The Role of Construction, Intuition, and Justification in Responding to Ethical Issues at Work: The Sense making-Intuition Model

The Role of Construction, Intuition, and Justification in Responding to Ethical Issues at Work: The Sense making-Intuition Model

The Role of Construction, Intuition, and Justification in Responding to Ethical Issues at Work: The Sense making-Intuition Model

Proponents of a popular view of how individuals respond to ethical issues at work claim that individuals use deliberate and extensive moral reasoning under conditions that ignore equivocality and uncertainty. I discuss the limitations of these “rationalist approaches” and reconsider their empirical support using an alternative explanation from social psychological and sensemaking perspectives. I then introduce a new theoretical model composed of issue construction, intuitive judgment, and post hoc explanation and justification. I discuss the implications for management theory, methods, and practice.

Several prominent theories claim that individ uals use deliberate and extensive moral reason ing to respond to ethical issues, such as weigh ing evidence and applying abstract moral principles. These “rationalist approaches” have flourished, in part, because of their cumulative research agenda and the absence of well developed alternative theoretical perspectives (Randall & Gibson, 1990). Despite their popular ity and usefulness, it is important to evaluate these approaches to understand their limita tions. I question several assumptions of ratio nalist approaches and answer scholars’ calls to develop alternative theoretical views (OTallon & Butterfield, 2005). I present a model based on social psychological and sensemaking perspec tives?something I call the “sensemaking intuition model” (SIM).

I argue that individuals engage in sensemak ing under conditions of equivocality and uncer tainty (Weick, 1979, 1995). Individuals’ expecta tions and motivations affect this process such that they vary in how they construct ethical is sues. Individuals then make intuitive judgments about their constructions of ethical issues. This view challenges the privileged status of moral

reasoning in rationalist models by claiming that responses to ethical issues are not always based on deliberate and extensive moral rea soning. Although in previous research scholars have found partial support for rationalist ap proaches, I reconsider these findings in light of an alternative explanation using social psycho logical and sensemaking perspectives.

I first briefly review rationalist approaches and then discuss the limitations of this view. I then outline the alternative assumptions of my approach and describe the key components of the SIM: issue construction, intuitive judgment, and explanation and justification. I conclude by detailing the implications of this research for theory, methods, and practice.

REVIEW OF RATIONALIST APPROACHES

Philosophers back to at least Plato took a strong interest in how individuals respond to ethical is sues, but social scientists have only recently be gun to focus on responses to ethical issues in the context of business. I briefly review three promi nent streams of research on how organizational actors respond to ethical issues?managers as “philosophers,” person-situation, and issue contingent approaches?and consider these as examples of rationalist approaches.

Managers As Philosophers Research Some of the earliest models of individuals’

responses to ethical issues borrowed directly

I thank Kathie Sutcliffe for her valuable feedback on this paper. I also thank Jane Dutton, Adam Grant, Patricia Wer hane, Phoebe Ellsworth, Marlys Christianson, former asso ciate editor Tom Donaldson, and the AMR reviewers for their comments on a previous draft. I presented an earlier version of this paper at the Society for Business Ethics 2004 annual meeting in New Orleans.

1022 Copyright o? the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s express written permission. Users may print, download, or email articles for individual use only.

This content downloaded from 131.170.21.110 on Thu, 30 Jan 2020 08:53:26 UTC All use subject to https://about.jstor.org/terms

2007 Sonenshein 1023

from traditional philosophical ethics. These views propose that managers, much like philos ophers, draw on normative theories to make de cisions about ethical issues. As a result, propo nents claim that individuals often use deliberate and extensive moral reasoning. For example, Fritzsche and Becker (1984) presented individuals with a series of vignettes and asked them how they would respond to a morally ques tionable behavior, and why (see also Premeaux, 2004, and Premeaux & Mondy, 1993). The re searchers classified individuals’ explanations into major schools of philosophy, such as deon tological and utilitarian reasoning. Hunt and Vitell (1986; see also Mayo & Marks, 1990) cre ated a similar model, in which “the ultimate underlying assumption is that people … do in fact engage in both deontological and teleolog ical evaluations in determining their ethical judgments, and ultimately, their behaviors” (Hunt & Vitell, 1986: 7). Both forms of reasoning involve ordinary organizational actors’ use of rather intricate philosophical theories to reason and make calculations about ethical issues. De ontological reasoning involves comparisons be tween potential behaviors and predetermined deontological norms (e.g., honesty, fair treat ment, etc.). Teleological evaluations involve cal culations, such as perceived consequences of alternative actions for stakeholder groups (Hunt & Vitell, 1986: 9).

