal detection theory, one could argue that manipulations of
social functionalist mind-sets have no effect on how people think
but that they do influence where people set their response thresh-
olds (betas) for expressing various categories of judgments
(Friedrich, 1993). Unaccountable and accountable decision makers
think in the same basic ways, but accountable decision makers are
more cautious about what they say. Or, to offer another example,
people in punitive mind-sets are not any less capable of differen-
tiating the innocent from the guilty, but they do place an unusually
high value on not letting the guilty go free.
In the spirit of contemporary work on construct accessibility and
spreading activation in associative networks, one could argue that
there is no need to invoke mysterious functionalist distinctions to
explain the effects of social functionalist manipulations when
simple associationist principles of the mind can do the same job.
Accountability, for example, may alter the decision calculus be-
cause it reminds people of the viewpoint of the person with whom
they expect to interact, or it reminds people of past episodes in
which they have been under pressure to justify their views, and the
coping strategies that worked come to mind. Reinvoking N. E.
Miller and Dollard (1941), social functionalist frameworks shed
light on what people think about; they tell us nothing about how
people think.
In the spirit of classic drive theory, one could argue that social
functionalist manipulations have drive- or arousal-inducing prop-
erties, and the resulting effects on cognitive performance fit into
familiar arousal-by-task-difficulty relationships that date back to
Hull-Spence days. The effects of accountability on judgment and
choice may be mediated by the effects of social facilitation arousal
on the cognitive processes required for task performance, or the
rigidity displayed by intuitive theologians who perceive sacred
values to be under siege may be consistent with the more general
literature on the effects of stress on information processing.
In the spirit of dual-process models of social cognition, one
could treat social functionalist manipulations as equivalent to other
incentives (such as money, consequences, involvement, or inter-
dependence) to engage in more effort-demanding and elaborate
information processing and thereby bring functionalist findings
into the explanatory orbit of dual-process models of social cogni-
tion (Chaiken & Trope, 1999).
There are, however, good logical and empirical grounds for
doubting the wisdom of full-scale reduction. On the empirical side,
it can be argued that the proposed reductionist equivalence either
does not hold or holds too loosely to have explanatory value. Here
the argument pivots on details of evidence. Some functionalist
manipulationsaccountability, social threat, sacred status of val-
uesdo undoubtedly influence willingness to express certain
465
FUNCTIONALIST PLURALISM
thoughts, but many effects cannot be so explained. In the account-
ability literature, there are five methodologically distinct sources
of evidence that the cognitive processes, activated by self-
criticisminducing forms of accountability, are both more compu-
tationally demanding and effort taxing than a simple response-
threshold-adjustment process of moving mindlessly toward the
middle would be. The evidence includes the following.
1. Self-criticisminducing accountability manipulations in-
crease the complexity of thought as revealed by content analysis of
confidential thought protocols and statistical modeling of cue
utilization (Ashton, 1992; Hagafors & Brehmer, 1983; Tetlock et
al., 1989).
2. Self-criticisminducing manipulations are more effective in
attenuating biases such as primacy, overattribution, and overcon-
fidence only when participants learn of being accountable prior to
viewing the evidence on which they base judgments (Tetlock,
1983b, 1985; Tetlock & Kim, 1987; E. P. Thompson, Roman,
Moskowitz, Chaiken, & Bargh, 1994). Timing should not matter if
people were simply shifting response thresholds; it should matter
if accountability effects were mediated by the initial interpretation
of evidence.
3. The power of such manipulations to attenuate bias is itself
attenuated by impositions of cognitive load that disrupt effort-
demanding forms of information processing (Kruglanski &
Freund, 1983)disruption that should not occur if people coped
with accountability by relying exclusively on low-effort attitude
shifting.
4. When such manipulations improve judgmental accuracy, the
effects are often too differentiated to reproduce by response-
threshold adjustment models. These effects include enhanced dif-
ferential accuracy from a Cronbach decomposition of person
perception accuracy (Tetlock & Kim, 1987; Mero & Motowidlo,
1995), improved correspondence between subjective confidence in
predictions and objective accuracywithout degradation in ability
to assign differential confidence to accurate and inaccurate predic-
tions (Siegel-Jacobs & Yates, 1996; Tetlock & Kim, 1987), and
closer correspondence between subjective importance of cues in
prediction tasks and objective weightings derived from multiple
regression (Hagafors & Brehmer, 1983).
5. The effects of such manipulations on thoughts persist even
after cancellation of the anticipated interview (Pennington &
Schlenkler, 1999).
Some social functionalist manipulations have drive- or arousal-
inducing properties, but it is also awkward to explain existing
accountability effects in this fashion. Consider the difficulties that
arise in assimilating the effects of accountability on judgment and
choice into a social-facilitation framework that joins assumptions
about the drive-inducing properties of the mere presence of an
audience with assumptions about the effects of drive arousal on
performance of inferential tasks of varying difficulty (Pelham &
Neter, 1995; Zajonc, 1965). From the standpoint of the SCM, this
reduction is fated to fail because a unidimensional arousal metric
cannot fit, even post hoc, the complex variations in accountability
effects that depend on qualitative features of the relationship
between decision maker and audience. Too much hinges on when
people learn of being accountable, whether they know the views of
the audience, and how exactly they perceive the accountability
requests or demands (J. S. Lerner & Tetlock, 1999). It is also
dauntingly difficult to reduce the diverse judgment-and-choice-
dependent variables to a single task-difficulty metric that permits
deduction of Hull-Spence Task-Difficulty Arousal interactions.
