Introduction

In 1951, Harold Lasswell—an architect of the policy sciences movement—identified the normative aspects of policymaking as integral to policy analysis when he advised analysts to engage in “a very considerable clarification of the value goals involved in policy” (p. 9). But he also offered a warning:

When the [policy] scientist is reminded to take note of value objectives, he quickly discovers conflicts ... [T]here is the demand to achieve a world community in which the dignity of man is realized in theory and fact. There is also the contradictory demand to make the world safe for ‘Aryan’ or white supremacy. … The dominant American tradition affirms the dignity of man, not the superiority of one set of men. … A glaring discrepancy between doctrine and practice in the United States is the mistreatment of Negroes and other colored peoples (p. 10).

While he didn’t use the modern language of social equity, Lasswell’s 70-year-old references to white supremacy and systemic racism suggest an appreciation of such issues. Indeed, Lasswell’s version of the policy sciences was explicitly normative, anchored in the pursuit of universal human dignity (Wallace & Clark, 2014). In the ensuing decades, however, the curriculum in most schools of public policy paid only limited attention to the creation of a workable methodology for clarifying value goals; the field instead prioritized the methods of microeconomics, empirical research, and decision analysis (Berman, 2022). By the mid-1990s, postpositivist scholars like Dryzek, Durning, Fisher, and Stone began to call for a recommitment to normative thinking and to public values in policy analysis. These scholars, however, struggled to create a well-defined curriculum for postpositive policy analysis (Dryzek, 2002). In turn, practicing policy analysts don’t have easy access to a suite of tools for normative analysis that serves Lassell’s vision of a contextualized, holistic, and interdisciplinary policy science.

Interest in normative issues, nonetheless, remains strong. In 2019, the National Academy of Public Administration identified pursuit of social equity as one of the Grand Challenges in Public Administration, arguing that “moving forward, public administrators and policymakers should develop a broader understanding of the elements and implications of social equity” (NAPA, n.d.). And several faculty in schools of public affairs have called for curricular reform to better address topics like equity, justice, and liberty (Berry-James, et al., 2021; Brady, 2019; McCandless & Larson, 2018).

To be fair, the field of public policy analysis has not ignored value-laden issues. Topics like social welfare, education, criminal justice, healthcare, housing, and inequality are mainstays of policy research. Within such topics, ethical and moral concerns are ubiquitous. My argument is not that policies with normative elements are missing from policy analysis education and practice, but rather that methods for consistent, rigorous, and holistic normative policy analysis need to be considerably strengthened. The problem is particularly acute when it comes to policy analysis, done under time, resource, and evidence constraints to inform specific policy decisions, or as Lasswell put it, to support the intelligence needs of policymakers (1971). But even scholarly policy research, which is less constrained and broader in scope than policy analysis, often focuses on narrow empirical questions while paying limited attention to comprehensive normative analysis (Ascher, 2017).

This commentary considers what could be done to better equip policy analysts to tackle normative issues. Unlike many of the postpositivists, I don’t suggest abandoning the standard model of policy analysis. Instead, I argue for an extension that leaves its foundations intact.

The fact/value dualism in policy analysis

To start, it may be helpful to situate normative policy analysis in a dualism that differentiates empirical and normative thinking, or facts and values. Empirical thinking describes how the world is understood to work while normative thinking focuses on how the world ought to work. Inattention to the fact/value dualism may impede policy analysis when empirical and normative thinking are conflated. Hume is typically regarded as the first to draw attention to the ‘is/ought’ problem (1739/1888). Consider the claim that ‘It is outrageous that 5,000 unhoused people live on the streets of Washington DC; prompt policy action must be taken.’ The number of unhoused people and the circumstances of their homelessness are empirical issues (i.e., fact-based) while the morality of the situation and the wisdom of policy action are normative issues (i.e., value-based). Both forms of inquiry are integral to high quality policy analysis, but each rests on a unique intellectual foundation. Policy analysts who mix up the two forms of thinking may use the wrong analytic tool in a fruitless attempt to clarify an issue.

Several scholars caution against sharply dichotomizing facts and values and instead urge a perspective that views them as complementary constructs (Ryan, 2022; Harmon, 2006; Putnam, 2002). The notion that normative and empirical reasoning are different forms of intellectual inquiry is generally not disputed. Instead, the argument is that credible policy conclusions reflect a collection of claims, some normative and others empirical, rather than one or the other. This intertwining of normative and empirical reasoning in policy analysis constitutes what Ryan calls a network of beliefs (2022). Because different methods are, however, used to validate them, analysts must be adept at distinguishing facts from values within policy analyses (Abel et al., 2021; Dunn, 2018).

In this framework, facts depend on empirical observation and logical inference. To validate a fact-based empirical claim, analysts apply techniques inspired by the social sciences. They gather evidence, subject it to rigorous analysis, develop theories to explain observed facts, describe residual uncertainties, and use empirical findings to answer policy-relevant questions. Although a specific belief in the mind of one person might seem factual to that person, scientific facts are those that have been “confirmed to such a degree that it would be perverse to withhold provisional assent” (Gould, 1983, p. 255). Such confirmation comes from other people, meaning that the process of establishing an empirical claim as valid depends on social processes (Berger & Luckmann, 1966; Kuhn, 1962). Empirical policy analysis thus aims to ascertain whether facts are provisionally true by virtue of having been broadly confirmed by knowledgeable and open-minded folks using sound methodologies (Linquiti, 2023).

In contrast, values reflect moral principles of right and wrong, along with judgements about the relative merits of alternative policy choices, backed by persuasive argument. Values are intangible, abstract, and reflect conclusions about what is important (Maio, 2017). Values are different from simple preferences (Amy, 1984). One needn’t, for example, justify a preference for watching Barbie rather than Oppenheimer. But value judgments—in the context of policy analysis—are expected to be supported by reasons sufficient to justify imposition of public policy.

Normative claims aren’t judged true or false, but instead are validated based on whether they reflect reasoning that is strong or weak, persuasive or unpersuasive, or sound or unsound (Chatfield, 2018; Amy, 1984; Anderson, 1979). For a value-driven normative claim, the analyst interrogates the principles and moralities upon which it is based while looking for inconsistencies and illogical combinations of premises and conclusions. Normative principles are tested against available facts to detect situations where a principle is either irrelevant or in need of restatement to enhance its persuasive force. In short, the analyst searches for and describes reasons that render some value judgements more compelling than others (Dunn, 2018), while resisting the temptation to settle normative disagreements by declaring one point of view true and others false (Lasswell, 1971).