Person-Situation Research

Trevi?o (1986) raised the possibility that indi viduals may not actually use philosophical the ories as the basis of moral reasoning; she more deeply rooted her theory in social science. Her “person-situation model” (see also Church, Gaa, Nainar, & Shehata, 2005) proposes that re sponses to ethical issues are a function of (1) a person’s cognitions determined by his or her stage of moral development (Kohlberg, 1981, 1984), (2) individual-difference moderators (ego strength, field dependence, and locus of control), and (3) situational moderators (organizational culture and characteristics of the work). While the person-situation model drops an explicit fo cus on philosophical reasoning, it retains a ra tionalist foundation by treating a manager’s stage of moral development as an important explanatory device in how individuals respond to ethical issues (Trevi?o, 1986: 604). For exam

pie, Trevi?o and Youngblood (1990) provided a management decision-making exercise for MBA students and found that individuals’ responses to ethical issues are affected by their reasoning processes, as well as by rewards and punish ments.

Issue-Contingent Research

Building on both the managers as philoso phers and person-situation models, Jones (1991) published an “issue-contingent model,” which calls attention to the importance of the charac teristics of ethical issues. Jones posits that the “moral intensity” of an issue explains how peo ple will react to ethical issues. Moral intensity is composed of differences in consequences (harms or benefits done to victims/beneficia ries), social consensus (degree of social agree ment that a proposed act is evil/good), probabil ity of effect (joint probability the act in question will happen and the predicted harm/benefit will occur), temporal immediacy (distance between the present and the onset of the act’s conse quences), proximity (feeling near to the victim/ beneficiary), and concentration of effect (the in verse function of the number of individuals affected by the act’s magnitude). One of the central claims of Jones’ (1991)

model, following Rest (1986), is that responses to ethical issues contain four components: recogni tion of a moral issue, ethical judgment, moral intent, and ethical behavior. All four compo nents are affected by an issue’s moral intensity. As an issue’s moral intensity increases, individ uals are more likely to recognize moral issues, make ethical judgments, establish moral intent, and engage in ethical behavior. Empirical re search shows that the social consensus and magnitude of consequences components receive the strongest support, with other factors receiv ing mixed support (O’Fallon & Butterfield, 2005).

LIMITATIONS OF RATIONALIST APPROACHES

Rationalist approaches have been useful for advancing scholarly understanding of how indi viduals respond to ethical issues. Unlike much of the work on ethical issues (Randall & Gibson, 1990), rationalist models are grounded using clear and testable theoretical claims that allow for theory extension. Not surprisingly, a recent review found 174 empirical articles where the

This content downloaded from 131.170.21.110 on Thu, 30 Jan 2020 08:53:26 UTC All use subject to https://about.jstor.org/terms

1024 Academy of Management Review October

dependent variable was one of the four stages of Rest’s (1986, and, by extension, Jones’ [1991]) model, including 32 using moral intensity (O’Fallon & Butterfield, 2005). Despite this atten tion, rationalist perspectives have important limitations that scholars rarely surface, in part because of a lack of theoretical alternatives (O’Fallon & Butterfield, 2005; Randall & Gibson, 1990).

In this section I discuss four limitations. Ra tionalist approaches tend to (1) fail to address the presence of equivocality and uncertainty common in natural settings, (2) view deliberate and extensive reasoning as a precursor for eth ical behavior, (3) underemphasize the construc tive nature of “ethical issues,” and (4) claim that

moral reasoning is used to make moral judg ments. Although not all three theories I re viewed above are subject to all four criticisms, they are historically intertwined with one an other and collectively illustrate the rationalist approach by emphasizing deliberate and exten sive reasoning and reflection.

Equivocality and Uncertainty

Equivocality involves the existence of several different, simultaneous interpretations (Weick, 1995: 91f). Uncertainty refers to a lack of informa tion that makes constructing a plausible inter pretation about a situation difficult. For exam ple, individuals may have imprecise estimates about the consequences of their current actions on the future (March, 1994), such as if committing an action will harm others.