To fit the data, one must make so many auxiliary assumptions that
the original advantage of parsimony is lost.
Some social functionalist manipulations induce mood or emo-
tional states, and it is tempting to posit that the effects on judgment
are mediated by the effects of emotion on the accessibility of
cognitions and on ability or willingness to engage in sober second
thought (theories that usually work from the default assumptions
that people are trying to understand the world, like intuitive sci-
entists, but are subject to certain emotional distortions; Forgas,
1995). This challenge is hard to rule out. It is difficult to conceive
of any experimental induction of the prosecutorial mind-set or any
other social functionalist mind-set that does not alter mood or
affect and prime supportive cognitions. Even here, though, reduc-
tion is easier to announce in principle than to implement in
practice. Defenders of the FBE model can argue that anger is not
necessary to produce punitiveness in some studies, and it is not
sufficient to produce it in others (cf. Goldberg et al., 1999; Tetlock
et al., 2002).
Empirical objections to reduction can, of course, always be
surmounted as new evidence and better theories become available,
but there are also deeper logical objections to reduction. The most
compelling of these objections is captured reasonably well by
Searles (1998) famous Chinese-room thought experiment, which
he advanced to demonstrate the logical impossibility of strong
artificial intelligence and of the associated notion that appropri-
ately programming a machine is sufficient to endow it with inten-
tional states. The thought experiment asks readers to suppose that
a monolingual English speaker is isolated in a cubicle and produc-
ing Chinese answers to Chinese questions well enough to mimic a
Chinese speaker, but doing so by following an algorithm written in
English. Although the occupant of the Chinese room can readily
pass the Turing test, most observers balk at saying either that this
person, or that a computer executing the algorithm, understands
Chinese. Searle argued that, far from irrational, this resistance
would be justified for any combinatorial algorithm because algo-
rithms are syntactically individuated in the sense that, like gen-
erative grammar, they represent unique, internally consistent rule
systems for manipulating symbols that need or have no external
referents. By contrast, intentional states are semantically individ-
uated in the sense that they make epistemic and other functional
references to the external world from which the occupant of the
Chinese room has, by definition, been isolated. Searle argued for
a nonreductive biological naturalism in which intentionality, like
qualia such as the liquidity of water or the beauty of the sunset, are
treated as high-level features that are caused by and realized in the
brain but are explicable only with reference to the interactions
between the brain and the external world.
Unapologetic reductionists (such as Pinker, 1997) have coun-
tered this argument by invoking history-of-science precedents in
which reductions of surface phenomena to deeper causal principles
that initially struck contemporaries as counterintuitive are now
universally accepted. Favorite examples include heat as nothing
but the kinetic energy of molecular motion or the color red as
nothing but photon emissions of 600 nanometers. Searles thought
experiment merely shows that the underlying mechanisms of lan-
guage comprehension are similarly jarringly counterintuitive. The
counter to this counterargument is to marshal alternative history-
466
TETLOCK
of-science precedents. Reducing language comprehension or other
social functions of thought to strictly internal brain-or-mind oper-
ations, to intrapsychic combinatorics, would be every bit as mis-
leading as reducing photosynthesis to strictly physical-chemical
processes and ignoring the embeddedness of these microprocesses
in macronetworks of causality represented in ecological and evo-
lutionary systems. Causal reduction, x is a function of y, is not the
same as ontological reduction, x is nothing but y, in case of
systemically embedded phenomena such as consciousness or pho-
tosynthesis. There is nothing inconsistent about conceding that the
psychological implementation of social functions must be reduc-
ible to strictly internal brain-or-mind operations but still insisting
that social functions, connecting individual minds to social sys-
tems, are ontologically irreducible. In this view, theorists who
advocate the ontological reduction of social functions (e.g., such
functions are nothing but intrapsychic combinatorics) have over-
extended history-of-science precedents in which causal and onto-
logical reduction were one and the same.
This argument still leaves one big reductionist contender stand-
ing: a higher order form of functionalism. Evolutionary psychol-
ogists are not at all averse to functionalist speculation or to the
notion that people might think in qualitatively different ways in
different contexts (Cosmides & Tooby, 1994). Qualitative varia-
tion of this sort reflects basic properties of the human mind that are
the product of millions of years of natural and sexual selection
pressures. The evolutionary program requires reverse engineering
thought experiments that pose the question, What types of adap-
tive problems was this mental software suited to solve in our
ancestral evolutionary environments?Many implementation rules
specified for functionalist theories of judgment and choice satisfy
the solvability and evolvability criteria of Cosmides and Tooby
(1994). People, by virtue of natural and sexual selection, should
come well prepared to solve such central life tasks as competing
for scarce resources (the economist), making predictively useful
inferences about physical and social causation (the scientist/stat-
istician/psychologist), getting along with key constituencies (the
politician), punishing norm violators (the prosecutor), and even
affirming their existential significance by protecting sacred values
(the theologian). The core if ..., then production rules for
implementing these functions are hardly so dauntingly difficult
that they could not have gradually emerged over thousands of
generations in a species as intensely interactive as our ownrules
such as avoid quarreling with fellow hominids who control valu-
able resources,”“think twice before making irreversible commit-
ments, and spread damaging gossip about rivals. The more
elaborate coping strategies of intuitive politicians, prosecutors, and
theologians may be cultural overlays on modular templates.
