12

THEORIES OF JUSTICE, RIGHTS, AND

DUTIES: NEGOTIATING THE
INTERFACE BETWEEN NORMATIVE
Copyright American Psychological Association. Not for further distribution.

AND EMPIRICAL INQUIRY

THOMAS A. SPRAGENS JR.

Count me among the firm believers in the great value—nay, the neces-
sity—of cohabitation and cross-fertilization between normative political
theory and empirical social science, for conceptions of justice are an impor-
tant subject of moral deliberation and assessment, and they play an impor-
tant role in real-world political events. Philosophers and social scientists who
deal with issues of justice can learn useful things from each other but only if
they understand the porous boundary between prescription and explanation
in social theory. Despite the logical gap between is and ought, empirical find-
ings can provide a prima facie case for theories of justice and can illuminate
whether they pass the test of feasibility. And normative analysts can provide
relevant ideal-type models to guide empirical inquiry and can help resolve con-
ceptual problems that may distort empirical interpretation and explanation.
Issues of justice, rights, and duties certainly represent one area in which
empirical and normative inquiry can overlap and interact very profitably.
Justice is not the only norm that concerns political theorists, but it clearly is

253

http://dx.doi.org/10.1037/10872-012
The Psychology of Rights and Duties: Empirical Contributions and Normative
Commentaries, edited by N. J. Finkel and F. M. Moghaddam
Copyright © 2005 American Psychological Association. All rights reserved.
 one of the most inescapable considerations for them. Prescriptive theories of
politics essentially seek to ascertain those social practices and institutions
that can, under the prevailing circumstances, permit the society's members
to lead flourishing lives consonant with the constraints of justice. Madison
(Hamilton, Madison, &Jay, 1787-1788/1961, p. 324) wrote in Federalist 51
that "justice is the end of government. It is the end of civil society. It ever has
been and ever will be pursued until obtained, or until liberty is lost in the
pursuit." And more recently, John Rawls (1999, p. 3) has insisted in parallel
fashion that "justice is the first virtue of social institutions." So every at-
Copyright American Psychological Association. Not for further distribution.

tempt to produce a comprehensive normative conception of politics has to

attend carefully to canons of right and justice.
Conceptions of right and justice, however, are not only theoretical no-
tions deployed by philosophical onlookers. Such conceptions inform the be-
havior, the commitments, and the attitudes of most of those who live in a
society and participate in its politics. These conceptions are, in short, em-
pirical forces that play an important part in shaping political events in the
real world. Empirical political scientists, sociologists, and psychologists, there-
fore, have to attend to these conceptions in their descriptions and explana-
tions of social behavior. And here then, surely, is a place where the essential
concerns of normative and empirical analysts converge and where the in-
sights and findings of each can be useful to the other.
Interchange across the boundary between empirical and normative in-
quiry into conceptions of justice, rights, and duties needs to take place, then.
But it needs to be done right. If what is possible and proper in this context is
not appreciated, the result can be confusion and ill-based claims on both
sides. One has to understand what each enterprise can do for the other, and
equally important, what each cannot do for the other. And to reach this
understanding of the proper way to negotiate the interface between facts and
norms when it comes to conceptions of right and justice one needs, or so I
argue here, to escape the long shadow of David Hume.

Hume's Shadow: The Gap Between Is and Ought

Writing in the 18th century, Hume made two famous arguments re-
garding the relationship between empirical findings and normative conclu-
sions. As it happens, these two arguments do not cohere very well with each
other, and they both have led to misunderstandings and misconceptions that
have impeded or distorted the interface between social science and social
ethics. One of these arguments has been, unsurprisingly, misread as prohibit-
ing any legitimate use of empirical data by normative theorists in the devel-
opment or justification of their arguments. The other, paradoxically, seems
to dissolve the independent status of moral theory and social ethics alto-
gether, reducing these normative enterprises into a subspecies of empirical
science. The first argument seems on its face to render social scientific find-

254 THOMAS A. SPRAGENS JR.

 ings altogether irrelevant to normative theory; the second argument seems
to accredit scientific inquiry with the capacity to generate dispositive nor-
mative standards for social practice. Each of these arguments contains an
important truth, but each is also seriously misleading. To negotiate the inter-
face between normative and empirical inquiry into the meaning, functions,
and validity of moral conceptions like justice, rights, and duties, one needs to
identify and exorcise the faulty inferences that Hume's arguments have gen-
erated; it is likewise necessary to recognize the genuine insights in these ar-
guments and to reconcile them with each other.
The first of Hume's (1739/1948) arguments is his famous insistence on
Copyright American Psychological Association. Not for further distribution.

what gets characterized as "the gap between is and ought." At the outset of
Book 3 of his Treatise of Human Nature, Hume offers the following observa-
tion (Part I, Section I):
In every system of morality which I have hitherto met with, I have al-
ways remarked that the author proceeds for some time in the ordinary
way of reasoning, and establishes the being of a God, or makes observa-
tions concerning human affairs; when of a sudden I am surprised to find,
that instead of the usual copulations of propositions, is, and is not, I meet
with no proposition that is not connected with an ought or ought not.
This change is imperceptible; but is, however, of the last consequence.
For as this ought or ought not expresses some new relation or affirmation,
it is necessary that it should be observed and explained; and at the same
time that a reason should be given for what seems altogether inconceiv-
able, how this new relation can be a deduction from others which are
entirely different from it. (p. 43)
There is an important truth in this passage relevant to the interface of con-
cern here, and that truth centers around the word deduction. As a purely
logical matter, it is in fact impossible to produce a valid prescriptive infer-
ence from premises that are entirely descriptive. All the indicative propor-
tions in the world cannot generate a logically compelling imperative conclu-
sion. So if I want to make a logically compelling argument on behalf of a
moral or prescriptive claim, I cannot piece together my justification for that
claim entirely from is and is not propositions. At least one of my premises
must itself be a moral premise. To take a simple example, 1 cannot as a purely
logical matter fully justify an assertion to you that you should give me $10 by
reminding you that you borrowed that amount from me recently and prom-
ised to pay me back. To complete my case syllogistically, I would need to add
the stipulation that as a moral matter one should keep one's promises.