Equivocality and uncertainty are common themes in organizational theory, including

Weick’s theory of organizing (1979, 1995) and Dutton’s (1997) view of strategic issues. While these views of equivocality and uncertainty

were developed outside the specific domain of ethical issues, there are likely to be important applications to ethical issues. For example, Mc Caskey (1982) has proposed several factors that lead to equivocal situations. His “differences in value orientations” is something especially sa lient for moral problems given moral pluralism, and “unclear or conflicting goals” often arise when individuals are forced to choose between right and right (Badaracco, 1997) and need to satisfy conflicting stakeholder needs (Phillips, 2003).

While organizational scholars have histori cally recognized the importance of equivocality and uncertainty, some rationalist models fail to address these concepts?specifically, those based on issue-contingent approaches. To illus trate this point, I examine the two components of

moral intensity that receive the most attention (and support) in the literature: (1) magnitude of consequences and (2) social consensus (May & Pauli, 2002).

Jones provides the following example for magnitude of consequences: an “act that causes 1,000 people to suffer a particular injury is of greater magnitude of consequence than an act that causes 10 people the same injury” (Jones, 1991: 374). This example assumes that whether 1,000 people or 10 people suffer, and to what degree, are known prior to individuals’ engag ing in (un)ethical behavior. But the hard part about ethical issues is often figuring out exactly what will happen from different (real or poten tial) behavioral responses. When individuals

make guesses about such outcomes, they may have a positive illusion that prevents them from recognizing that their decision will even harm others (Taylor & Brown, 1988). Alternatively, in dividuals may develop several interpretations about the issue, which vary in the degree to which an act will have harmful consequences. In the former case, an absence of any interpre tation involving harm exists, leading to uncer tainty. In the latter case, multiple interpreta tions about potential harms lead to equivocality. A second example comes from Jones’ idea of

social consensus: “the evil involved in bribing a customs official in Texas has greater social con sensus than the evil involved in bribing a cus toms official in Mexico” (1991: 375). Others have pointed out that this conception of social con sensus is a “relatively objective issue-related factor” (Butterfield, Trevi?o, & Weaver 2000: 990), implying that the issue varies, not its interpre tation. But individuals are likely to vary in how they understand a social consensus. An individ ual who has already committed a bribe in Mex ico (and got caught) may infer a different social consensus than someone who did not get caught. Other individuals may not have any in terpretation (i.e., they are uncertain) about the social consensus of bribery in Mexico compared to Texas. They might try to make predictions about the results of a poll of the relevant refer ent group on their thoughts of offering a bribe in

This content downloaded from 131.170.21.110 on Thu, 30 Jan 2020 08:53:26 UTC All use subject to https://about.jstor.org/terms

2007 Sonenshein 1025

either Texas or Mexico. But what is the relevant referent group anyway?workgroup, organiza tion, industry, society? Do the referent groups provide conflicting answers that create equivo cality?

In addition to equivocality and uncertainty about the “facts” of an issue, equivocality and uncertainty may exist in the abstract rules and principles used in moral reasoning. Kohlberg’s (1981,1984) conventional stage of moral develop

ment?where he thinks most adults peak? stresses the importance of adopting the moral standards of others, including society. Kohlberg is less concerned with the specific content of those standards and focuses more on individu als’ orientations to rules. But from a practical standpoint, such standards are not always known and clear to individuals, and there may even be conflicts (e.g., general social norms ver sus a local community). Whose standards apply in these cases? While Kohlberg implies that de ontological reasoning is more advanced than utilitarian moral reasoning (Kohlberg, 1981, 1984), this itself is a controversial philosophical and psychological claim that does not seem readily resolvable (Thomas, 1997). Even assum ing individuals agree on a set of moral stan dards, figuring out what these standards mean, along with their application, is often so difficult (Sonenshein, 2005; Walzer, 1987) that it may only rarely happen inside organizations.

Ethical Behavior Requires Deliberate and Extensive Reasoning

Rationalist approaches assume that deliber ate and extensive reasoning is required to en gage in ethical behavior. For example, Street, Douglas, Geiger, and Martinko’s (2001) “cogni tive elaboration model” posits that when both the motivation and ability to engage in moral reasoning are low, individuals will be less likely to recognize the ethical implications of issues. But when such reasoning is high, issues

will be resolved in accordance with rationalist models (e.g., Jones, 1991; Rest, 1986). In short, the cognitive elaboration model claims that deliber ate and extensive reasoning is a prerequisite for moral awareness and subsequent (actual or po tential) behavioral responses consistent with a

moral viewpoint (see also Moore & Loewenstein, 2004).