The ultimate relevance of this framework is not in doubt:
Virtually all psychologists subscribe to some variant of evolution-
ary theory. The position of functionalist pluralism taken here is
also strikingly compatible with modularity hypotheses advanced
by theorists who challenge the once conventional wisdom that
people possess general-purpose reasoning skills and who insist that
the mind is compartmentalized into separate systems for solving
specific adaptive challenges such as linguistic communication,
cheater detection, and social perspective taking (cf. Haselton &
Buss, 2000; Hirschfeld & Gelman, 1994). That said, the reduc-
tionist move still appears premature. Post hoc data fitting rightly
earns a theory fewer pointsby most standards of epistemological
accountingthan generating novel hypotheses of the sort that can
still be derived from social functionalist theories that focus on
contemporary adaptive challenges rather than on those of the
Pleistocene.
Strict Falsificationists
Falsificationists who subscribe to a Popperian philosophy of
science demand that new functionalist frameworks justify their
existence by making predictions that clearly diverge from those of
traditional frameworks (Suppe, 1974). Functionalist research pro-
grams are, however, better viewed as exhaustible than as falsifi-
able. Even if one testable theory grounded in a functionalist
framework is falsified, it remains possible to amend the theory or
to advance alternative theories (Laudan, 1997). Moreover, the
track record for drawing sharp distinctions between functionalist
research programs has not been too inspiring in social psychology.
Tetlock and Levi (1982) documented how an eclectic array of
motivational theories of attribution bias (which stressed protecting
self-image and social image) was once regularly pitted against cold
cognitive explanations organized around the scientist metaphor.
Tetlock and Manstead (1985) documented how conceptual off-
shoots from the intrapsychic cognitive-consistency traditionthe-
ories such as cognitive dissonance, reactance, and equitywere
once pitted against impression-management theories organized
around the dramaturgical metaphor. Although these empirical con-
frontations did stimulate methodological innovations and notewor-
thy discoveries that shed light on boundary conditions for the
applicability of rival theories, after each round of hypothetico-
deductive confrontations, each side could still claim vindication
by fine-tuning auxiliary assumptions. For instance, some disso-
nance theorists maintained that evaluative audiences heighten
self-awareness and thus magnify dissonance (stealing the im-
pression management prediction that accountability to an audience
magnifies social identity-protection motives) and some impression-
management theorists argued that people try to please internalized
audiences (stealing the intrapsychic prediction that defensive-
ness should disappear when people were unaccountable). Strict
scorekeepers are disinclined to permit such borrowing, sometimes
disparaging the practice as a form of cheating that violates sacro-
sanct conceptual boundaries. Tetlock and Manstead (1985) replied,
however, that investigators will just have to increase their toler-
ance of ambiguity: There are no sharp boundaries in the fuzzy-set
world of functionalist frameworks, and what looks like a transpar-
ently defensive maneuver from one point of view may later gen-
erate a string of discoveries that will be credited to the vibrant
problem-solving power of the research program.
Boundary blurring of this sort is already apparent in reductionist
challenges to social functionalism that treat accountability effects
as merely response threshold adjustments or by-products of drive
or arousal states. It has already been shown that considerable
conceptual stretching was required to assimilate social functional-
ist frameworks into mainstream ones. The analysis also suggests
that observers will often disagree on when stretching has been
taken to the breaking pointdisagree for the simple reason that,
for any given guiding functionalist metaphor, there is a substantial
range of permissible theories specifying how well or poorly people
cope with the designated adaptive challenges. The resulting fuzzi-
ness of category boundaries guarantees that, in the domain of
467
FUNCTIONALIST PLURALISM
judgment and choice, most stretching exercises will neither clearly
succeed nor fail but rather fall in the gray zone of indefinite
intellectual stalemate. Moreover, the ultimate arbitrariness of these
boundaries cannot be concealed by quasi-formal algorithms for
calculating whether a research program has become degenera-
tive by, say, dividing the frequency with which the negative
heuristicof the program is invoked to deal with anomalies by the
frequency with which the positive heuristic makes boldly precise
predictions that turn out to be accurate (cf. Meehl, 1990).
A Compromise Position: Functionalist Pluralism
The argument thus far can be briefly summarized: (a) Waiting
for the grand reduction of research programs to a common con-
ceptual language will turn out to be a waiting-for-Godot exercise;
(b) the undisciplined proliferation of research programs is an
invitation to intellectual anarchy (a conclusion that thoughtful
relativists never denied; Feyerabend, 1993); and (c) the quest for
crucial experiments is tempting given the professional incentives
confronting young scholars but is unlikely to yield anything much
more conclusive than has already been accomplished by traversing
this path. The metatheoretical position taken here, functionalist
pluralism, can be viewed as a form of compromise among these
options. It concedes to relativists the need to acknowledge multiple
possible benchmarks of rationality (although it insists on con-
straints on the number of permissible standards), it concedes to
reductionists the usefulness of documenting the precise psycho-
logical and eventually neurological mechanisms underlying the
activation and implementation of functionalist mind-sets (although
it insists on the inevitable incompleteness of accounts that do not
recognize the external adaptive challenges that functionalist mind-
sets are directed to meet), and it concedes to falsificationists the
usefulness of documenting boundary conditions on the applicabil-
ity of rival functionalist accounts (although it insists on the inev-
itability of boundary blurring given the variety of theoretical
instantiations that can be advanced for each guiding functionalist
metaphor).