Bridging the Gap: Counsels of Prudence

What is misleading about Hume's argument, however, turns about the
words inconceivable and entirely different. These terms seem to imply that nor-
mative and empirical discourse are so utterly remote and distinct that they

JUSTICE, RIGHTS, AND DUTIES 255

 can have no relevance to each other. And that misleading implication has
been acted on both by crude meta-ethical theories such as the "emotive theory
of ethics,"1 which depicted moral claims as cognitively empty expressions of
feeling, and by equally crude conceptions of wertfrei social science,2 which
insisted that social scientific findings had no practical bearing on questions
of social policy and social morality.
Now if all propositions that come packaged in the indicative language
of is and is not referred solely to static and discrete states of affairs such as the
"red, here, now" description that Bertrand Russell offered as the paradig-
matic data report, the belief that logical distinctness implied mutual irrel-
Copyright American Psychological Association. Not for further distribution.

evance might be justified. That conclusion would seem especially reasonable

if all ought and ought not propositions represented moral stipulations such as
Kant's imperatives of pure practical reason and never represented what the
user of such language sees as peculiarly compelling words to the wise such as
the counsels of Aristotelian phronesis.
When we consider, however, that the propositions of empirical social
science are not simply and entirely bald descriptions of static and discrete
facts such as "40% of Catholic voters in Detroit are registered Democrats" or
"20% of North Carolina's industrial output is in textiles" and when we rec-
ognize that many of the most important propositions of normative social
theory are not stipulative moral imperatives but what J. W. N. Watkins,
characterizing Hobbes's "laws of nature," called "doctor's orders of a pecu-
liarly compelling kind," it turns out that it is not at all "inconceivable" as
Hume insists but instead rather obvious how moral theorists can move with-
out impropriety from the indicative to the prescriptive mode in their argu-
ments. This is especially unmysterious when we recall that some of the com-
mon indicative propositions are, as Hume noted, assertions about "human
nature." For human beings are not merely pieces of flotsam and jetsam that
drift passively through life in mute and unresisting acquiescence to external
forces. They are, instead, somewhat teleological organisms whose actions are
motivated in large part by natural and socially conditioned desires, needs,
and passionately sought goals. And these human beings are not only the
subject matter of social scientific propositions but also the makers and the
audience of these propositions.

'The central claim of the emotivist theory of ethics is that ethical propositions are not genuinely
cognitive claims about a reality external to the one who utters them but that they are instead
expressions of the emotions of approval or disapproval, of liking and abhorrence, aroused in the
utterer by the state of affairs he or she is calling good or bad, right or wrong. The classic statement of
this viewpoint is perhaps that found in Ayer's (1936) Language, Truth, and Logic, and the best
secondary account is Urmson's (1968) The Emotive Theory of Ethics.
2
Advocates of the wertfrei ideal of social science sought a body of empirical knowledge about human
and social phenomena that had been scrupulously cleansed of normative bias, moral assumptions, or
ideological influence. Max Weber gave one of the paradigmatic expressions of this ideal in his essays
on "The Meaning of Ethical Neutrality in Sociology and Economics" and "Objectivity in Social
Science and Social Policy." These essays are translated and edited by Edward Shils and Henry Finch in
Max Weber's (1949) The Methodology of the Social Sciences.

256 THOMAS A. SPRAGENS JR.

 If doctors tells patients that they ought to quit smoking two packs a day
and give as the basis for their imperative claims the purely indicative asser-
tions that heavy smokers are three times more likely than nonsmokers to
develop lung cancer and that lung cancer leads in a high percentage of cases
to pain, debility, and death, they have committed the logical fallacy Hume
deplores in the famous passage 1 quoted above. But few patients would in an
instance like this chastise their physicians for violating the canons of logic.
And they would refrain from such objections not because they are them-
selves logical incompetents but because they understand the tacit normative
Copyright American Psychological Association. Not for further distribution.

premise at work here. The physicians could complete their logical cases and
close the syllogistic gap by saying that patients ought to prefer a healthy life
to a miserable death. Physicians do not do so, though, because it seems a
waste of breath; they simply assume that "normal" and "rational" people have
the desires physicians take for granted. Logically, of course, patients could
dismiss their doctors' strong counsel, expressed in the imperative language
of ought, by allowing that they were indifferent about life and death or
wanted to die or loved high-stakes gambling. But most patients would not
avow such unusual preference orderings and would accordingly understand
and accept without objection the normative import of their doctors' in-
dicative references.
Now normative social theory is replete with important instances of cases
logically parallel to this homely example. One perennial normative issue
that Aristotle argued about with Plato, an issue that was central to ideologi-
cal disputation during the Cold War and that still provides one important
axis of partisan division within advanced industrial societies, concerns the
best way to institutionalize a society's economic activities. Should these be
left to private exchanges in the free marketplace, or should public agencies
plan and regulate these activities by law? Part of this issue is a freestanding
moral issue about the preinstitutional moral desert and about the nature and
extent of moral entitlements to property. But much of the battle turns about
empirical questions. Do unregulated market economies produce the most ef-
ficient allocation of resources and hence the highest possible gross domestic
product (GDP), a full employment equilibrium, and innovations in produc-
tive technology? Or do they lead to chronic instability, underemployment,
structural poverty, and warfare among stratified social classes? The way that
normative theorists depict the structural features of the good society will
depend on the answers they endorse to these empirical questions, not be-
cause ought follows deductively from is but because a good society is one in
which human lives can flourish and because human lives can flourish better
in peaceful and prosperous settings than in poor and acrimonious ones.
Moreover, as a philosophical maxim goes, "ought implies can." That is,
no valid obligation can be imposed on anyone to do something that is impos-
sible. And although empirical social science cannot definitively establish on
the basis of its evidentiary base what are the limits of the possible (e.g., even