What remains an open question is how fre quently individuals use deliberate and exten sive reasoning. After all, research in psychology posits important boundaries to individuals’ cog nitive capacities (Simon, 1955), and business ethics scholars make related claims about indi viduals’ “bounded moral rationality” (Donald son & Dunfee, 1994). Moreover, social psycholog ical research shows that individuals rarely engage in the deliberate, extensive reasoning proposed by rationalist models (Bargh & Char trand, 1999; Greenwald & Banaji, 1995). Instead, individuals engage in mental processes outside their conscious awareness and guidance (Bargh & Chartrand, 1999).

This research, which is beginning to trickle into discussions about business ethics (Bazer

man & Banaji, 2004), challenges the idea that individuals reason deliberately and with great effort to respond to ethical issues (Banaji, Bazer man, & Chugh, 2003). For example, the “bounded personal ethics” model claims that individuals are often unaware of the ethical implications of their actions, which leads them to favor their self-interest (Murnighan, Cantelon, & Elyashiv, 2001). But the current literature shows a general belief that satisfactory responses (actual or po tential behaviors consistent with a moral view point) cannot emerge from anything other than deliberate and extensive reasoning (Moore & Loewenstein, 2004; Street et al., 2001).

Objectivity, Construction, and Interpretation

Issue-contingent models are predicated on the idea that previous research does not do more “than hint that characteristics of the moral issue itself will affect the moral decision-making pro cess” (Jones 1991: 369). This observation was in tended to point out a limitation in prior re search?namely, that individuals will “decide and behave in the same manner whether the issue is the theft of a few supplies… or the release of a dangerous product” (Jones 1991:371). However, to make this claim, it became neces sary to emphasize the properties of issues (i.e., moral intensity [Jones, 1991]). In fact, rationalist approaches sometimes suggest that the “pro cess begins with a problem” (Jones, 1991: 380) and that “the individual reacts to an ethical dilemma” (Trevi?o, 1986: 602). This emphasizes the properties of issues, which exist prior to (and independent of) individuals’ actions. Whether

This content downloaded from 131.170.21.110 on Thu, 30 Jan 2020 08:53:26 UTC All use subject to https://about.jstor.org/terms

1026 Academy of Management Review October

ethical issues are actually out there and indi cated by a set of objective indicators (such as moral intensity) is a deep philosophical ques tion that takes the reader into debates between realist and antirealist ontologies. My purpose is not to settle this important question here (see Fairclough, 2005, and Tsoukas, 2000). Instead, I posit the more tempered (epistemological and not ontological) claim that even if ethical issues are mind independent, individuals can still dif fer in how they interpret such issues (Best, 1995) and will do so based on their motivations and expectations. Accordingly, it is important that scholars study the interpretive processes that construct ethical issues out of social stimuli in the environment. Research shows that individuals frequently

develop subjective interpretations of issues that go beyond the objective features (assuming, for now, that such features exist) of those issues (Ross, 1987,1989; Ross & Nisbett, 1991) because of their expectations and motivations. Expecta tions, often encapsulated in scripts, place limi tations on what individuals perceive (Fiske & Taylor, 1991; Gioia, 1992). This challenges their ability to consider all relevant information when making decisions. At the same time, individuals have strong motivational drives that affect how they see social stimuli, making them very parti san actors (Pittman, 1998), often without being aware of it (Griffin & Buehler, 1993).

One way that rationalist models have tried to incorporate the idea of interpretive variation of ethical issues is with the “moral awareness” (Butterfield et al., 2000) or the “recognizing moral issues” (Jones, 1991) phase. In fact, Rest’s (1986: 5)

model acknowledges that individuals will inter pret issues differently, and it even notes that “relatively simple situations” can nevertheless lead to interpretive difficulties. While these parts of rationalist models are more consistent with the alternative I present below, moral awareness is often viewed as binary?you ei ther recognize the ethical issue or you fail to do so (Jones, 1991: 380; May & Pauli, 2002: 97). Con sequently, research has tended to focus on whether moral awareness is present or absent as a precondition for activating the other stages of rationalist models (Jones, 1991: 383), and not on how individuals construct their own interpre tations of “issues” in more nuanced ways.