From the standpoint of functionalist pluralism, mixed metaphors
may be ugly but they are the next step in theory building. The
looming challenge will be to develop viable mixed-metaphor mod-
els of how people resolve conflicts among functionalist impera-
tives. A useful first step is to specify more precisely the properties
of the self-regulatory machinery to be integrated.
One promising approach, drawing on work on category acces-
sibility (Higgins, 1996) and personsituation interactionism (Mis-
chel & Shoda, 1995), is to treat functionalist orientations as emo-
tionally charged sets of production rules that specify the contextual
preconditions that must be satisfied to activate particular coping
strategies and goal states. The relative importance of functionalist
orientations should, in this view, hinge on the momentary salience
of situational cues for priming mind-sets as well as on the chronic
accessibility of these mind-sets. Cross-functional conflicts should
be especially intense when two functions are roughly equally
accessible and activate processing rules that lead to contradictory
conclusions.
Even at this early stage of articulating social functionalist the-
ories, it is possible to identify numerous contexts in which cross-
functional conflicts seem to arise. Intuitive statisticians seek pre-
dictively useful cues but stop when doing so undercuts their
self-image as racially egalitarian. Intuitive prosecutors consider
extenuating circumstances but stop when the social order seems to
be under siege. People defend the normative order but stop when
the rules start to feel intolerably oppressive. Intuitive economists
seek to maximize expected utility but stop when utility maximi-
zation requires monetizing sacred values that they are supposed to
treat as infinitely important. People want the approval of key
constituencies, but some draw the line at compromising basic
convictions. The general theme seems to be cybernetic: People
discover they have had enough of a functionalist mind-set only
after they have had more than enough.
Kunda (1999) offered one framework for modeling how people
deal with such conflicts. Like other theorists (e.g., Kruglanski,
1990), she has focused on the tension between nondirectional
accuracy goals (the quest for the truth that directs the thought
processes of ideal-type intuitive scientist) and directional goals
(the desire to reach a foreordained conclusion). Directional goals
can bias encoding, retrieval, and subsequent interpretation, but
there are plausibility constraints on the ability of noninstitutional-
ized adults to draw whatever conclusions they want. Even when
they are swept along by a motivated current of thought, people
keep in mind the countervailing need to reach conclusions that
they could justify to what Kunda called a dispassionate observer
(the homunculus who personifies the accuracy motive). Kundas
proposal assigned a mental-watchdog status to the intuitive scien-
tist: prevent people under the sway of directional goals from
slipping into self-serving delusional worlds.
An alternative framework, however, turns this argument on its
head. Personifications of social directional goalsthe theologian,
prosecutor, and politicianmight just as easily serve as watchdogs
on the mental operations of the intuitive scientist and economist,
checking lines of thought that undercut social values. For example,
one could even go so far as to assign metacognitive primacy to the
intuitive politician. People do not care about accuracy per se; they
care only about justifiability, and justifiability is a profoundly
relational construct that hinges on the identity of the audience and
its evaluative standards. Kundas dispassionate observer is a rare
special case of justifiability that arises when the evaluative audi-
ence cares solely about procedural fairness or rationality.
Of course, the choice is not eitheror. One could argue that both
monitoring processes need to operate concurrently for people to
adjust reasonably to the surrounding social world. Pressures to
balance conflicting functionalist imperatives could be formally
captured by parallel-constraint-satisfaction models in which (a)
each functionalist orientation is a network structure with informa-
tional elements (e.g., beliefs, values) and relations of support or
conflict between elements that can have varying intensity and (b)
the psycho-logic of constraint satisfaction allows each element to
simultaneously influence and be influenced by all other elements
to which it is directly or indirectly connected. These simulations
often converge on stable asymptotic balances that maximize the
consistency of the entire network (cf. Read & Marcus-Newhall,
1993).
The battle for explanatory primacy will obviously not suddenly
end with the adoption of an integrative agenda. Multifunctional
frameworks must still make potentially tendentious priority assign-
ments to functions and their modes of implementation. Function-
alist pluralists insist, however, that these disputes are best viewed
as matters of degree, not of kind. The relative importance of
468
TETLOCK
mind-sets varies across individuals and contexts, and research
effort is better devoted to (a) documenting these moderators and
(b) shedding light on the self-regulatory dynamics of shifting from
one mind-set to another and of striking compromises between
mind-sets. This analysis fits with yet another, albeit higher order,
political metaphor. Functionalist pluralists model the mind as a
polyglot polity populated by semiautonomous functional fiefdoms,
each with its own operating principles, in uneasy coexistence. They
tend to be suspicious of monistic proposals, such as Greenwalds
(1980) totalitarian ego, viewing them as special cases in which a
particular set of functionsthose stressing cognitive order, conti-
nuity, and self-image protectionhave achieved self-regulatory
primacy. For most people, most of the time, the human psyche is
a patchwork quilt of ever-shifting functionalist compromises.
Skeptics might counter that mixed-metaphor theories take on the
Promethean aspirations of classic theories of personality. Not
enough is known about the psychological implementation of spe-
cific functions to merge these diverse facets of human nature.