Lasswell uses the word value more broadly than I do here. In his vocabulary, values do not necessarily reflect metaphysical or ethical considerations (Clark, 2002). Instead, values are the gratifying outcomes that motivate humans’ efforts to make themselves better off (Clark & Wallace, 2015; Ascher, 1986). He claims that eight such values drive behavior: power, enlightenment, wealth, well-being, skill, affection, respect, and rectitude (Lasswell, 1970). Someone who, for example, values wealth might pursue a high income or invest in productive assets as a means of achieving their valued outcome. When it comes to policy, a person who values wealth might advocate for private property ownership or a right to an adequate standard of living (Lasswell, 1971). Lasswell acknowledges the expansive nature of his definition when he notes that “[a]mong the many terms that are at least partial equivalencies of ‘values’ are ‘preferences,’ ‘needs,’ ‘desires,’ ‘drives,’ or ‘ends’” (1970, p. 10).

I focus here not on the full array of personal factors that motivate human behavior but only on the normative arguments—couched in terms of right and wrong, morality, and legitimacy—that may be invoked to provide a rationale for public policy. Lasswell does not ignore such considerations; he simply uses a different vocabulary. To that end, Lasswell refers to “transempirical propositions” that are used to justify norms of conduct, with such propositions originating in metaphysics, ethics, and theology (Lasswell, 1971, p. 41). While he counsels policy scientists to avoid passing judgment on the validity of transempirical claims, he also recognizes the importance of critically analyzing them.

Treatment of normative issues in canonical policy analysis texts

To figure out how normative policy analysis might be enhanced, one could start with the ways in which it is taught in graduate public affairs programs. But empirically describing how thousands of students in dozens of programs around the world are being trained to address normative issues poses a daunting research challenge. As a proxy for the content of ‘the’ policy analysis curriculum, I reviewed texts often used in graduate policy analysis courses. Many texts exist, but six are in sufficiently widespread use that they can be called canonical.Footnote 1

These texts share some common elements. The fact/value dualism is evident in all of them. Bardach and Patashnik, for example, explain that policy analysis has “two interconnected but separable plotlines, the analytic and the evaluative. The first is all about facts and disinterested projections of consequences, whereas the second is all about value judgments” (2020, p. 31). Kraft and Furlong concur, arguing that policy analysis that ignores normative issues “is incomplete and inadequate” (2018, p. 191). And Stone, who puts a premium on politically driven normative analysis, nonetheless suggests that “policy disputes entail some disputes over facts …” (2012, p. 312).

While all six books acknowledge the importance of normative thinking in policy analysis, they do so to differing degrees. Bardach & Patashnik offer only a few pages on the topic. Books by Dunn, Kraft & Furlong, Patton et al., and Weimer & Vining address some normative issues, including a distinction between equality of opportunity and of outcomes, permutations of distributional analysis, the role of the political process in settling normative disputes, and variations on the idea of equity. These four texts also touch on foundational topics like liberty, justice, human rights, and material welfare, but none cover all such topics in a comprehensive and in-depth manner. And Stone provides an extensive discussion of normative concepts like equity, efficiency, welfare, liberty, and security, and to tensions among them, arguing that the primary task of a policy analyst is to “reveal and clarify the underlying value disputes so that people can see where they differ and move toward some reconciliation” (Stone, 2012, p. 14).

At least three of the books discuss why values are often given short shrift in the curriculum. With origins in empirical disciplines like economics and operations research, the field of policy analysis has long focused on objective and neutral analysis at the expense of ethical analysis (Patton et al., 2013). Kraft and Furlong agree: “… it is easier for analysts to stress criteria such as effectiveness and efficiency, where an assessment can be based on hard data such as measurable costs and benefits” (2018, p. 191). They also suggest that inattention to normative analysis may reflect a belief that values are not amenable to rational analysis, meaning that an attempt to introduce them into policy analysis will devolve into an irresolvable cycle of moral relativism.Footnote 2 Finally, Bardach and Patashnik, after acknowledging the fact/value distinction, decline to provide a means of addressing it: “[E]ven if everyone accepts the same facts, … each person may apply a different evaluative framework to these facts. … [T]here are no obvious or accepted ways to resolve philosophical differences of this type” (2020, pp. 2–3).

In sum, none of the six books provide specific guidance on how to conduct normative analysis in ways that resemble the specificity found in microeconomics and statistics textbooks.

Competencies for normative policy analysis

The limited focus of canonical texts on normative issues suggests an opportunity to enhance the teaching and practice of policy analysis. I propose seven competencies related to normative policy analysis that students, practitioners, and policy scholars should master.Footnote 3 In brief, they should be able to:

  1. (1)

    Understand where normative issues enter the classical policy analysis model;

  2. (2)

    Understand normative issues adjacent to the classical policy analysis model;

  3. (3)

    Master a vocabulary for normative policy analysis;

  4. (4)

    Understand the nature of moral cognition as it relates to public policy;

  5. (5)

    Apply the tools of analytic political philosophy to policy issues;

  6. (6)

    Characterize empirically the nature and structure of personal moral values; and

  7. (7)

    Locate and characterize policy-specific normative concerns.

Competency 1: understand where normative issues enter the classical policy analysis model

With some variations, the six canonical textbooks suggest a similar approach to policy analysis (the classical model). As depicted below, it comprises six steps.

  1. (1)

    Problem definition

  2. (2)

    Alternative policies

  3. (3)

    Evaluation criteria

  4. (4)

    Projected outcomes

  5. (5)

    Tradeoff analysis

  6. (6)

    Reporting of results

Most of the texts confine normative analysis primarily to Step 3 (evaluation criteria), but close inspection suggests that normative analysis is relevant to every step in the model. Readers are advised to start by defining a problem to be addressed. The word problem indicates that the status quo is somehow unsatisfactory, meaning that there exists a gap between two states of the world: the empirical, as-is condition and a normative, to-be condition. Without a value-based specification of the to-be condition, there is no problem to be solved.

As a second step, analysts are told to develop policy alternatives to mitigate the problem. Doing so can entail empirical study of policies used in other jurisdictions, policies proffered by advocates, or evaluations of existing programs. But the list of distinct options potentially relevant to a particular problem is often a long one and the classical model asks the analyst to winnow the list before analyzing each option in detail. Judgements about which options to study (and which to omit) invariably rest, at least in part, on value-based considerations.

The design of individual policies also reflects normative choices. Deciding on the target population of a policy (individuals, groups, firms, or institutions whose situations are thought to require change), on the balance between coercion and flexibility within a policy, and on the behaviors to be incentivized and disincentivized by the policy is largely a normative task (Schneider & Ingram, 1993).

The third step in the model is the selection of evaluation criteria to judge the pros and cons of the options under consideration. This step is almost entirely normative; the choice of evaluation criteria is inherently a statement about which aspects of public policy are valued, and which are not. A construct converted into an evaluation criterion is, by definition, relevant and important in policymaking; as a practical matter, a construct not used as a criterion is functionally irrelevant.