To further account for differences in percep tions of issues, extensions of rationalist models

have started to move toward perceived moral intensity. Singer (1996) concludes that different respondent groups may vary in their weighting of moral intensity components. And Morris and McDonald (1995) criticize Jones (1991) for assum ing that moral intensity is issue specific as op posed to based on individuals’ perceptions.

Moral Reasoning and Moral Judgment

One of the central tenets of rationalist ap proaches is that individuals use deliberate and extensive moral reasoning to make moral judg ments about how to respond to issues (e.g., Kohl berg, 1981, 1984; Rest, 1986). This view considers individuals to be pensive moral deliberators who are “gathering facts [and] applying moral principles” (Jones, 1991: 384) in order to evaluate a specific course of behavior. Some rationalist approaches acknowledge that intuitions may arise during the early response to an issue, but they consider these “primitive cognitions … [as] poor guides for action” (Rest, 1986: 6).

Despite the prominence of deliberate and ex tensive reasoning in rationalist theories, less clear from empirical findings is the degree to which such reasoning actually occurs. For ex ample, in tests of the managers as philosophers view, researchers asked research participants to rate the likelihood they would engage in ques tionable moral behavior and then to explain why they made that decision (Fritzsche & Becker, 1984; Premeaux, 2004; Premeaux & Mondy, 1993). When respondents answered “against company policy” (Premeaux, 2004: 273), researchers interpreted this explanation as evi dence for “rule utilitarian reasoning.” But some one simply indicating an explanation for why he or she acted (e.g., against company policy) does not mean that the individual used rule utilitari anism, nor that any reasoning (generally) in formed his or her judgment (the “reasoning” could be an explanation that emerged after the fact).

The questionable empirical support for ratio nalist approaches becomes more problematic when juxtaposed with growing social psycho logical evidence that “moral reasoning is rarely the direct cause of ethical judgment” (Haidt, 2001: 815). Consistent with this argument is re cent neurological research proposing that indi viduals frequently use nonconscious pattern matching, which recognizes not only the de

This content downloaded from 131.170.21.110 on Thu, 30 Jan 2020 08:53:26 UTC All use subject to https://about.jstor.org/terms

2007 Sonenshein 1027

scriptive aspects of ethical issues but also pre scribed ways for resolving them (Reynolds, 2006). How individuals respond to issues can come from moral intuitions?immediate reac tions that contain an affective valence (good or bad; Zajonc, 1980)?without requiring conscious awareness of the process. Although scholars ob serve individuals engaging in moral reasoning, such reasoning may be an attribution that cre ates only the illusion of considered reasoning (Haidt, 2001). Put another way, post hoc attribu tions can reflect individuals’ engagement in sensemaking processes (Weick, 1995), in which individuals infer their own behavior after the fact through self-perception (Bern, 1967). Moral reasoning serves as a means for indi

viduals to explain/justify their own behavior, but it does not necessarily cause that behavior; indeed, moral reasoning may even be a conse quence of that behavior. On this account, indi viduals’ rational descriptions of their judgments can reflect normative standards of how moral decision making ought to proceed?carefully and deliberately (Kohlberg, 1981, 1984; Rest, 1986). Thus, empirical research that supports such rationalist approaches is subject to an al ternative explanation: individuals have intui tive reactions to issues but then engage in post hoc moral reasoning to explain/justify those judgments. To illustrate the idea of explanation and jus

tification, suppose that a purchasing manager is offered a “payment” to facilitate completing a contract with another company. The purchasing manager has an immediate affective valence come into her conscience (this feels bad) and then acts: she kindly turns the offer down. How ever, the purchasing manager wants to explain her behavior?why she acted the way she did. In searching for an explanation, she turns to the rational processes often prescribed in resolving moral issues, or decision making, generally (Ba zerman, 2002)?for example, abstract reasoning, the consideration of alternatives, complex com putations, and so forth. However, these rational processes occur after the purchasing manager intuitively judged that taking the bribe would be

wrong?limiting the use of moral reasoning to a way of explaining and/or justifying her re sponse. In the former case, the purchasing man ager may start reasoning about the bribe imme diately after her judgment as a way of making sense out of why that judgment entered into her

consciousness. In the latter case, the purchasing manager uses moral reasoning as a way of pro viding legitimacy to her instantaneous feeling that something was wrong.1