However, the counterclaim is that these functionalist facets blur
into each other. Unifunctionalist tunnel vision blinds the research
community to empirical and normative boundary conditions on
basic effects. Inconvenient though it is, people are multifunctional
entities that demand cumbersomely complex explanations.
References
Aberle, D. F., Cohen, A. K., Davis, A. K., Levy, M. J., Jr., & Sutton, F. X.
(1951). The functional prerequisites of a society. Ethics, 3, 60111.
Allport, G. W. (1985). The historical background of social psychology. In
G. Lindzey & E. Aronson (Eds.), The handbook of social psychology
(Vol. 1, pp. 356). Cambridge, MA: Addison-Wesley. (Original work
published 1968)
Andersen, S. M., & Glassman, N. S. (1996). Responding to significant
others when they are not there: Effects on interpersonal inference,
motivation, and affect. In R. M. Sorrentino & E. T. Higgins (Eds.),
Handbook of motivation and cognition (pp. 263321). New York:
Guilford.
Ashton, H. R. (1992). Effects of justification and a mechanical aid on
judgment performance. Organizational Behavior and Human Decision
Processes, 52, 292306.
Baron, J., Gowda, R., & Kunreuther, H. (1993). Attitudes toward managing
hazardous waste: What should be cleaned up and who should pay for it.
Risk Analysis, 13, 183201.
Baron, J., & Ritov, I. (1993). Intuitions about penalties and compensation
in the context of tort law. Journal of Risk and Uncertainty, 7, 1735.
Baumeister, R. F. (1998). The self. In D. T. Gilbert, S. Fiske, & G. Lindzey
(Eds.), Handbook of social psychology (Vol. 1, pp. 680740). New
York: McGraw-Hill.
Baumeister, R. F., & Leary, M. F. (1995). The need to belong: Desire for
interpersonal attachments as a fundamental human motive. Psychologi-
cal Bulletin, 117, 497529.
Benartzi, S., & Thaler, R. H. (1995). Myopic loss aversion and the equity
premium puzzle. Quarterly Journal of Economics, 110, 7392.
Boring, E. G. (1963). History, psychology, and science: Selected papers.
New York: Wiley.
Brunsson, N. (1989). The organization of hypocrisy: Talk, decisions, and
actions in organizations. New York: Wiley.
Burger, J. M. (1981). Motivational biases in the attribution of responsibil-
ity: A meta-analysis of the defensive-attribution hypothesis. Psycholog-
ical Bulletin, 90, 496512.
Chaiken, S., & Trope, Y. (1999). Dual-process theories in social psychol-
ogy. New York: Guilford Press.
Cosmides, L., & Tooby, J. (1994). Origins of domain specificity: The
evolution of functional organization. In L. A. Hirschfeld & S. Gelman
(Eds.), Mapping the mind: Domain specificity in cognition and culture
(pp. 85116). New York: Cambridge University Press.
Curley, S. P., Yates, J. F., & Abrams, R. A. (1986). Psychological sources
of ambiguity avoidance. Organizational Behavior and Human Decision
Processes, 38, 230256.
Durkheim, E. (1976). The elementary forms of the religious life (2nd ed.).
London: Allen & Unwin. (Original work published 1925)
Edgerton, R. B. (1985). Rules, exceptions, and social order. Berkeley:
University of California Press.
Edwards, W. (1962). Subjective probabilities inferred from decisions.
Psychological Review, 69, 109135.
Feyerabend, P. (1993). Against method. New York: Verso.
Fiske, A. P. (1991). Structures of social life: The four elementary forms of
social relations. New York: Free Press.
Fiske, A. P., & Tetlock, P. E. (1997). Taboo trade-offs: Reactions to
transactions that transgress spheres of justice. Political Psychology, 18,
255297.
Forgas, J. P. (1995). Mood and judgment: The affect infusion model.
Psychological Bulletin, 117, 3966.
Friedrich, J. (1993). Primary error detection and minimization strategies in
social cognition: A reinterpretation of confirmation bias phenomena.
Psychological Review, 100, 289319.
Geertz, C. (2000). Local knowledge: Further essays in interpretive anthro-
pology. New York: Basic Books.
Gergen, K. J. (1986). Correspondence versus autonomy in the language of
human understanding. In D. W. Fiske & R. Schweder (Eds.), Metatheory
in social sciences (pp 136162). Chicago: University of Chicago Press.
Gigerenzer, G., Todd, P., & The ABC Research Group. (1999). Simple
heuristics that make us smart. New York: Oxford University Press.
Gilbert, D. T. (1991). How mental systems believe. American Psycholo-
gist, 46, 107119.
Gilovich, T., Griffin, D. W., & Kahneman, D. (2002). Inferences, heuris-
tics and biases: New directions in judgment under uncertainty. New
York: Cambridge University Press.
Gilovich, T., Medvec, V. H., & Savitsky, K. (2000). The spotlight effect in
social judgment: An egocentric bias in estimates of the salience of ones
own actions and appearance. Journal of Personality and Social Psychol-
ogy, 78, 211222.
Glucksberg, S., & McGlone, M. S. (1999). When love is not a journey:
What metaphors mean. Journal of Pragmatics, 31, 15411558.
Goldberg, J. H., Lerner, J. S., & Tetlock, P. E. (1999). Rage and reason:
The psychology of the intuitive prosecutor. European Journal of Social
Psychology, 29, 781795.