The fourth step in policy analysis—the projection of outcomes—is sometimes empirical and sometimes normative. If an evaluation criterion reflects a tangible outcome (e.g., the number of homeless persons after a new housing program), then projection falls into the empirical domain. But if the analyst is projecting whether the new program enhances equity across socioeconomic lines, then outcomes projection depends at least in part on normative reasoning.

Having projected outcomes, the fifth step entails describing tradeoffs across options. There are two variants of this step: one empirical, the other normative. If the analyst stays within the empirical realm, a tradeoff is framed by juxtaposing pros and cons of the options, and the tradeoff is left for the client to make. In the normative variant of this step, the analyst makes tradeoffs and offers a recommendation. To do so, the analyst may simply use his or her own values (or their client’s) to derive a recommendation, obscuring the tradeoff process. Alternatively, a summative multi-attribute technique can help make tradeoffs more transparent. But even then, multi-attribute analyses inevitably embed normative choices about the relative importance of the criteria being analyzed (i.e., the ‘weight’ for each criterion) and, in the case of cost–benefit analysis, the role that economic efficiency should play in policymaking.

The final step is to communicate the analyst’s results. This process could be fully descriptive, devoid of normative overtones. The analyst might describe the entire analysis with high fidelity to what was actually done. But this mode of reporting results is the exception rather than the rule, practiced by a few government agencies and some think tanks. Policy analysts want their work to matter, to have an impact on policymaking. In turn, analysts may be tempted to highlight aspects of their work that support their conclusions while downplaying those that don’t. Some textbooks even instruct policy analysts to tell a story or focus on narrative policy analysis. Interesting stories and compelling narratives can make for entertaining reading, but they can also obscure important normative considerations (Weimer, 1998).

Competency 2: understand normative issues adjacent to the classical policy analysis model

Even a policy analyst attentive to normative elements in the classical policy analysis model may miss important insights if they neglect the spaces adjacent to the model. As shown below, there are several other aspects of policy analysis in which normative choices play a key role.

  • Agenda setting

  • Standing

  • Distribution analysis

  • Intersectionality

  • The arc of history

  • Infrapolicies & systemic inequity

Agenda setting entails deciding which issues deserve analysis. Scholars have worked assiduously for years to develop empirically derived, theoretical explanations of how the policy process works (Weible & Sabatier, 2018). These models typically ground agenda setting in a dynamic mix of political, social, and institutional forces. An individual policy analyst may have little control over their analytic agenda, either because broad societal forces shape the agenda or because the client directs the analyst to focus on a specific topic. But policy analysts should appreciate that agenda setting has important normative implications; subjecting one issue to analysis rather than another is akin to deciding that the first is more important than the second.

Another normative concern often not made explicit is the question of standingFootnote 4 (Guy & McCandless, 2020). A person or group with standing matters for purposes of analysis (Robert & Zeckhauser, 2011). In turn, their problems, and the impacts of policies on them, are deemed suitable subjects for analysis. But impacts experienced by those who lack standing don’t matter (at least for purposes of analysis). A person can have standing by sheer accident of birth (Wright, 2019). Someone born in Juarez, Mexico, for example, rather than four miles north in El Paso, Texas, would not have standing in an analysis focused on U.S. citizens. And whether a fetus has standing is an important element in the abortion debate. Failure to explicitly address standing may obscure an analytic choice with normative implications.

The third normative limitation of the model is that it does not emphasize distributional analysis. Both Kraft & Furlong and Bardach & Patashnik are generally silent on distributional analysis. The other four canonical books note the importance of distributional analysis but provide few details on how to conduct such analyses. Patton et al. provide some broad suggestions for deciding which socioeconomic subgroups should be subjected to distributional analysis.

Inattention to distributional analysis might be appropriate in a homogenous society with a history of equal treatment of its residents and where the pros and cons of policy fall more or less equally across the public.Footnote 5 But in a society with significant current and historical disparities in wealth, status, and access to opportunity, inattention to distributional analysis means that the equity of a policy cannot be fully determined (Acland & Greenberg, 2023; Revesz & Yi, 2022; Hammitt, 2021; Guy & McCandless, 2020).

A fourth option to normatively extend the standard model is to consider how policies operate at the intersection of multiple identities. Kimberle Crenshaw was the first scholar to theorize intersectionality as a critical element in law and public policy (Crenshaw 1989). Her initial work focused on mechanisms by which race and gender interact to disadvantage Black women, but intersectionality is now understood to apply to multiple identities (Guy & McCandless, 2020). The topic has garnered significant scholarly attention with, for example, over one hundred peer-reviewed research studies of the links between intersectionality and public policy published between 2010 and 2017 (Garcia & Zajicek, 2022).

While intersectionality is a nuanced concept, its essence is straightforward. Policies, practices, and circumstances of oppression (and privilege) that originate in individual and group identities cannot be fully understood when interpreted one at a time. Instead, we need to consider how multiple identities come together and interact to determine key outcomes. Intersectional analysis has been described as multiplicative rather than additive (Hankivsky & Cormier, 2011). The impact of, for example, race cannot simply be added to the impact of gender to understand how a person holding both identities is affected by public policy. Instead, the two identities interact to produce a race/gender-driven impact that often cannot be explained only on the basis of race or gender acting in isolation from each other.Footnote 6

The fifth situation in which the classical model may overlook important normative factors stems from its approach to historical circumstances. While the model doesn’t require an analyst to ignore the past, it also does not demand consideration of how the past affects the present. The model focuses the analyst’s attention on current conditions and asks what it would take to improve them. The concern here is that inequities of the past often have lingering consequences in the present. At an individual level, a person’s life opportunities today depend at least in part on the opportunities afforded to their ancestors. If we’re interested in equal opportunity for someone today, we need to investigate whether there was equal opportunity yesterday for their forebears (Wright, 2019).Footnote 7

The arc of history is also relevant to normative analysis at the societal level. During the twentieth century, for example, government policy and commercial banking practices created widespread residential segregation in the U.S. (Rothstein, 2017). Federal legislation and related court decisions have since rendered such practices illegal, but their residual impacts persist, including lost opportunities to accumulate housing wealth, communities unable to afford adequate municipal services, and lingering intergenerational trauma within families that have faced profound inequity. In short, if current conditions have been materially shaped by prior inequities, then a policy analysis that overlooks the past is incomplete (Meyer et al., 2022).

The final normative element often given short shrift in policy analysis education is the operation of infrapolicies (Linquiti, 2023). Infrapolicies are like physical infrastructure; they provide the context in which other policies exist. Examples include property rights, civil rights, law enforcement, zoning, tax codes, elections, social welfare programs, public education, and transportation investments. The Jim Crow system, for example, was built on infrapolicies that together, disenfranchised, oppressed, and often terrorized persons of color living in the post-Civil War south. Like all policies, infrapolicies embed normative judgments. Most important here is their cumulative effect. If only one infrapolicy is inequitable, then its consequences are likely more modest than if multiple inequitable infrapolicies interact to create systemic inequities.