JUSTICE, RIGHTS, AND DUTIES 257

 if all swans are found by empirical investigation to be white, that in itself
does not prove that a black swan is an impossibility, and even if history pro-
vides no known instances of classless societies, that does not prove that it
could never be done), nevertheless one's best assessments of the limits of
social possibility have to be based on one's best understanding of the struc-
tural features and dynamics of the real world that we can know only by em-
pirical inquiry. Therefore, when sociologists like Roberto Michels began
making arguments about an "iron law of oligarchy" and when political scien-
tists studying public opinion and voting behavior documented a consider-
able gap between the knowledge and competencies of the electorate and the
Copyright American Psychological Association. Not for further distribution.

requisites placed on the citizens of the idealized democratic polity, these

empirical findings tended to cut the props out from under or at least to com-
plicate and compromise the normative claims of populist and participatory
democratic theory (e.g., Kariel, 1970).
One final significant bridge from is to ought is worth noting here, par-
ticularly in the context of this edited volume and its topical focus. Norma-
tive theory is concerned with the human good, and one standard definition
of the human good is "the object of rational desire." To the extent that is so,
it clearly seems relevant to know what people out there in the world actually
desire and consider to be good for them. And that information can only be
provided by empirical investigation, canvasing a range of respondents to find
out what people in fact accredit as good and whether their responses are
largely convergent or instead marked by disagreement. The empirical data
here cannot definitively adjudicate normative issues, but they at least shift
the burden of proof toward those theorists who insist on establishing stan-
dards most of their fellow human beings reject.

Empirical Beliefs and Moral Validity

This logic provides the opening wedge for the second Humean argu-
ment that needs considered criticism. This is his avowed intention and rec-
ommendation "to introduce the experimental method of reasoning into moral
subjects." (This language comes from the subtitle Hume, 1739/1948, gave to
his A Treatise of Human Nature.) On its face, such an attempt seems to vio-
late Hume's own admonition against moralists who move from is to ought.
But his thinking ran as follows. The good, he argued (Hume, 1751/1948, Part
I, Section IX, p. 249), is that which people find to be "useful or agreeable to
themselves or others." (It is not entirely clear whether this represents a
stipulative endorsement of a utilitarian conception of the good like Mill's
statement of the creed in his essay on Utilitarianism or whether it is offered
as a kind of Wittgensteinian endorsement of the meaning of terms as grounded
in their ordinary use as he understands it or whether it is some combination
of moral assertion and acquiescence in prevalent conventions.) Accordingly,
to determine what the human good is in concrete terms, one needs to pro-

258 THOMAS A. SPRAGENS JR.

 ceed "experimentally," that is, empirically, to determine what it is that people
find to be in fact useful and agreeable.
But this proposal, which would seem to turn over to empirical social
science the task and the power to settle normative questions, runs into diffi-
culties of both principle and execution. The problem of principle is that
some widely held views about what is right or good may be repugnant to
many people's moral intuitions. In these cases, one would and should hesi-
tate to accredit such views simply because of their empirically ascertained
acceptability to large numbers of people in certain settings. Just because large
Copyright American Psychological Association. Not for further distribution.

sectors of the German populace endorsed or acquiesced in Hitler's genocidal

anti-Semitism certainly does not provide moral validity for the Holocaust.
Or to take a similar more recent case, the fact that many Iraqis see nothing
amiss with Saddam's use of poison gas on the Kurdish population does not
make such a policy morally licit. (A recent critique of American propaganda
efforts used the denunciations of Saddam for human rights violations on ra-
dio messages broadcast into Iraq as one example of the propagand ists' inepti-
tude because they failed to understand that many of the targets of the propa-
ganda would not find these actions improper and delegitimating, as they
assumed would be the case.)
As for the practical problem of carrying out Hume's proposed project,
the phenomenon of moral disagreement poses serious difficulties. Either
empirical inquiry must be taken as leading to moral cynicism or to a debili-
tating relativism—not what Hume intended—or else large blocks of respon-
dents have to be disqualified in their judgment on the grounds that they are
in some sense not adequately rational. Hume (1751/1948) resorted to the
latter strategy of selective data counting to save his project, discrediting as
deluded and hence discountable the moral judgments of those he mocked as
"delirious and dismal hair-brained enthusiasts" (p. 251). In his Enquiry Con-
cerning the Principles of Morals (Part I, Section IX), Hume proposed to include
only the responses of "men of sense" who based their judgments about the
human good on "their natural unprejudiced reason without the delusive glosses
of superstition and false religion." The problem with this strategy of disquali-
fication of respondents, however sincere though it may be, is that it under-
cuts the ostensive authority of the empirical evidence: Whether the evi-
dence is to count or not is decided by the application of a priori philosophical
standards, and if we can accredit these standards on the basis of philosophi-
cal judgments about what is rational, we could just as well proceed to our
substantive conclusions directly from these judgments without taking what
turns out to be a contingent and qualified detour through a gerrymandered
body of "evidence."
Hume thus offers us a pair of apparently incompatible arguments about
the relationship of the empirical and the normative, both of which turn out
on critical examination to be seriously misleading. He insists on the logical
gulf between empirical findings and normative judgments, but in the process