Summary

Rationalist models have made important con tributions to the study of how individuals re spond to ethical issues. Yet in the preceding arguments I cast doubt on some of the assump tions of these approaches. Organizational life is often equivocal and uncertain, and the very con struction of a particular issue reflects (at least in part) each individual’s expectations and motiva tions. I also challenged whether individuals al

ways use moral reasoning and moral principles to make moral judgments and engage in ethical behavior. Indeed, individuals often describe their reactions in rationalist terms, but I offered the alternative theoretical explanation that in dividuals first use intuitions and then use post hoc (moral) reasoning.

THE SIM In this section I articulate an alternative to

rationalist approaches that overcomes some of the limitations raised above. This alternative perspective, the SIM, is composed of three stages (see Figure 1 for an overview of the mod el): issue construction, intuitive judgment, and explanation and justification.

Individuals construct issues from social stim uli in equivocal and uncertain environments, and these constructions are affected by their expectations and motivations. Recall that I held in abeyance ontological claims about the mind independent properties of ethical issues and

made the more moderate epistemological claim that individuals will vary in their interpretation of ethical issues (regardless of whether or not such issues truly exist). As soon as an individual constructs an ethical issue, that individual in stantaneously makes an intuitive judgment. Such intuitions come from an individual-level factor (experience) and a collective-level factor

1 In the case that the purchasing manager has a positive affective reaction (this feels good), she will likely accept the bribe. Here her post hoc justification will likely rationalize her reaction by pointing out that there was nothing morally wrong with it (Ashforth & Anand, 2003).

This content downloaded from 131.170.21.110 on Thu, 30 Jan 2020 08:53:26 UTC All use subject to https://about.jstor.org/terms

1028 Academy of Management Review October

FIGURE 1 The Sensemaking-Intuition Model (SIM)

Expectations Motivational drives

Experience”

Issue construction

Individual

Intuitive judgment Explanation and

justification

Social pressures

Social anchors Representation

Collective

(social pressures). After this intuition emerges, an individual explains and justifies his or her response to him/herself and others. Before expli cating each phase of the model, I outline two assumptions of the SIM. I then discuss issue construction, intuitive judgment, and explana tion/justification.

Starting Assumptions

First, I assume that the SIM is triggered when individuals use sensemaking to respond to con ditions of equivocality and uncertainty (Weick, 1995, especially Chapter 4). When equivocality involves multiple interpretations, individuals engage in sensemaking because of confusion? they simply do not know how to mediate among interpretations (Weick, 1995). For uncertainty, sensemaking arises because individuals lack

access to a plausible interpretation?they can not see (or are unsure) how their actions will affect the future (Weick, 1995). In either case, sensemaking gets triggered because it is diffi cult to determine a course of action.2

Second, while sensemaking never ends?as soon as we make sense of something we must make sense out of that something?individuals

2 In cases where individuals tolerate uncertainty or equiv ocality, they may have an artificially high degree of confi dence in what actions to take. As a result, they may develop intuitive reactions (discussed below) to the artificially cer tain interpretation. In such situations individuals often react in dangerous ways because they cannot create order out of the equivocality and uncertainty in their environments (Weick, 1993). When there is an absence of uncertainty or equivocality, rationalist approaches may have more merit because the strength of social stimuli may overpower indi viduals’ expectations and motivations.

This content downloaded from 131.170.21.110 on Thu, 30 Jan 2020 08:53:26 UTC All use subject to https://about.jstor.org/terms

2007 Sonenshein 1029

reach temporary resting points where they have a depiction of a social object; they bracket the phenomenon to simplify equivocality and uncer tainty and focus on a plausible account of an issue that can induce action (Weick, Sutcliffe, &

Obstfeld, 2005). Pressed for time, individuals’ search for the accurate answer (assuming that one can even be reached) will stifle action, something encapsulated in the speed versus ac curacy trade-off most managers make (Fiske, 1992). However, rationalist approaches seem to discourage such a trade-off, since they require “time and energy” for morally intense issues (Jones, 1991: 384).