Competency 3: master a vocabulary for normative policy analysis

Many elements in the policy analysis canon are precise and unambiguous. In microeconomics, students are taught that consumer surplus is the area under the demand curve and above a horizontal line equal to the market price. But when it comes to matters of normative analysis, ambiguity abounds.

Take, for example, the idea of social equity. Some authors frame the concept narrowly, as a variant of equality in which equal access, not just equal opportunity, is the defining characteristic of social equity. Others are more expansive, including in their definition the concepts of due process, fairness, justice, representative government, equality of outcomes, and equal access to the political process. Social equity has been characterized as “fraught with definitional confusion” (Murray & Davis, 2001, p. 578), lacking “a standard definition” (Durant & Rosenbloom, 2017, p. 723), “difficult to define” (Johansen, 2019, p. 3), and “one of the least understood concepts in public administration” (Guy & McCandless, 2020, p. vii). More than a dozen different definitions can be found in the global public administration literature (Johansen, 2019). And a systematic literature review of 145 articles related to equity in public services found that many lacked an explicit definition of equity. Among studies that did provide a definition, the concept was often defined in different or sometimes conflicting ways (Cepiku & Mastrodascio, 2021).

If scholars, practitioners, and citizens attach different meanings to the key terms used in policy debates, the odds of productive dialogue are diminished.Footnote 8 Accordingly, I propose that rather than starting with one construct—equity—and then assembling all the concepts that fit within it, we reverse the process by parsing policy debates to identify the more granular normative concepts at play. To be useful for aspiring policy analysts, terms in this vocabulary should be collectively exhaustive to ensure all normative concepts that actually arise in ‘real world’ policy debates are represented. The terms, however, needn’t be mutually exclusive. Normative concerns typically arise within a network of concerns, creating close connections among core concepts (Ryan, 2022); overlaps are inevitable. Drawing inspiration from Stone (2012), the normative concepts listed below can be viewed as elements of a vocabulary for normative policy analysis.Footnote 9

  • Rights & Duties

  • Freedom: Liberty & Democracy

  • Equality: Opportunity, Access, Outcomes

  • Justice: Procedural, Distributional, Restorative

  • Domestic Tranquility: Security & Stability

  • Human need & Material Well-being

I offer this list with trepidation. Trying to describe concepts like justice or equality in a few paragraphs is an act of intellectual hubris. Each has been debated over centuries. This review is meant only to describe their relevance to normative policy analysis, not provide a thorough exploration of each. Moreover, additional empirical work is needed to confirm that these concepts collectively exhaust the range of normative claims typically heard in ‘real world’ policy debates.

Rights and duties codify other normative constructs (e.g., a right to equality or freedom). Defining a normative expectation as a right elevates it to an entitlement that must be respected; someone whose rights have been denied has been unambiguously wronged. All rights create a duty or obligation on the part of other people or government to deliver on the right, to satisfy the entitlement implied in the right. Whether a right has been formally codified is another key question for normative policy analysis. Many rights are asserted, and may be morally persuasive, but if they are not guaranteed in constitutional or statutory law, then their impact on policy will be more limited.

Also important is the degree to which rights conflict with one another. Political discourse in the U.S. is often framed as a competition of rights (Stone, 2012). Greene agrees and cites the debate about affirmative action in higher education as a prime example (2021). On the one hand, longstanding racial discrimination implies a right to a legal remedy for past injuries (e.g., affirmative action). On the other hand, intentional race-based preferences for one group over another may deny the right of the latter group to equal treatment.

When there exist multiple conflicting rights and when rights are seen as absolute entitlements, the inescapable result is polarization (Greene, 2021). In essence, a conflict of rights becomes a zero-sum game. One partial solution is to identify fewer rights entitled to strong legal protection and to allow the political process, rather than the judiciary, to mediate among rights by relaxing their absolute nature to allow multiple rights to be recognized, albeit each more weakly, at the same time.

A commitment to freedom is a cornerstone of the post-war Western ethos.Footnote 10 Simply put, freedom implies a lack of constraint. In public policy, it can be viewed as the right to self-determination, either for the individual or for the community (Wright, 2019). At an individual level, freedom manifests as personal liberty. At the collective level, it manifests as democracy.

Personal liberty can be divided into two variants (Stone, 2012; Berlin, 1958). The negative variant is the freedom to do as one pleases without being encumbered by others (as long as one’s actions do not inappropriately constrain the liberty of others). The positive variant suggests that true self-determination is only possible if one has the social and material resources to lead a fulfilling life (Wright, 2019).

At the community level, democracy provides a collective mechanism by which a community self-determines the public policies to which it will subject itself, rather than being subject to the decisions of a ruling elite or an authoritarian regime. Giving citizens voice through democracy is a key characteristic of social equity (Johansen, 2019), as is easy access to political and electoral processes (NAPA, n.d.). Threats to such processes include gerrymandering, burdensome registration and voting procedures, and denial of voting rights to some citizens. Democracy is more than a means of ensuring collective freedom. It also provides a mechanism for legitimating decisions that may in turn foster public acceptance of those decisions (Stivers & McDonald, 2023).

Equality is also a cornerstone of the Western ethos, but without further elaboration, the concept isn’t particularly helpful for normative policy analysis. There are many circumstances where unequal treatment of people is normatively acceptable. Most would agree that an employer can treat qualified and unqualified job applicants differently when filling a vacant position or that a police officer needn’t treat drivers who run red lights and those who stop at them equally.

The first task of a policy analyst examining equality is thus to discern which groups (or classes) of people merit equal treatment and the situations in which they merit such treatment. Another aspect of equality relevant to normative policy analysis originates in the “Equality of What? debate” (Wright, 2019, p. 10). In that debate, some advocates argue for protecting only equality of opportunity, others for ensuring equality of access, and still others for fostering equality of outcomes.

Equality of opportunity implies that all are treated in an identical fashion, without regard to how they come to be in their current position. From this perspective, it would be normatively acceptable in college admissions to prefer applicants with higher grades or test scores over those who lack such credentials, even if they were primarily white rather than persons of color. But if the basis of the admission decision wasn’t the strength of the application, but the applicant’s race or ethnicity, then the principle of equal opportunity wouldn’t countenance the situation.

In contrast, equality of access does not require that everybody be offered identical opportunities (McCandless & Guy, 2020). Instead, it aims to mitigate prior circumstances that impede a person’s ability to access a beneficial situation that is accessible to others who didn’t face such impediments (e.g., affirmative action in higher education). Equality of access allows for disparate treatment of folks, as long as the disparate treatment equalizes their ability to access some benefit or service (Johansen, 2019).