JUSTICE, RIGHTS, AND DUTIES 259

 leaves us with the misleading implication that facts about the world are not
importantly relevant to nonnative claims on that world; he leaves mysteri-
ous and unacknowledged the way that our understanding of the limits of
what is possible, the direction and force of human desires, and the causal
relationships governing the attainment of various goods all create what
Charles Taylor has called a "value slope" that goes a long way toward deter-
mining what practical political goals and policies seem reasonable and proper
to pursue. Then he counsels people to reach their moral conclusions through
an "experimental method" to determine what is right and good by canvasing
Copyright American Psychological Association. Not for further distribution.

public opinion. In doing so, however, he fails to make clear the way that any
conclusions so derived are contingent on prior normative judgments of his
own about what the human good in the abstract consists of (i.e., that the
good is the useful and agreeable) and about what opinions out there are to be
accredited as sensible and reasonable. Where does this leave one, then, with
regard to negotiating the interface between normative and empirical inquiry
into subjects such as justice, rights, and duties?

WHAT PHILOSOPHERS CAN LEARN FROM

EMPIRICAL STUDIES OF JUSTICE

It leaves one, in the first place, telling normative theorists that they
have much to learn from empirical studies of the kind included in this vol-
ume, for these findings are important to their enterprise. But second, it leaves
one telling the nonnative theorists that they cannot expect to find their
questions answered there and that they should resist any intimations to the
contrary. In trying to ascertain what is right and good to do, it seems reason-
able to acknowledge that what empirical social scientists discover about
"commonsense justice" and conventional beliefs about what is humanly de-
sirable add up to a prima facie case but by no means a dispositive case on
behalf of those moral beliefs that they find to be widely endorsed.
Widespread endorsement of certain elements or standards of fairness,
rights, and justice, such as what Stuart Hampshire (2000) has argued, is the
almost universal acceptance of procedural norms of adversarial deliberative
argumentation and constitutes a prima facie case on behalf of these stan-
dards for three basic reasons. The first of these reasons is found in the logic
underlying the venerable notion of the ius gentium. The concept of ius gen-
tium originated with Roman law, and it referred to the settled customary
rules of peoples with whom the Romans had contact. Roman law gave recog-
nition and accorded legitimacy to these rules when they did not conflict
with provisions of Roman law itself. That notion later evolved into the idea,
as in Grotius, that those customary norms widely endorsed by civilized peoples
should be honored as the fruit of the moral sense or natural reason shared by
all rational human beings. In the present context, then, the idea is that norms

260 THOMAS A. SPRAGENS JR.

 of justice and fairness widely embodied in the customs and conventions of
peoples must be seen as carrying a presumption on their behalf by virtue of
this empirically ascertained breadth of their acceptance. Second, a similar
presumption would seem to follow from the liberal and democratic principles
that ground legitimacy in the consent of the governed. And third, to the
extent that people are willing to accord consideration to the moral judgment
of their fellow creatures, it seems proper to respect, as Jefferson put it in the
Declaration of Independence, the "decent opinion of mankind."
This prima facie case can, however, be overridden for legitimate and
Copyright American Psychological Association. Not for further distribution.

compelling reasons. What people generally believe is not always right, both
as to fact and as to morality. Because people in Salem believed in witchcraft
did not mean that anyone there was really a witch. Because most American
colonists condoned the prevailing institutions of personal bondage did not
mean that slavery was a morally acceptable practice. And just because it is a
democratic principle that the people have the right, as Justice Holmes once
wrote, to "go to hell in a handbasket if they so choose," does not mean that
moral theorists, even those who recognize the legitimacy of democratic pro-
cedures, are obliged to acquiesce in making the lineaments of hell norma-
tive. Normative theorists cannot get lazy and simply count votes or test pub-
lic sentiments. They are both entitled and obligated to deploy their powers
of reasoning and their own critically examined moral intuitions in develop-
ing their conceptions of social justice and the public good. Thus normative
theorists need to resist any intimations, some of which appear in places in
the empirical chapters of this volume, that the absence of unanimity across
cultures about principles of human rights means that belief in universally
applicable principles of human rights is ipso facto untenable, or conversely,
that the empirical finding that most people endorse certain ideas of Tightness
(e.g., meritocracy) validates such ideas and insulates them from critical at-
tack. The finding, for example, that something like 75% of the American
public endorses distributive norms at odds with Rawls's difference principle
does not mean that Rawls's argument in A Theory of Justice (Klosko, 2000)
must be abandoned. Likewise, if jurors are widely inclined to ignore judicial
norms of evidence or standards for judgments of guilt or criteria for the defi-
nition of particular crimes, that does not mean that these standards are some-
how wrong or inappropriate: The "common sense" of the jurors may be well
intentioned but erroneous judgment produced by inexperience and a lack of
understanding.
Another reason for normative theorists to attend to empirical studies
of existing moral sensibilities is that normative theories depend for their pur-
chase not only on claims of moral validity but also on considerations of prac-
tical feasibility and efficacy. Persuasive normative ideals have to be plausibly
attainable and not merely morally attractive. For that reason, they come
trailing empirical strings, as it were, to functional anchoring points that are
needed to make them workable. Normative theorists usually recognize these

