Volume 5, May 2005
Epistemic and Non-epistemic Aspects of the Factfinding
Process in Law
* Professor of Law, Hofstra University School of Law
Legislators, regulators, and judges
attempt to create factfinding processes that integrate both epistemic and
non-epistemic goals. Moreover, the rule of law requires that those factfinding
processes be principled, equitable, and reasonably transparent. This complex
endeavor therefore produces some of the best-documented examples of societal
factfinding. This essay analyzes the major decisional elements in such
factfinding processes, with attention to common sub-tasks, distinct decision
makers, and points of divergence for institutional procedures and safeguards.
The fundamental point of this essay is
that any factfinding process in a governmental institution is designed to
balance the epistemic objective against relevant non-epistemic objectives. The
epistemic objective is to produce findings of fact that are as accurate as
possible and that are warranted by the evidence legally available to the
factfinder. The non-epistemic objectives include many that are common across
governmental institutions (such as procedural fairness to parties and
administrative efficiency), as well as many that vary by institution and by
area of law (such as public health, safety in the space shuttle program, low
inflation, protection of labor unions, and corrective justice). The background
hypothesis, which is not argued in the essay, is that all truth-seeking processes of human beings exhibit a similar
blending of epistemic and non-epistemic aspects, although few collective
factfinding endeavors are as carefully engineered as the factfinding processes
Factfinding in law is always pragmatic,
in the sense that it always occurs in a context in which governmental action is
at stake. First, the range of possible actions that any particular governmental
institution is empowered to take is usually well settled. Such an action might be
issuing a final regulation, sentencing someone to prison, ordering compensation
or payment of a fine, issuing a report, or making recommendations. Furthermore,
the positive substantive law generally identifies which findings of fact are
required in order to justify particular actions. For example, statutes and
implementing rules identify the factual issues that the Environmental
Protection Agency must establish before that agency can lawfully suspend or
cancel an existing approval (“registration”) of a pesticide that leaves a
residue in food.
Judicial precedents establish the factual elements that a torts plaintiff must
prove before the court can lawfully order a defendant to pay compensation.
Legislative bodies, regulatory agencies engaged in rulemaking, and courts
applying common law decide which factual predicates justify which actions, and
then those engaged in adjudication or enforcement know which factual findings
are legally significant. The propositions at issue in any given proceeding
determine the relevance of any evidence submitted for the record in that
The epistemic objective in factfinding
is to ensure that the factfinder will be accurate in declaring (“finding”)
propositions to be true for purposes of the legal proceeding, and that the
evidence that is legally available warrants or adequately supports those
findings of fact. But the factfinding process cannot be divorced from the
pragmatic context in which it occurs, and which justifies expending resources
on the factfinding process. The non-epistemic objectives therefore influence
what evidence is ruled to be “legally available,” when the factfinder should be
allowed to find that the available evidence “adequately supports” a finding,
what standard of proof the factfinder should use in selecting a finding, what
kinds and levels of uncertainty are acceptable in factfinding, and so forth.
Rules governing such decisions should always balance substantive policies,
procedural fairness, and administrative efficiency against the epistemic
objective. Some non-epistemic policies trump the epistemic objective (for
example, the minimal demands of due process), while others only weigh in the
balance, or only apply when the epistemic objective is unachievable. This essay
will discuss a few of the major decisional nodes in judicial and administrative
It is important at the outset to
clarify the phrases “issue of fact” and “issue of law.” These are terms of art
that indicate who the decider is,
rather than the nature of what is to be decided. “Issues of fact” present
decisions for the factfinder – which in a given proceeding may be a jury, a
judge, an administrative hearing examiner, or a regulator. The factfinder is
supervised in each case by a presiding judge or other official. “Issues of law”
are decided in the first instance by the judge or regulatory official who
presides over the factfinding proceeding, with an appropriate opportunity for
further review. Further review of issues of law is almost always available, and
incorporates various degrees of deference to the presiding judge’s or
official’s ruling. In this essay, the neutral phrase “proposition at issue”
refers to the content of a contested issue, which might be decided in a
particular case through a finding by the factfinder (as an issue of fact) or
through a ruling by the presiding or reviewing authority (as an issue of law).