A final variant of equality focuses on outcomes. Equality of outcomes requires that the distribution of, for example, income, education, or healthcare be similar across people and groups. Proponents of equal outcomes, however, usually don’t insist on identical outcomes for all; instead, the objective is to confirm that disparate outcomes result only from normatively acceptable factors, perhaps hard work, choices freely made by responsible adults, or natural talent. But if unequal outcomes result from normatively objectionable factors, like overt discrimination or unequal access to needed resources—then inequality of outcomes may be a matter of public concern.

Justice can be defined as “getting one’s due” (McCandless & Guy, 2020, p. 3). At least three forms of justice are relevant to policy analysis. First, procedural justice implies that government processes respect agreed principles and do so in a consistent, predictable, and transparent fashion. Moreover, government processes must operate the same for all in society, irrespective of wealth, power, or privilege.

The second variant of justice—distributional justice—focuses on the allocation of resources, opportunities, burdens, and obligations across groups and individuals. The idea here is that it is not enough to ensure procedural justice; we need to also consider whether public policy actually delivers equitable outcomes (Dunn, 2018; Weimer & Vining, 2017; Patton et al., 2013). At the center of distributional justice is the notion of deservingness. In a just world, people get what they deserve and deserve what they get (Davis & Wilson, 2022; Swift, 2019). In turn, an undeserved benefit, or a deserved but undelivered benefit, are examples of distributional injustice.

The final variant of justice—restorative justice—aims to make things right after a wrong has taken place. Both the civil and criminal court systems provide restorative justice, albeit on a narrow scale. The construct of restorative justice can also be applied more broadly. Doing so requires more than ending current injustices and preventing new injustices. If a prior injustice has lingering consequences (e.g., twentieth century de jure residential segregation) or has gone unacknowledged by society writ large (e.g., the 1921 Tulsa Race Massacre), then providing restorative justice implies that more must be done. Designing suitable restorative public policies is not simple, but it can be done (Gregory, Halteman, Kaechele, & Satterfield, 2023). Examples include payment of reparations, targeted public investments in communities affected by prior injustice, and truth and reconciliation processes where stories of prior injustices are publicly acknowledged, victims are identified, and perpetrators are named.

Delivering security and stability by ensuring domestic tranquility is one of the Constitution’s core purposes. Stone defines security as the absence of insecurity, or a “worry that something bad will happen” (Stone, 2012, p. 130). Using policy to keep bad things from happening, however, is not easy.

Threats to society’s collective sense of stability originate in many ways. Crime can destabilize a community. Abusive policing may not only victimize individuals but contribute to a climate of mutual hostility. Civil unrest—an insurrection or widespread rioting after a police incident—can leave communities physically and psychically damaged. Political sectarianism, hate crimes, and toxic social media may undermine citizens’ sense of community and trust. Natural disasters like hurricanes and pandemics, and anthropogenic disasters like oil spills and cyber-attacks, disrupt society’s sense of equanimity and order. Immigration may bring new residents and their lifestyles to a once homogenous community that sees the changes as unwelcome disruptions to its way of life. Tracking of online behavior by technology companies and use of facial recognition software by government may represent invasions of privacy. Finally, new technologies like artificial intelligence may portend unsettling changes in human relations.

While empirical evidence is certainly relevant to debates about security and stability, these debates rest heavily on normative concerns. When a policy analysis touches on issues related to security and stability, an analyst unfamiliar with human needs for order and tolerance for disorder, along with attitudes toward the ‘other,’ may miss important aspects of the problem and potential solutions.

Life is about more than income and wealth, but a minimal level of material wellbeing is surely a prerequisite for a basic quality of life. Material well-being is integral to normative concerns like equality, distributional justice, and the positive version of liberty. The Universal Declaration of Human Rights articulates a right to a standard of living adequate for health and wellbeing (UN General Assembly, 1948). Poverty also has dimensions that go beyond the individual. Communities with high levels of poverty are also likely to have limited tax resources and face fiscal challenges in providing municipal services. Fewer investments in schools, healthcare, and other social services may reinforce cycles of poverty as new generations struggle with the effects of deprivation.

Defining a minimally acceptable standard of living, and the appropriate role of government in securing it, however, can be a challenge for a policy analyst. A person unable to afford food, clothing, shelter, and healthcare would seem to be in need. But this simple idea masks a tougher question: How much is enough? Stone notes that even if there is general agreement that people in need deserve help, the construct of ‘need’ is so morally powerful that it leads to intense debates about how public policy should help those in need (2012). Empirical evidence underscores the point. About 45 percent of Americans say that government aid to the poor does more harm than good by making people too dependent on aid while 54 percent say such aid does more good than harm because people can’t escape poverty until basic needs are met (Pew Research Center, 2022). To analyze normative issues related to human need and material wellbeing, policy analysts need an empirical understanding of the nature, distribution, and correlates of poverty (and affluence). They also need a working knowledge of concepts like income, wealth, and inequality and the ways in which taxes and transfers affect material wellbeing.

Competency 4: understand the nature of moral cognition as it relates to public policy

Policy analysts need to understand how people—clients, policymakers, advocates, citizens—form normative judgements on matters of public policy. The focus here is on morality-based value claims, not claims motivated by material self-interest (though there is surely much overlap between the two). Value claims are part of the reason-giving process to justify and elicit support for public policy.

Two schools of thought are evident in efforts to describe moral cognition: moral reasoning and moral intuition (Martin et al., 2021; Maio, 2017; Patton et al., 2013). To use the language of dual process theory (Kahneman, 2011; Shafir & LeBoeuf, 2002; Epstein, 1994), moral reasoning originates in slow, deliberative System 2 thinking while moral intuition originates in fast, instinctive System 1 thinking (Haidt, 2007).

Moral reasoning treats normative thinking as a methodical process, with conclusions deduced or inferred in a logical and systematic manner. Moral reasoning comprises at least two distinct approaches: consequentialism and deontologism.Footnote 11 Consequentialism defines the moral virtue of an action based on its outcomes, not on the morality of the methods by which the outcomes are achieved (Tanner et al., 2008). By contrast, deontologism judges the moral virtue of an action based on its fidelity to moral principles which ought not be breached, irrespective of the outcome (Maio, 2017).

The second school of thought holds that normative conclusions result not from reason and logic, but from the expression of emotion and intuition originating in System 1 processes (Graham, et al., 2018). This idea is not a new one. Three hundred years ago, Hume argued that

“Morals excite passions, and produce or prevent actions. Reason of itself is utterly impotent in this particular. The rules of morality, therefore, are not conclusions of our reason” (1740, p. 457).

Hume’s perspective—that moral conclusions are driven by human emotions—has been confirmed by empirical work done by cognitive psychologists (Graham et al., 2018; Maio, 2017; Haidt, 2001). This research is empirical, not theoretical. The claim is not that folks should use emotion to draw normative conclusions, but rather that they do rely on emotion to do so (Graham, et al., 2018; Mair, et al., 2019). In this research, moral intuition has been likened to an instinctive normative judgement that appears spontaneously in the mind and is not the result of effortful and deliberative cognitive processes (Haidt, 2001). Dunn also links policy-relevant moral reasoning to empirical work in social psychology (2018).