JUSTICE, RIGHTS, AND DUTIES 261

 necessities, but they tend to treat them too summarily or speculatively. At
this juncture, their empirically oriented colleagues could help them out.
Two relevant examples of this nexus between empirical findings and
normative force arise in the context of John Rawls's project of grounding
democratic practices on moral foundations provided by the sense of justice.
As Rawls has emphasized in his later work, his aspiration is not simply to
provide our democratic society with compelling moral ideas but also to es-
tablish for it an axis of political stability. He has practical prudential hopes
for his theory, not purely idealistic ones. Here the crucial claim is that con-
Copyright American Psychological Association. Not for further distribution.

temporary pluralistic democratic societies cannot establish their legitimacy

(in the empirical Weberian sense of that term) by reference to a common
conception of the human good (e.g., as Aristotle proposed) because these
societies are home to multiple and incompatible conceptions of the good
life. But that inability, he insists, does not doom these societies to become
mere pacts among citizens animated solely by self-interest, a situation that
he argues is dangerously unstable because its bases of social cohesion are so
fragile and problematic to sustain. Instead, contemporary democracies like
the United States can ground their moral legitimacy and political cohesion
in a broad "overlapping consensus" on particular principles of social justice
that he specifies. His account of this putative consensus, however, is essen-
tially speculative and impressionistic, an abstract model offered without much
in the way of evidentiary warrants. Clearly, here is an open invitation and
need for some empirical investigation. What evidence, if any, is out there to
support the claim that such a consensus currently exists among the demo-
cratic populace or that, alternatively, some forces are visible that could rea-
sonably be expected to create such a consensus in the foreseeable future (e.g.,
Klosko, 2000)?
The other important empirical question that is directly pertinent not
only to the practical aspirations Rawls has for his theory but also to all argu-
ments regarding the best strategies for creating and sustaining the legitimacy
of democratic regimes concerns the nature and power of moral motives in
determining human behavior. The empirical question here is this: "What is
the strength and the basis of moral motives as compared with self-interested
motives in human behavior?" And if moral considerations in fact have moti-
vational power, can this power derive from adherence to abstract principles
of justice or are moral motives derived from this source relatively weak by
comparison with those that flow from communal solidarity and devotion to
public good? Here, surely, is an area of empirical investigation within the
provenance of social psychology that carries vast implications for democratic
theory in general and for the persuasiveness of the practical claims advanced
on behalf of Rawlsian "political liberalism" in particular.
To sharpen this point a bit and to clarify its theoretical significance,
consider Patrick Neal's (1997, pp. 195-196) critical observations on the logic

262 THOMAS A. SPRAGENS JR.

 of what he calls the "stability arguments" of political liberals like Rawls and
Charles Larmore. He wrote that these arguments
assume not simply a link, but an extremely powerful one, between moral
belief and motivation, one sufficient to override the force of whatever
non-moral motivations move us. In the absence of an explanation of the
assumed link between moral belief and motivated behavior, I cannot see
why there is any reason to accept the argument of Rawls and Larmore
that their . . . accounts are superior to a modus vivendi account... in
terms of stability.
Copyright American Psychological Association. Not for further distribution.

But, he continued, despite this need neither of them

actually gives an account of moral motivation, though each seems to
assume himself to have done so.... The link between belief and motiva-
tion is simply assumed on their account; they give an account of belief,
but treat it as if it were an account of motivation.
To supply evidence on behalf of the crucial link between belief and motiva-
tion—and perhaps for explanatory theory as to why that evidence either is or
is not forthcoming—one has to make recourse to the methods, the findings,
and the theoretical arguments of empirical psychology. Do deontological
political liberals like Rawls and Larmore have the stronger case when it comes
to the requisites of political stability in contemporary democratic regimes?
Or do modus vivendi liberals like Neal and David Gauthier (1986) have the
better case? For the life of me I cannot see how these arguments can be as-
sessed and adjudicated without making some serious forays into the vine-
yards of empirical social psychology.

WHAT SOCIAL SCIENTISTS CAN LEARN FROM

MORAL PHILOSOPHY
In these several ways, then, normative theorists of politics need to at-
tend carefully to what empirical social scientists can tell them about social
dynamics and human behavior. Without solid anchoring in social reality,
normative theories become adrift and ineffectual or positively misleading.
The wisdom and prudence of phronesis cannot come about by immaculate
conception. What, however, of the reciprocal relationship? Can social scien-
tists learn anything useful from normative theory? Can the analytical meth-
ods or substantive arguments of political philosophy contribute to the clarity
and relevance of empirical inquiry into social behavior? My own sense is that
normative theorists have less to teach our empirically oriented colleagues
than they have to teach us. Nonetheless, there are a few ways in which it can
be profitable for social scientists to attend to the work of normative theorists
or to collaborate with them. Let me conclude this chapter by specifying three
of these.

JUSTICE, RIGHTS, AND DUTIES 263

 First, normative theory can provide those who seek to investigate and
understand "commonsense" moral beliefs and attitudes with empirical mod-
els, categories, and paradigms to guide their inquiry. The normative disputa-
tion that takes place among moral philosophers and social theorists is an
excellent source to use in seeking to identify the most important moral ori-
entations available to those who seek moral reference points for their ac-
tions. These orientations can then serve as Weberian "ideal types" by refer-
ence to which questions posed to experimental participants can be formulated,
data can be gathered, and analytical categories can be deployed to set up
Copyright American Psychological Association. Not for further distribution.