Participants in a legal proceeding can usually examine the applicable law and
agree at an early stage on the list of propositions at issue. The entire
process in which various decision makers resolve those issues, for the
pragmatic purposes of the relevant legal context, is the factfinding process.
The remainder of this essay examines
the structure of the factfinding process by discussing three major decisional
roles, which I will call the province of the factfinder, the province of the
presiding authority, and the province of the reviewing authority.
Province of the Factfinder
The province of the factfinder has as
its main task the evaluation of all the evidence produced and the declaration
of which propositions will constitute the findings of the proceeding (called
“verdicts” in the case of jury findings). In one sense, the factfinder makes
these decisions with considerable freedom, and there is little “law” about how
to accomplish this task. There are few if any rules about how to assess the
credibility of a fact witness’s testimony, how much probative value to assign
to a document or to an expert’s theory, or how to integrate (for example) the mathematical
probabilities derived from DNA evidence with the likelihood that the
investigating police department contaminated or planted that DNA evidence. But
to say that there is little law about how the factfinder should reach decisions
within the zone of factfinding
discretion is largely a truism, if by “law” is meant decisional rules adopted
by presiding and reviewing authorities. Such rules do channel the factfinder’s
decisions, but those rules operate outside
the province of the factfinder, and therefore outside the zone of factfinding
discretion. The process of factfinding is a dynamic interaction between
factfinder decision-making and constraining rules, with the general trend in
many areas of law being in the direction of adding more constraining rules.
Probably the largest category of
constraining rules consists of substantive rules of law, although such rules
are not always recognized as constraints on the factfinder. Consider a torts
case in which the plaintiff alleges that the defendant acted negligently. A
basic proposition at issue (and presumptively for the jury) is whether the
defendant engaged in any conduct that is considered negligent. There are many
judicial rules, however, that create necessary conditions under this factual
issue. For example, many courts hold that unless the defendant knew or
reasonably should have known about the risks created by a course of conduct,
then the defendant could not have been negligent in engaging in that conduct.
This rule therefore generates a disjunctive proposition about whether the
defendant had “notice” (“knew or should have known”), which any plaintiff is
required to prove in order to establish that negligence occurred. Substantive
judicial rules can therefore generate new propositional issues to be resolved
by the factfinder, yet at the same time impose new constraints on the
discretion of the factfinder.
Another category of constraining rules,
closely related to substantive rules, consists of definitional rules. In a
civil or criminal trial, for example, the presiding judge instructs the jury
concerning the issues of fact about which the jury must make findings. The
presiding judge leaves undefined most of the words employed in those
instructions, for definitions always lead to more words, and defining must end
somewhere. The meanings of most words, therefore, are left to the factfinder to
determine, on the basis of the factfinder’s background knowledge. With respect
to certain critical terms, however, courts may adopt rules of definition. Using
the example of negligence in tort law, judges routinely tell juries that the
law defines “negligence” as “lack of ordinary care” and “a failure to use that
degree of care that a reasonably prudent person would have used under the same
The judge, however, probably will not define further such terms as “ordinary,”
“degree of care,” and “prudent.” Taken together, the substantive rules and
definitional rules identify for the factfinder those propositions that need to
be resolved. And at the same time those rules place boundaries on the
factfinder’s role, and to some extent may influence the factfinder’s decisions
within those boundaries.
A third category of constraining rules
addresses process aspects of the factfinder’s task. Chief among these are rules
concerning the standard of proof and burden of persuasion that the factfinder
is to employ. For every issue of fact to be decided, there is a standard of
proof to be met before a finding should be made. Standards of proof describe
for the factfinder the quality of support required between the available
evidence and the finding. For most issues in civil cases, the standard of proof
is a “preponderance of the evidence”: the factfinder is to make a finding that p if, but only if, the evidence supports
p more than it supports p’s negation, not-p.
For some issues of fact, the law imposes a “clear-and-convincing-evidence”
standard of proof, while criminal cases employ the familiar
“beyond-a-reasonable-doubt” standard of proof.