Because emotions and subjectivity play a key role in policy-relevant thinking, a purely rational approach to policy analysis may overlook important considerations (Ascher, 2007). Stone draws a link between emotion and policy analysis, arguing that traditional policy analysis models often come up short because they “ignore our emotional feelings and moral intuitions, both powerful parts of human motivations and precious parts of our life experience” (2012, p. 11). The topic has garnered increased scholarly attention, with multiple studies confirming the ubiquitous and important role played by emotion in public policy and administration (Maor & Capelos, 2023).

Three other insights from cognitive psychology are relevant to moral intuition. The first is the challenge of cognitive dissonance (Cooper, 2007; Festinger, 1957). Dissonance may arise in the mind when someone encounters a discrepancy between an idea or belief that they hold to be true and a competing idea that they also hold to be true or that is endorsed by their social reference group (friends, family, colleagues, or a political party or leader). Such dissonance can trigger an unpleasant emotional reaction that motivates the person to dismiss one of the dissonant thoughts in order to restore a sense of psychic equilibrium. If the person happens to be a policy analyst and the dismissed thought is relevant to public policy, then its dismissal threatens the quality of the analyst’s work.

The second key insight from cognitive psychology is the phenomenon of directionally motivated reasoning, in which someone allows their preferred conclusion to drive how they think about an issue. In the presence of motivated reasoning, the evidence a person considers, the logic they apply, and the tradeoffs they make are shaped in ways to reach a preferred conclusion. The operation of directionally motivated reasoning has been confirmed in multiple empirical studies (Lerman & Acland, 2020; Sheffer et al., 2018; Baekgaard et al., 2017; Kahan et al., 2017). Most relevant here is a concern that people who engage in directionally motivated reasoning are often unaware of doing so, instead believing that they are engaged in a neutral analytic process (Kunda, 1990). Analysts oblivious to the impact of their own preferences on their substantive work may be unable to provide clients with the best possible advice.

Third, the fact that someone’s normative conclusions originate from intuitive emotions, the suppression of cognitive dissonance, or motivated reasoning, and not rational deliberation, does not mean that he or she can acknowledge them as emotionally driven. Rather, in a process of post-hoc rationalization, most folks are adept at combining seemingly relevant evidence with plausible logic to explain their point of view in ways that resemble moral reasoning (Haidt, 2007). Haidt postulates that when people encounter a situation with overtones of right or wrong, they unknowingly experience an affective response which triggers moral judgment that is then followed by post-hoc rationalization. In other words, the justification offered to defend a policy position may have little to do with what actually motivated the conclusion (Yia-Anttila, 2023). The prospect of post-hoc rationalization threatens transparent policy discourse. If a normative position is driven by a speaker’s emotional intuition but incorrectly described by the speaker as the result of an evidence-based logical process, meaningful debate may be difficult.

Competency 5: apply the tools of analytic political philosophy to policy issues

Policy-relevant claims are sometimes the product of methodical deliberation. And sometimes, they only seem so but are actually post hoc rationalizations. Either way, aspiring policy analysts need to be able to rigorously assess the validity and persuasive power of policy-relevant arguments. Analytic political philosophy, a subfield of philosophy that emerged early in the twenty-first century, provides a framework for doing so (Barnutiu, 2023; Abel et al., 2021; Swift, 2019; Wolff, 2013).

The first requirement of analytic political philosophy is that key terms be described clearly and unambiguously; else, rigorous analysis is impossible. Policymakers often invoke concepts like equality or liberty, for example, to justify their positions despite different understandings of what such words actually mean (Swift, 2019). Part of the analyst’s job is to clarify the language of the debate. Perhaps a politician is referring to strict equality of opportunity while another is talking about equitable access.Footnote 12 Perhaps a politician has confusingly conflated freedom, liberty, and democracy to generate more applause lines in their stump speech. The goal for the analyst here is not to decide which politician is right, only to crystallize what each is saying (or, if an ambiguous claim is made, to call it out as such).

With terms clearly defined, the next step is to form one or more logical syllogisms to characterize the reasoning offered in a policy debate. A syllogism combines premises to reach a conclusion. If the premises are true and the reasoning is sound, the conclusion is deemed persuasive (Chatfield, 2018). When it comes to public policy, a syllogism must combine value-based normative claims with fact-based empirical claims to reach a policy conclusion (Abel et al., 2021).

Exhibit 1 demonstrates how analytic political philosophy and the tools of syllogistic reasoning might be used to explicate the normative issues associated with two policies related to parental rights and children’s healthcare: mandatory vaccination (vax) and a prohibition on gender-affirming care for transgender (trans) children. This example comes from my policy analysis class when students and I explored how it might be possible for someone to support parental rights when it comes to vaccination but not when it comes to trans care. Rather than dismiss the argument as hypocrisy, we used syllogistic reasoning to develop internally consistent arguments for each of four potential policy stances. (These are not the only possible syllogisms that could be used to explain the reasoning at work here; they are meant only as illustrative examples.)

Immediately apparent is that two types of policy are being compared. One requires an action (vaccination) while the other prohibits an action (trans care). This distinction requires logical syllogisms that vary in content. This example also underscores that policy conclusions depend on both fact-based empirical inquiry (Are vaccines safe? Is trans care harmful?) and value-driven normative reasoning (What is the nature of parental rights? How do they compare to the health of children and the community?).

While empirical inquiry and normative reasoning demand different analytic methodologies, they also share a common trait. Irrespective of whether analysts evaluate empirical or normative reasoning, they:

… can follow a similar procedure in the face of any claim that strikes as problematic, whatever its nature: inquire into its bases, its supports, learn something about the network of beliefs of which it is a part (Ryan, 2022, p. 7).

Exhibit 1 suggests that the use of analytic political philosophy in policy analysis is unlikely to yield a single conclusion. Rather, an analyst may need to construct multiple syllogisms, each with differing premises, to capture competing points of view heard in policy debates. By juxtaposing the syllogisms, the analyst can highlight areas of agreement and disagreement (Abel et al., 2021).

Exhibit 1
figure 1

Applying Analytic Political Philosophy to Possible Policy Stances about Vaccinations and Gender-Affirming Care: An Illustrative Example

Full size image

Competency 6: characterize empirically the nature and structure of personal moral values

Reflecting on his education in philosophy, one scholar colorfully describes a key limitation on analytic political philosophy:

Reason cut through all the bullshit. … It asked its questions, and it continued asking them till it had reached truth. … [Given] your belief in the truth of something, reason can tell you what else is true. … What it can’t do is tell you what’s true to begin with (Sanklecha, 2023, p. 1).