empirical descriptions and explanations. Philosophers are human beings, too,

and their models tend to embody intellectually refined versions of moral
intuitions that arise and function among the general populace. If philoso-
phers debate the relative merits of deontological and utilitarian and
eudaemonistic and conventionalist theories of justice, therefore, it is reason-
able to expect that "commonsense" conceptions of justice will tend to be
somewhat cruder and less articulate versions of those positions. Some people
out there will be commonsense deontologists who approach questions of dis-
tributive justice and social policy first and foremost in terms of what rights
they claim for themselves and attribute to others. Others reason in terms of
what they see as the common good. Others repair to customary rules and
standards, and so on. Empirical students of everyday moral beliefs and their
influence on behavior, therefore, are likely to find in normative literature
one of the best possible sources of the categories and frameworks they need
to help them sort through the booming and buzzing confusion of their nearly
infinite potential data sets. Several of the empirical contributors to this vol-
ume exhibit an awareness of this relevance of normative categories to their
analytical work, as do other practitioners of this craft like the political scien-
tist Jennifer Hochschild (1981) and the experimental social psychologists
Norman Frolich and Joe Oppenheimer (1990). All other things being equal,
it can be expected that the more detailed and refined the knowledge of nor-
mative theoretical arguments possessed by empirical social scientists, the better
their analytical categories are likely to be.
Normative theorists can also contribute to understanding some of the
issues examined in the kinds of empirical investigation contained in this
volume by drawing upon their own "empirical" knowledge base, which is
their familiarity with both contemporary and historical arguments regarding
justice or rights and with the political and ideological uses of those argu-
ments. These contributions can in some instances add clarity to the dis-
cussion and in others may complicate it in useful ways. Let me give two
examples.
The first example regards the apparent perplexity occasioned by
Jefferson's language in the Declaration that "it is their right, it is their duty"
to throw off the yoke of British rule. This claim seems puzzling if the lan-
guages of duties and rights are compatible only correlatively and not through

264 THOMAS A. SPRAGENS JR.

 overlap or identity because the one references a permissive claim and the
other an obligation. In this instance, however, the widespread belief among
the colonists, as seen in the sermonic literature of the day that sought to
provide exegeses of scriptural doctrine concerning legitimate power, was that
they were both entitled to throw off British rule on the grounds that their
rights had been violated and also under a positive obligation to do so because
of a divine imperative to resist tyrants (who violated God's mandate by arro-
gating to themselves an absolute power that was properly God's alone). This
piece of intellectual history helps not only to resolve the apparent puzzle of
Copyright American Psychological Association. Not for further distribution.

the particular text but also to indicate that any attempt to understand the
empirical functioning of discourses of rights and duties cannot proceed on
the assumption that these discourses represent either-ar alternatives that may
not in some instances be embraced simultaneously by political actors.
The second example concerns the claim made in one of the empirical
contributions to this volume to the effect that "the prevalence of rights talk
may be a sign of relative empowerment" of disadvantaged or subordinate
groups in society, whereas "duties talk may be characteristic of advantaged-
group discourse" (see Louis & Taylor, chap. 5, this volume, p. 116). The
logic here is that dominant groups characteristically talk of the duties of
obedience to them and of the constituted authority that sustains their privi-
leges, whereas groups that are trod on invoke rights claims to validate their
attempts to get their oppressors' boots off their necks. Certainly there are
examples that comport with that logic, as in what we often refer to as the
civil rights revolution in this country.
However, the record of normative theories of rights and duties and the
political uses of these arguments provides some evidence to the contrary and
suggests that the attempt to align rights appeals with subordinate groups and
appeals to duties with dominant groups may be too hasty and potentially
misleading. As Mary Ann Glendon (1991, p. 24) argued in Rights Talk, the
pivotal and decisive paradigm for rights discourse in this country was
Blackstone's absolute and unnuanced defense of property rights in his Com-
mentaries. The classes most attracted to the discourse of rights, then, were
property-owning classes that are not subordinate groups. Rights discourse
could be used as a fence to shelter property owners from the reach of the state
(which can be "liberal" or "conservative", I suppose, depending on whether
that state is a monarchy wanting to levy taxes without consent or a demo-
cratic government wanting to redistribute economic resources), but it also
could be used and was used to defend the established property rights and
possessions of better off groups against the design of have-nots against these
holdings— against, as Locke (1690/1924) put it in Second Treatise of Civil
Government (Chap. V, Section 34, p. 133), "the fancy or covetousness of the
quarrelsome and contentious." Both of these uses were accepted as legiti-
mate by the Founders, and both are on display in Madison's deployment of
rights language in Federalist 10.

JUSTICE, RIGHTS, AND DUTIES 265

 The contemporary analogue of this deployment of rights discourse on
behalf of the self-protective claims of property holders is found in conserva-
tive and libertarian arguments such as those of Richard Epstein (1985) and
Robert Nozick (1971). Nozick began his argument with the words "Individu-
als have rights, and there are things no person or group may do to them
without violating their rights," (p. ix) and these rights, he asserted, include
an absolute entitlement to any property honestly gained or freely given. This
is not the discourse of the disenfranchised. Contemporary arguments over
affirmative action, to take another example, also seem to confute the claim
Copyright American Psychological Association. Not for further distribution.

that subordinate groups invoke rights, whereas dominant groups invoke du-
ties. Here it is the critics of state action to privilege historically subordinate
groups who build their case around claims of rights infringement, whereas
the principal arguments in defense of such policies are currently grounded in
social utility and the social values of inclusiveness and "diversity." The gen-
eral conclusion to be drawn from these examples taken from intellectual
history and recent moral philosophy is this: One cannot be entirely comfort-
able with any claim made on the basis of particular empirical examples to the
effect that one can reliably associate rights claims with subordinate groups or
assume that dominant groups are somehow likely to abandon invocations of
rights and repair to insistence on duties. That is not logically necessary, and
there are too many examples that confute this alleged relationship of social
status and moral languages.