Green, M., Visser, P., & Tetlock, P. E. (2000). Coping with accountability
cross-pressures: Low-effort evasive tactics and high-effort quests for
complex compromises. Personality and Social Psychology Bulletin, 26,
13801391.
Greenberg, J., Pyszcynski, T., Solomon, S., Simon, L., & Breus, M. (1994).
Role of consciousness and accessibility of death-related thoughts in
morality salience effects. Journal of Personality and Social Psychology,
67, 627637.
Greenwald, A. G. (1980). The totalitarian ego: Fabrication and revision of
personal history. American Psychologist, 35, 603618.
Hagafors, R., & Brehmer, B. (1983). Does having to justify ones judg-
ments change the nature of the judgment process? Organizational Be-
havior and Human Performance, 31, 223232.
Haidt, J. (2001). The emotional dog and its rational tail: A social intuition-
ist approach to moral judgment. Psychological Review, 108, 814834.
Hamilton, V. L. (1980). Intuitive psychologist or intuitive lawyer: Alter-
native models of the attribution process. Journal of Personality and
Social Psychology, 39, 767773.
Haselton, M. G., & Buss, D. M. (2000). Error management theory: A new
469
FUNCTIONALIST PLURALISM
perspective on biases in cross-sex mind reading. Journal of Personality
and Social Psychology, 78, 8191.
Heider, F. (1958). The psychology of interpersonal relations. New York:
Wiley.
Higgins, E. T. (1989). Self-discrepancy theory. In L. Berkowitz (Ed.),
Advances in experimental social psychology (Vol. 22, pp. 93136). New
York: Academic Press.
Higgins, E. T. (1996). Knowledge activation. In E. T. Higgins & A.
Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp.
133168). New York: Guilford Press.
Hirschfeld, L. A., & Gelman, S. A. (1994). Mapping the mind: Domain
specificity in cognition and culture. New York: Cambridge University
Press.
Horowitz, D. L. (1985). Ethnic groups in conflict. Berkeley: University of
California Press.
Jacobs, J., & Potter, K. (1998). Hate crimes: Criminal law and identity
politics. New York: Oxford University Press.
James, W. (1983). The principles of psychology. Cambridge, MA: Harvard
University Press. (Original work published 1890)
Kahn, B. F., & Baron, J. (1995). An exploratory study of choice rules
favored for high-stakes decisions. Journal of Consumer Psychology, 4,
305328.
Kahneman, D., & Tversky, A. (2001). Choice, values, and frames. New
York: Cambridge University Press.
Kelley, H. H. (1971). Attribution in social interaction. In E. E. Jones, D. W.
Kanouse, H. H. Kelley, R. E. Nisbett, S. Valins, & B. Weiner (Eds.),
Attribution: Perceiving the causes of behavior (pp. 121). Morristown,
NJ: General Learning Press.
Kelley, H. H., & Thibaut, J. W. (1978). Interpersonal relations: A theory
of interdependence. New York: Wiley.
Keltner, D., & Buswell, B. N. (1997). Embarrassment: Its distinct form and
appeasement functions. Psychological Bulletin, 122, 250270.
Kennedy, J. (1993). Debiasing audit judgment and accountability: A frame-
work and experimental results. Journal of Accounting Research, 31,
231245.
Kruglanski, A. W. (1990). Lay epistemics and human knowledge. New
York: Plenum.
Kruglanski, A. W., & Freund, T. (1983). The freezing and unfreezing of
lay-inferences: Effects on impressional primacy, ethnic stereotyping,
and numerical anchoring. Journal of Experimental Social Psychology,
19, 448468.
Kruglanski, A. W., & Webster, D. (1996). Motivated closing of the mind:
Seizing and freezing. Psychological Review, 103, 263268.
Kunda, Z. (1999). Social cognition: Making sense of people. Cambridge,
MA: MIT Press.
Lakatos, I. (1970). Falsification and the methodology of scientific research
programs. In I. Lakatos & A. Musgrave (Eds.), Criticism and the growth
of knowledge (pp. 8101). Cambridge, England: Cambridge University
Press.
Laudan, L. (1997). Progress and its problems. Berkeley: University of
California Press.
Lerner, J. S., Goldberg, J. H., & Tetlock, P. E. (1998). Sober second
thought: The effects of accountability, anger and authoritarianism on
attributions of responsibility. Personality and Social Psychology Bulle-
tin, 24, 563574.
Lerner, J. S., & Tetlock, P. E. (1999). Accounting for the effects of
accountability. Psychological Bulletin, 125, 255275.
Lerner, M. J., & Lerner, S. C. (1981). The justice motive in social behavior:
Adapting to times of scarcity and change. New York: Plenum Press.
McGuire, W. J. (1983). A contextualist theory of knowledge: Its implica-
tions for innovation and reform in psychological research. In L. Berko-
witz (Ed.), Advances in experimental social psychology (Vol. 16, pp.
147). New York: Academic Press.
Mead, G. H. (1934). Mind, self, and society. Chicago: University of
Chicago Press.
Meehl, P. (1990). Appraising and amending theories: The strategy of
Lakatosian defense and two principles that warrant it. Psychological
Inquiry, 1, 108141.
Mero, N., & Motowidlo, S. (1995). Effects of rater accountability on the
accuracy and favorability of performance ratings. Journal of Applied
Psychology, 80, 517524.
Miller, D. T., & Vidmar, N. (1981). The social psychology of punishment
reactions. In M. J. Lerner & S. C. Lerner (Eds.), The justice motive in
social behavior (pp. 145168). New York: Plenum Press.