Sanklecha’s final sentence hints at a challenge for normative policy analysis: the need to explain the starting points of the analysis (e.g., the normative claims in Exhibit 1). Moral reasoning typically begins with values or principles that are not defended based only on empirical evidence or derivation from a demonstrably true conclusion.Footnote 13 The starting points originate elsewhere, typically with moral belief.

This challenge comes as no surprise to moral intuitionists who locate normative thinking in humans’ instinctive emotional systems, rather than in the methodical processes of the political philosopher. But even the most ardent practitioners of analytic philosophy will benefit from understanding the empirical moral foundations of normative value judgements and how moral claims originate in the human mind. It is not enough to expose students only to moral principles developed by learned philosophers; we also need to introduce them to the value systems of their fellow citizens (Barnutiu, 2023).Footnote 14

The two most fully developed, empirically validated theories of moral judgment come from the fields of psychology and sociology. The first is Personal Values Theory (PVT), developed by Shalom Schwartz and colleagues, and the second is Moral Foundations Theory (MFT), developed by Jonathan Haidt, Jesse Graham, and colleagues (Feldman, 2021). There has been limited work to link the two theories, although the comparative work that has been done tends to confirm connections between the two, along with a few instances where moral judgement is theorized in different ways in the two approaches (Yia-Anttila, 2023; Feldman, 2021; Zapko-Wilmes, Schwartz, Richter, & Kandler, 2021). Until more work is done to integrate the two theories, policy analysts will benefit from understanding both.

Personal Values Theory has been validated in 200 + samples from 80 countries (Scharfbillig, et al., 2021). It is built around a set of personal values that operate as guiding principles in the lives of those who hold them. Applicable in multiple situations and at different points in a person’s life, they are central to one’s sense of self and are seen as intrinsically worthy, desirable, and good (Sagiv & Schwartz, 2022). Individuals typically prioritize particular values differently, embracing some, rejecting some, and being indifferent to some.

As shown in Exhibit 2, PVT organizes values into a system that captures the relationships among them. The framework contains two pairs of higher order values, with each value in the pair standing in opposition to the other. One pair comprises conservation (a preference for stability) and openness to change (an inclination toward progress). The other pair comprises self-enhancement (a focus on the self) and self-transcendence (a focus on others). Within these higher order values are nineteen specific values. Empirical work suggests that these personal values are correlated with altruistic behavior, aggressive behavior, political ideology, voting, tolerance, and prejudice (Sagiv & Schwartz, 2022).

Exhibit 2
figure 2

Personal Values Theory Scharfbillig, et al., (2021), based on Schwartz, et al., 2012)

Full size image

A key feature of PVT is its arrangement of values into a circle (see the middle ring of Exhibit 2).Footnote 15 Values adjacent to one another on the circle are similar in nature and likely to be held together. As one moves around the circle to more distant values, the less likely it is that both values will be seen positively by the same person. Moreover, values depicted across the circle from one another reflect opposing values, implying that endorsing one entails rejecting the other.

Scholars working for the European Commission extended the PVT framework with examples of each of the nineteen values (Scharfbillig, et al., 2021). These examples, in the outer ring of Exhibit 2, demonstrate how personal values may connect to policy preferences. A person whose values are anchored in conservation and security, for example, is likely to embrace a strong defense while someone grounded in self-transcendence and universalism will prefer policies aimed at peace and conflict resolution.

Moral Foundations Theory is similar to PVT in that it offers a taxonomy for moral thinking. But its approach reflects the idea that normative positions are often post-hoc rationalizations of instinctive emotional reactions, rather than the result of deliberative moral reasoning (Haidt, 2001).Footnote 16 The theory sorts morally relevant intuitions into categories, or foundations, that capture the origins of instinctive moral judgment (Graham, et al., 2018). Much of the work is empirical, using surveys of different populations around the world, to develop the classification scheme (Atari, et al., in press; Kivikangas et al., 2021).

The latest version of MFT comprises six moral foundations:

  • “Care: Intuitions about avoiding emotional and physical damage to another individual;

  • Equality: Intuitions about equal treatment and equal outcomes for individuals;

  • Proportionality: Intuitions about individuals getting rewarded in proportion to their merit or contribution;

  • Loyalty: Intuitions about cooperating with ingroups and competing with outgroups;

  • Authority: Intuitions about deference to legitimate authorities and the defense of traditions, all of which are seen as providing stability and fending off chaos; and

  • Purity: Intuitions about avoiding bodily and spiritual contamination and degradation” (Atari, et al., in press, p. 13).

These six foundations fall into two higher-order constructs: individualizing foundations and binding foundations (Zapko-Wilmes, Schwartz, Richter, & Kandler, 2021). Individualizing foundations are often endorsed by political liberals and include care, equality, and proportionality. Binding foundations are often endorsed by political conservatives and include loyalty, authority, and purity (Feldman, 2021; Graham, et al., 2013; Kivikangas et al., 2021). Moreover, in two studies with over 24,000 participants, researchers found that the degree to which people endorsed different moral foundations predicted attitudes about ‘culture war’ issues like abortion, immigration, and same sex marriage (Koleva et al., 2012). When it comes to normative policy analysis, MFT provides a framework for helping policy analysts characterize and understand the nature of others’ moral frameworks, especially those that are dissimilar to their own (Graham, et al., 2013).

Giving analysts an appreciation of the nature and structure of personal moral values as exemplified by PVT and MFT is not a formulaic means of resolving disagreements. Value-based disputes will persist, even in the wake of high-quality policy analysis. But disputes may be more productive if areas of agreement and disagreement are clearly identified and linked to underlying values. Indeed, as noted with respect to Competency 5, doing so is integral to analytic political philosophy. Scharfbillig and his colleagues suggest that policymakers can benefit significantly from a deeper understanding of how personal values shape the language of policy debates (2021). Such insights may be particularly important in a polarized political climate in which the deeper drivers of disagreement can be hard to discern (Mair, et al., 2019). And Ryan, who argues that facts and values necessarily co-exist within a network of beliefs, suggests that open-minded discussion of competing values is an important antidote to “end of the line thinking” that leads participants in policy debates toward a polarized, irreconcilable impasse (2022, p. 17).

Competency 7: locate and characterize policy-specific normative concerns

Before policy analysts can take Lasswell’s advice to clarify value goals, they first need to know how to find and characterize values relevant to a particular issue (Weimer, 1998). And they must decide whose normative values belong in the analysis. Sometimes, there is only one important set of values, those of the analyst, or of their client or employer. Sometimes, relevant values come embedded in the analyst’s assignment, when for example, a client identifies a single policy option to be analyzed or is concerned only with a single element of a multi-faceted policy problem. Working with a single normative view, self-introspection or a conversation with the client/boss will suffice to identify relevant normative concerns.

Things get murkier when a multi-perspective approach is called for. Perhaps the client isn’t one person, but a legislative body, trade association, or advocacy group whose members hold divergent views. If the analyst has ready access to those members, he or she can engage them to ascertain their perspectives. Characterizing such perspectives fairly—and reconciling them if possible—then becomes an important task for the analyst.