CONCEPTUAL CLARITY AND EMPIRICAL INVESTIGATION

A last way in which social scientists who study the role of moral beliefs
and political ideals in shaping political behavior can benefit from collabora-
tion with normative theorists involves the research component that Siegfried
Hoppe-Graff and Hye-On Kim (chap. 3, this volume) refer to as "conceptual
work." Useful empirical investigation and explanation of social behavior can-
not proceed by a random compilation of facts and observations. The strategy
of data collection is driven by a conceptual framework of some sort that di-
rects the attention of researchers and suggests to them what they need to
look for. And the role of concepts is even more fundamental than that, for as
philosophers of science have taken pains in recent decades to remind us, the
observations that produce research data are themselves in some inescapable
ways "theory laden."
So the "conceptual work" aspect of empirical social science is funda-
mental to its success. Fuzzy or inappropriate or incoherent concepts will cor-
rupt and disable at its base our understanding of what is going on in the social
phenomena we are examining. In Wittgenstein's view, indeed, this problem
was what accounted for the difficulties and explanatory incapacities of the
social sciences by comparison with the natural sciences. From Comte and

266 THOMAS A. SPRAGENS JR.

 Mill on to the present day, the relative weakness of the social sciences has
been attributed to their relative newness and immaturity. They are, as the
recurrent phrase goes, "infant sciences." But that argument has, some two
centuries later, worn thin and unconvincing, Wittgenstein insisted. The prob-
lems of the social sciences will not be solved by time and the aggregation of
ever more data. The deeper problems of the social sciences result from diffi-
culties in formulating concepts and relating them to data. Psychology,
Wittgenstein (1958, p. 232) wrote, and he could have said the same thing of
the other social sciences, is not an infant science; instead, the problem is
that conceptual difficulties cause "problem and method [to] pass one another
Copyright American Psychological Association. Not for further distribution.

by."
To deal with the difficulties of good concept formation and the appo-
site linking of concepts and data, social scientists can make collaborative
efforts with normative social theorists that can be useful and productive—
especially concerning some of the concepts that are specifically normative
ones like rights and duties. Normative theorists have experience not only in
dealing with these concepts but also in the whole process of sorting through
concepts, clarifying them, and seeking to minimize linguistic confusion, as
these are standard tasks of the trade. This kind of clearing of conceptual
underbrush is what Locke referred to as "under laborer" work, and philo-
sophical students of politics and society engage in it continually, even if not
always terribly effectively.
Political philosophers have the methodological tools and the back-
ground, therefore, to serve as useful interlocutors and critics of the concepts
that empirical students of social behavior seek to deploy in the context of
their investigations. No doubt the queries and objections that we are likely
to raise in this context will at times be experienced as an irritating impedi-
ment to moving ahead on the empirical front, but confronting these concep-
tual issues can nonetheless be very helpful in heading off fruitless lines of
inquiry and preventing research from bogging down in conceptual quagmires.
In the context of the empirical research agenda of the contributors to this
volume, normative analysts might raise cautionary queries and challenges in
a number of ways. Consider a few examples for purposes of illustration.
Political philosophers can be useful, first, in helping to sort out and
distinguish issues that are empirical and those that are conceptual. This is
not always an easy thing to do in the social sciences because human actions
possessed of meaning and intentionality can be said to be "lived concepts."
As Peter Winch (1958, p. 23) has put it, "social relations are expressions of
ideas about reality." Hence, the empirical and the conceptual overlap and
intermingle in social scientific descriptions and explanations: They are, as it
were, the internal and external faces of the same "event" or social "phenom-
enon." The logical distinction still remains, however, and it is important
methodologically. One problem that arises here is that social scientists have
a tendency at times to construe conceptual issues as though they were em-

JUSTICE, RIGHTS, AND DUTIES 267

 pirical questions to be adjudicated by gathering relevant observations when
in fact these issues are impervious to "external" evidence of this sort and can
only be addressed by an "internal" analysis of the logic of the concept and the
language game in which it plays. The result is misguided effort and in some
instances conceptual confusion and misleading claims.
In the present volume, for example, most philosophers would surely be
inclined to say that it is an example of this confusion of the relative func-
tions of concepts and data to pose as an empirical question: "How is it that
rights need not be exercised or can be waived whereas waiving duties is not
Copyright American Psychological Association. Not for further distribution.

so easily done?" This in fact is not an empirical question at all but one that
can be answered and can only be answered by explicating the meaning and
use of the concepts involved. It is internal to the concept of rights, part of
their logic, that they can be foregone, just as it is internal to the logic of
duties that they cannot be ignored on one's own choosing. Rights are entitle-
ments and permissions and protections—goods owned by the holder for his
or her own benefit. Hence the holder can without objection opt to waive or
give away rights he or she possesses because doing so can only be at the holder's
expense and not at cost to some other party. And such charity is permitted us
or at least no moral objection can be raised against it. It is part of the very
meaning of duties, conversely, that they are binding on those subject to them.
They are constraints on people's range of action, social obligations imposed
for the benefit of others. (The etymological stem of the word obligation comes
from ligare, which means "to bind.") Strictly speaking, then, empirical inves-
tigation is irrelevant to answering the question cited above as it is posed.
That question can be addressed only and fully by the tools of conceptual
analysis, and the energies of empirical inquiries can be directed toward puzzles
amenable to the kind of evidence only they can provide.
Social scientists are also prone to construing as technical problems is-
sues that arise in the context of "operationalizing" concepts and "coding"
responses when these issues have conceptual and philosophical dimensions
that technical virtuosity alone cannot settle. Is it appropriate, for example,
to categorize as duty based the act of feeding the parking meters of other
people's cars when they are about to be ticketed? In part, this is an empiri-
cal question, albeit not one that can be adjudicated by purely external ob-
servation of the acts in question. We have to know the actor's state of mind
or motives. We would, therefore, have to interrogate the person who did
this to know why he or she fed the meters. And then the philosophical or
conceptual question would come into play: Is this behavior properly con-
struable as a response to a sense of duty? All morally driven acts, after all,
do not have to be pushed into one of two boxes: rights or duties. Perhaps
this was understood by the actor as, and hence was in fact, an act of "ran-
dom kindness," that is, a morally noncompulsory act of charity or
superogatory virtue. Or possibly it was an action taken simply to impede
and frustrate the punitive exactions of authority figures, a thumb to the eye