Miller, N. E., & Dollard, J. (1941). Social learning and imitation. New
Haven, CT: Yale University Press.
Mischel, W., & Shoda, Y. (1995). A cognitiveaffective system theory of
personality: Reconceptualizing situations, dispositions, dynamics, and
invariance in personality structure. Psychological Review, 102, 246
268.
Murphy, R. (1994). The effects of task characteristics on covariation
assessment: The impact of accountability and judgment frame. Organi-
zational Behavior and Human Decision Processes, 60, 139155.
Nagel, E. (1961). The structure of science: Problems in the logic of
scientific explanation. New York: Harcourt, Brace, & World.
Nisbett, R. E., & Cohen, D. (1996). Culture of honor: The psychology of
violence in the South. Boulder, CO: Westview Press.
Pelham, B. W., & Neter, E. (1995). The effect of motivation of judgment
depends on the difficulty of the judgment. Journal of Personality and
Social Psychology, 68, 581594.
Peng, K., & Nisbett, R. (1999). Culture, dialectics, and reasoning about
contradiction. American Psychologist, 51, 252264.
Pennington, J., & Schlenker, B. R. (1999). Accountability for consequen-
tial decisions: Justifying ethical judgments to audiences. Personality and
Social Psychology Bulletin, 25, 10671077.
Pinker, S. (1997). How the mind works. New York: Norton.
Polinsky, A. M., & Shavell, S. (1999). Punitive damages: An economic
analysis. Harvard Law Review, 111, 869895.
Radcliffe-Brown, A. (1952). Structure and function in primitive society.
London: Cohen and West.
Read, S. J., & Marcus-Newhall, A. (1993). Explanatory coherence in social
explanations: A parallel distributed processing account. Journal of Per-
sonality and Social Psychology, 65, 429447.
Ridley, M. (1994). The Red Queen: Sex and the evolution of human nature.
New York: Macmillan.
Samuelson, W., & Zeckhauser, R. (1988) Status quo bias in decision
making. Journal of Risk and Uncertainty, 1, 759.
Schlenker, B. R. (1982). Translating actions into attitudes: An identity-
analytic approach to the explanation of social conduct. In L. Berkowitz
(Ed.), Advances in experimental social psychology (Vol. 15, pp. 194
248). New York: Academic Press.
Schlenker, B. R. (Ed.). (1985). The self and social life. New York:
McGraw-Hill.
Schwarz, N. (1994). Judgment in a social context. In M. P. Zanna (Ed.),
Advances in experimental social psychology (pp. 123162). San Diego,
CA: Academic Press.
Scott, M. B., & Lyman, S. (1968). Accounts. American Sociological
Review, 33, 4662.
Searle, J. (1998). The rediscovery of the mind. Cambridge, MA: MIT Press.
Semin, G. R., & Manstead, A. S. R. (1983). The accountability of conduct:
A social psychology analysis. London: Academic Press.
Shafir, E., Simonson, I., & Tversky, A. (1993). Reason-based choice.
Cognition, 49, 1136.
Siegel-Jacobs, K., & Yates, F. (1996). Effects of procedural and outcome
accountability on judgment quality. Organizational Behavior and Hu-
man Decision Processes, 1, 117.
470
TETLOCK
Simonson, I. (1989). Choice based on reasons: The case of attraction and
compromise effects. Journal of Consumer Research, 16, 158174.
Simonson, I., & Nye, P. (1992). The effect of accountability on suscepti-
bility to decision errors. Organizational Behavior and Human Decision
Processes, 51, 416446.
Simonson, I., & Staw, B. M. (1992). Deescalation strategies: A comparison
of techniques for reducing commitment to losing courses of action.
Journal of Applied Psychology, 77, 419426.
Skitka, L., & Tetlock, P. E. (1993). Providing public assistance: Cognitive
and motivational processes underlying liberal and conservative policy
preferences. Journal of Personality and Social Psychology, 65, 1205
1224.
Sniderman, P., Fletcher, J., Russell, P., & Tetlock, P. E. (1996). The clash
of rights: Liberty, equality, and legitimacy in liberal democracy. New
Haven, CT: Yale University Press.
Sniderman, P., & Tetlock, P. E. (1986). Interrelationship of political
ideology and public opinion. In M. Hermann (Ed.), Handbook of polit-
ical psychology (pp. 232260). San Francisco: Jossey-Bass.
Staw, B. (1980). Rationality and justification in organizational life. In B.
Staw & L. Cummings (Eds.), Research in organizational behavior
(Vol. 2, pp. 157). Greenwich, CT: JAI.
Steele, C. M. (1988). The psychology of self-affirmation. In L. Berkowitz
(Ed.), Advances in experimental social psychology (Vol. 21, pp. 261
302). San Diego, CA: Academic Press.
Stryker, S., & Statham, A. (1985). Symbolic interaction and role theory. In
G. Lindzey & E. Aronson (Eds.), The handbook of social psychology
(Vol. 1, pp. 311378). New York: Random House.
Sunstein, C. R., Kahneman, D., & Schkade, D. (1998). Assessing punitive
damages. Yale Law Journal, 107, 20712125.
Suppe, F. (1974). The structure of scientific theories. Urbana: University of
Illinois Press.