Analysts may also opt to take a society-wide perspective, in hopes of characterizing the full array of normative values relevant to the issue under study (Ascher, 2017). In such cases, the relevant normative views are not held by a single client or easily identified members of a particular group. Instead, the focus is on a more amorphous group, like the residents of a neighborhood, city, or state. Analysts in such situations must reflect in their analysis value judgments that are not per se their own or their client’s. Rather, they must cast “a wide net” to discern the values expressed by those from across the political, ideological, and socioeconomic spectrum (Ascher, 2007, p. 142).

Analysts seeking to characterize the normative contours of a particular policy issue might start with what Bardach and Patashnik call the “issue rhetoric” or the “ordinary language of debate and discussion in the client’s political environment” (2020, p. 4). Sometimes, specific normative claims are explicit in the rhetoric; in other cases, they must be identified by carefully parsing the discourse. As a quick entry to a policy debate, I ask my students to enter a policy issue into the search function of Google News. The result is a collection of news stories, opinion pieces and half-truths that ranges across the ideological spectrum. Students can then organize multiple disputed normative values into what Robert and Zeckhauser call a taxonomy of disagreement to facilitate further analysis (2011).

One downside of relying solely on secondary sources for normative insight is that the analyst may overlook the views of marginalized communities that rarely participate in policy debates. The analyst may also fail to discern the nuances and subtleties of the normative concerns at issue. In some cases, analysts and affected communities may need to work in partnership to identify and articulate normative concerns (Ascher, 1986; Stone, 2012). In short, policy analysts cannot do their work “in isolation. As they search for relevant social values, they must engage others who are a part of society in conversations to find out what they value” (Weimer, 1998, p. 119). There are multiple approaches—arrayed on a continuum of consultation—that an analyst might use to elicit and understand a community’s normative concerns (Linquiti, 2023):

  • Public participation establishes a mechanism to receive public input, perhaps by taking comments through a website or public hearing. The consultation is one-way, without back-and-forth discussion to deepen either side’s understanding of the issue. Moreover, public servants are likely to hear only from folks paying enough attention to realize that input is being sought.

  • Public engagement entails meetings between self-identified stakeholders and policy analysts. While such discussions entail a two-way conversation that may offer a deeper understanding of stakeholders’ normative views, they may still provide the analyst an incomplete picture of what is most important if people affected by policy problems and proposed policy changes are not well represented in the conversation.

  • Public inclusion requires that policy analysts think carefully about who in a community is affected by a problem or potential remedies and then intentionally reach out to them (or to community leaders who can speak authoritatively on their behalf) to have a two-way conversation about the community’s normative values and policy-relevant concerns.

It may also be the case that relevant communities lack fully formed preferences that policy analysts simply discover and then incorporate in their analysis. Instead, the analyst may need to work with community members to more fully articulate values, preferences, and goals, and to search for common interests, as well as irreconcilable differences (Linquiti, 2023; Stone, 2012; Clark, 2002).

Efforts to characterize a community’s values are likely to be more productive if the policy analyst engages in perspective taking, putting their own views aside to meaningfully consider the opinions of others. Doing so is a two-step process. The first step entails the development of “standpoint awareness” (Wallace & Clark, 2014, p. 147). Its purpose is to allow analysts to examine their own view on the topics being studied in order to better understand their interpretation of and reaction to arguments made by others (Clark, 2002; Lasswell, 1971). This intentional and systematic approach to introspection is sometimes referred to as reflexivity.

The second step in perspective taking entails going beyond an awareness of one’s own standpoint to make a concerted effort to genuinely consider the normative outlook of others. This step is akin to John Rawls’ veil of ignorance test in which we are rendered ignorant of our standing in the world before opining about matters of right and wrong (Lovett, 2011).Footnote 17 We might be privileged or marginalized, young or old, white or of color, LGBTQ + or not; we just don’t know. In turn, from behind a veil of ignorance, our normative views aren’t influenced by our identities or by our standing in the world. Of course, fully abandoning our identity and standpoint is likely an impossible task, but even the attempt to do so may help an analyst better understand normative views that they do not personally share.

Perspective taking can be enhanced by adopting the principle of charity, which holds that arguments heard during a debate should be interpreted to maximize the likelihood of their truth or validity (Speaks, 2021). If an argument is ambiguous or poorly expressed, the principle of charity holds that it should be recharacterized in the way the speaker likely intended it to be interpreted (The Ethics Centre, 2017). Particularly when it comes to thoughtful policy analysis of competing normative values, analysts must guard against dismissing dissenting points of view simply because of their form and expression.

My argument that policy analysts ought to learn how to find and charitably characterize competing normative views is not the same as arguing that they must embrace the validity of such views (Berlin, 2001). Instead, I am suggesting only that they recognize that genuine moral disagreements are pervasive and often a driving force in policy debates (Rai & Fiske, 2011). In turn, an analyst’s work is not well served if dissenting moral claims are summarily dismissed or left unscrutinized. As Laswell suggested, however, when we ask analysts to analyze values, we need to warn them to expect internal conflict (1951). All social communities operate within a moral framework; policy analysts—like anthropologists—must be adept at understanding moral frameworks “that are not their own, and that they may even find personally offensive” (Graham et al., 2018, p. 17).

In short, except for the most specious of claims, if a normative argument is being heard, it belongs in the analysis (Anderson, 1979). There are at least two reasons why this is the case. First, the goal of policy analysis is not to eliminate moral disagreement, but to frame disagreements using a vocabulary that both sides can understand (Yia-Anttila, 2023). Second, if someone aims to oppose values with which they disagree, then as a strategic matter, they must recognize such values are often morally motivated, “not simply errors in judgment, limitations of knowledge, or failures of self-control” (Rai & Fiske, 2011, p. 58).

Conclusion

The foregoing is only an approximation of what is needed to deliver on Lasswell’s aspiration that analysts possess the skills to bring clarity to disputed value goals. Reconfiguring the policy analysis canon requires answers to several questions. What can we learn from academic programs already moving in this direction? What would a comprehensive empirical discourse analysis of current policy disputes tell us about the ways in which normative policy claims are framed, debated, and addressed (or left unreconciled)? Are there competencies that should be added or deleted from the list proposed above? For competencies worth keeping, how might they be revised to be more useful for practicing policy analysts?

Moreover, increased attention to the methods of normative analysis ought not come at the expense of critical empirical analysis within the classical six-step model of policy analysis. Normative analysis is a complement to, not a substitute for, the classical model. An analyst not steeped in the classical model has little chance of successfully applying the competencies of normative analysis proposed here. And, conversely, an analyst not steeped in normative analysis has little chance of completing policy analyses that can deepen policymakers’ understanding of policy issues and facilitate progress toward their resolution.