268 THOMAS A. SPRAGENS JR.

 of the officer poised to write the traffic ticket. There is in fact: important
empirical work to be done here, but we also need to keep our concepts clear
and refined in order to reach appropriate conclusions about the role that
notions of rights and duties (and other moral notions, perhaps) play in
shaping human behavior.
There are two final conceptual observations. The first of these is that it
is important not to conflate the notions of something that is right to do and
something that you have a right to do (e.g., see chap. 3, this volume, for two
such locutions). As my political theory colleague William Galston never
tires of reminding us, we may very well have a right to do something that may
Copyright American Psychological Association. Not for further distribution.

nonetheless not be the right thing to do. We have the right to waste our
natural talent, to engage in political demagoguery, and to stand idly by and
watch someone drown. But it is not right to do any these things. The second
conceptual observation is that it is potentially misleading to deploy the terms
individualism and collectivism uncritically. These are at best rather crude and
conceptually fuzzy ideal-type constructs. It is difficult to specify exactly what
these terms mean in a concrete and specific sense; also, real-world societies
embody a great multiplicity of patterns of relating people to families, groups,
institutions of civil society, and the state. If theoretical distinctions are to be
invoked to characterize and distinguish among the possibilities in this re-
spect, they will have to be nuanced and numerous rather than simple and
dichotomous.

CONCLUSION

In the many ways and for the many reasons canvased above, then, all of
us who seek to understand moral ideas and how these bear on social behavior
both as legitimate standards and as causal determinants have much to learn
from those who labor on the other side of the divide between philosophy and
science, between the normative and the empirical. Those of us on the side of
philosophy need to attend carefully to the findings of our empirically ori-
ented colleagues to know which of our ideals and aspirations can reasonably
hope to achieve the grounding within and the purchase on the world we live
in that they require to be effective and which of our philosophically gener-
ated moral claims and aspirations are destined to flit uselessly about as mere
ghostly phantasmagoria of the academic imagination. Conversely, our famil-
iarity with normative argumentation and its history can provide useful assis-
tance to empirical social scientists as they seek to give their inquiries theo-
retical sophistication and practical relevance. The kind of interdisciplinary
engagement the editors of this volume have sought to create is, therefore, as
welcome as it is rare. We can hope that others will follow their lead and be
motivated to contribute from their own resources to the intellectual cross-
fertilization that can benefit us all.

JUSTICE, RIGHTS, AND DUTIES 269

 REFERENCES

Ayer, A. J. (1936). Language, truth, and logic. London: Gollancz.

Epstein, R. (1985). Takings: Private property and the power of eminent domain. Cam-
bridge, MA: Harvard University Press.
Frolich, N., & Oppenheimer, J. (1990). Choosing justice in experimental democra-
cies with production. American Political Science Review, 84, 461-480.
Gauthier, D. (1986). Morak by agreement. Oxford, England: Oxford University Press.
Glendon, M. A. (1991). Rights talk. New York: Free Press.
Copyright American Psychological Association. Not for further distribution.

Hamilton, A., Madison, J., & Jay, J. (1961). The federalist papers. New York: New
American Library. (Original work published 1787-1788)
Hampshire, S. (2000). Justice is conflict. Princeton, NJ: Princeton University Press.
Hochschild, J. (1981). What's fair? American beliefs about distributive justice. Cam-
bridge, MA: Harvard University Press.
Hume, D. (1948). A treatise of human nature. In H. D. Aiken (Ed.), Hume's moral
and political philosophy (pp. 1-169). New York: MacMillan. (Original work pub-
lished 1739)
Hume, D. (1948). Enquiry concerning the principles of morals. In H. D. Aiken (Ed.),
Hume's moral and political philosophy (pp. 173-291). New York: MacMillan.
(Original work published 1751)
Kariel, H. (Ed.). (1970). Frontiers of democratic theory. New York: Random House.
Klosko, G. (2000). Democratic procedures and liberal consensus. Oxford, England: Ox-
ford University Press.
Locke, J. (1924). Second treatise of civil government. London: ]. M. Dent and Sons.
(Original work published 1690)
Neal, P. (1997). Liberalism and its discontents. New York: New York University Press.
Nozick, R. (1971). Anarchy, state, and Utopia. New York: Basic Books.
Rawls, J. (1999). A theory of justice (Rev. ed.). Cambridge, MA: Harvard University
Press.
Urmson, J. O. (1968). The emotive theory of ethics. New York: Oxford University
Press.
Weber, M. (1949). The methodology of the social sciences. (E. Shils & H. Finch, Eds. &
Trans.). New York: Free Press.
Winch, P. (1958). The idea of asocial science. London: Routledge & Kegan Paul.
Wittgenstein, L. (1958). Philosophical investigations. (3rd ed., G. E. M. Anscombe,
Trans.). New York: Macmillan.

270 THOMAS A. SPRAGENS JR.