Taylor, K. (1975). Testing credit and blame attributions for choices under
ambiguity. Organizational Behavior and Human Performance, 64, 128
137.
Tesser, A. (2000). On the confluence of self-esteem maintenance mecha-
nisms. Personality and Social Psychology Review, 4, 290299.
Tetlock, P. E. (1983a). Accountability and complexity of thought. Journal
of Personality and Social Psychology, 45, 7483.
Tetlock, P. E. (1983b). Accountability and perseverance of first impres-
sions. Social Psychology Quarterly, 46, 285292.
Tetlock, P. E. (1985). Accountability: A social check on the fundamental
attribution error. Social Psychology Quarterly, 48, 227236.
Tetlock, P. E. (1992). The impact of accountability on judgment and
choice. In M. Zanna (Ed.), Advances in experimental social psychology
(Vol. 25, pp. 331376). New York: Academic Press.
Tetlock, P. E. (1998). Losing our religion: On the collapse of precise
normative standards in complex accountability systems. In R. Kramer &
M. Neale (Eds.), Influence processes in organizations: Emerging themes
in theory and research (pp. 168189). Thousand Oaks, CA: Sage.
Tetlock, P. E. (1999). Accountability theory: Mixing properties of human
agents with properties of social systems. In J. Levine, L. Thompson, &
D. Messick (Eds.), Shared cognition in organizations: The management
of knowledge (pp. 255278). Hillsdale, NJ: Erlbaum.
Tetlock, P. E. (2000a). Cognitive biases and organizational correctives: Do
both disease and cure depend on the political beholder? Administrative
Science Quarterly, 45, 293326.
Tetlock, P. E. (2000b). Coping with trade-offs. In S. Lupia, M. McCubbins,
& S. Popkin (Eds.), Political reasoning and choice (pp. 237262).
Berkeley: University of California Press.
Tetlock, P. E., & Boettger, R. (1989). Accountability: A social magnifier
of the dilution effect. Journal of Personality and Social Psychology, 57,
388398.
Tetlock, P. E., & Boettger, R. (1994). Accountability amplifies the status
quo effect when change creates victims. Journal of Behavioral Decision
Making, 7, 123.
Tetlock, P. E., & Kim, J. (1987). Accountability and judgment in a
personality prediction task. Journal of Personality and Social Psychol-
ogy: Attitudes and Social Cognition, 52, 700709.
Tetlock, P. E., Kristel, O., Elson, B., Green, M., & Lerner, J. (2000). The
psychology of the unthinkable: Taboo trade-offs, forbidden base rates,
and heretical counterfactuals. Journal of Personality and Social Psy-
chology, 78, 853870.
Tetlock, P. E., & Lerner, J. (1999). The social contingency model: Iden-
tifying empirical and normative boundary conditions on the error-and-
bias portrait of human nature. In S. Chaiken & Y. Trope (Eds.), Dual
process models in social psychology (pp. 542551). New York: Guilford
Press.
Tetlock, P. E., & Levi, A. (1982). Attribution bias: On the inconclusiveness
of the cognitionmotivation debate. Journal of Experimental Social
Psychology, 18, 6888.
Tetlock, P. E., & Manstead, A. S. R. (1985). Impression management
versus intrapsychic explanations in social psychology: A useful dichot-
omy? Psychological Review, 92, 5977.
Tetlock, P. E., Peterson, R., & Lerner, J. (1996). Revisiting the value
pluralism model: Incorporating social content and context postulates. In
C. Seligman, J. Olson, & M. Zanna (Eds.), Ontario Symposium on Social
and Personality Psychology: Values (pp. 3764). Hillsdale, NJ: Erlbaum.
Tetlock, P. E., Skitka, L., & Boettger, R. (1989). Social and cognitive
strategies for coping with accountability: Conformity, complexity, and
bolstering. Journal of Personality and Social Psychology, 57, 632640.
Tetlock, P. E., Visser, P., Singh, R., Polifroni, M., Scott, A., Elson, B., &
Mazzocco, P. (2002). People as intuitive prosecutors: The fair-but-
biased-and-occasionally-erratic model. Unpublished manuscript, Uni-
versity of California, Berkeley.
Thompson, E. P., Roman, R. J., Moskowitz, G. B., Chaiken, S., & Bargh,
J. A. (1994). Accuracy motivation attenuates covert priming: The sys-
tematic reprocessing of social information. Journal of Personality and
Social Psychology, 66, 474489.
Thompson, L. (1995). They saw a negotiation: Partisanship and involve-
ment. Journal of Personality and Social Psychology, 68, 839853.
Thompson, L. (2001). The mind and heart of the negotiator. Englewood
Cliffs, NJ: Prentice-Hall.
Tyler, T. R., & Smith, H. J. (1998). Social justice and social movements.
In D. Gilbert, S. Fiske, & G. Lindzey (Eds.), Handbook of social
psychology (Vol. 2, pp. 595629). New York: McGraw-Hill.
Vygotsky, L. S. (1978). Mind in society. Cambridge, MA: Harvard Uni-
versity Press. (Original work published 1934)
Wegener, D. T., & Petty, R. E. (1995). Flexible correction processes in
social judgment. Journal of Personality and Social Psychology, 68,
3651.
Zajonc, R. B. (1965, July 16). Social facilitation. Science, 149, 269149.
Received March 9, 2000
Revision received March 19, 2001
Accepted July 31, 2001
471
FUNCTIONALIST PLURALISM