In contrast to the standard of proof, the burden of persuasion instructs the
factfinder as to which party loses if the evidence does not satisfy the
standard of proof. For example, if an issue of fact is to be decided by a
preponderance of the evidence, but the relevant evidence in the case is equally
weighted as between p and not-p, then, if the plaintiff has the
burden of persuasion on p, the jury
must find not-p, against the
plaintiff. Typically, rules of law allocate the burden of persuasion to some
particular party, for every factual issue in the case.
A fourth category of constraining rule
on factfinders consists of default inference rules. For example, a legal
presumption is a rule of law that describes what inference the factfinder
either must or may draw, once the factfinder finds some specified proposition
to be true. An example is the presumption that a person missing for some fixed
period of time is dead, absent evidence to the contrary.
Presumptions may be either mandatory (“If you (the factfinder) find p to be true, then you must find q to be true”) or permissive (“If you find p to be true, then you may
find q to be true”).
Yet a fifth category of constraining
rules consists of relevant-factor rules – rules of law that prescribe which
factors the factfinder either must or may consider in arriving at a finding.
For example, in determining whether the conduct of an actor was negligent, the
factfinder should take into account the magnitude of the risk involved and the
utility of the act or of the manner in which the act was done.
There are undoubtedly additional types
of rules that constrain factfinder discretion, but the categories mentioned
above supply enough examples to illustrate the following point. Substantive
rules of law, legal definitions, process rules, presumptions, and
relevant-factor rules are all devices to structure the factfinder’s role in
deciding the propositions at issue. Some rules provide the grounds for taking a
propositional issue away from the factfinder altogether (as discussed in the
next section). Some are best understood as explicit or implicit commands (“…
you must …”) or permissions (“… you may …”) addressed to the factfinder. There
is of course no guarantee that the factfinder will understand those
instructions, let alone follow them. The surest safeguard, other than the
good-faith efforts of the factfinder, is the system of oversight provided by
the presiding and reviewing authorities. As discussed below, those authorities
are additional decision makers, whose provinces of decision interact with the
province of the factfinder in complicated ways.
Province of the Presiding Authority
In a judicial proceeding, the presiding
authority is a trial judge. In an administrative adjudication, it may an
administrative law judge or a hearing examiner. In an administrative
rulemaking, it is the presiding regulatory official, who may chair a commission
or a board. In any case, the presiding authority is the decision maker who
presides over the creation of the official evidentiary record and who oversees
the participation of the factfinder. The presiding authority also decides in
the first instance which legal rules are applicable in the particular
proceeding, and decides whether and how to enforce the applicable legal rules.
This may involve instructing the factfinder about the legal rules as it is
appropriate to do so, or granting or denying the motions of participating
In many factfinding proceedings, a
single individual functions as both the factfinder and the presiding authority.
In judicial proceedings, examples are “bench trials” (in which the judge hears
and decides the case without a jury) and hearings on preliminary matters (such
as hearings on motions to exclude expert testimony).
In administrative adjudications, the administrative law judge generally decides
the case without a jury. In all of these proceedings, however, the distinction
between the functions of factfinder and presider is still vitally important.
When the judge in a bench trial makes a finding as the factfinder, her decision
is entitled to the deference due to any factfinder’s decision. When that same
judge makes a ruling as a matter of law, however, her ruling is subject to the
same scope of appellate review as any trial judge’s ruling on an issue of law.
The fact that the same person often plays two roles does not cause the line
between those roles to vanish. This duality of role is possible in large part
because the reviewing authority enforces the distinction between issues of fact
and issues of law. The different scopes of appellate review for issues of fact
and issues of law are discussed below, under the province of the reviewing
Because the presiding authority decides
all issues of law in the first instance (subject to appellate review), he or
she can make rulings on every category of legal issue that was discussed above
under the province of the factfinder. On the motions and arguments of
participating parties, and occasionally sua
sponte, a presiding authority will decide which substantive rules of law,
definitional rules, process rules, default inference rules, and relevant-factor
rules are in force within the jurisdiction and applicable in the proceeding.
Even in proceedings where the same person acts as both factfinder and presiding
official, a party may move for rulings of law concerning propositional issues
on which the same judge will ultimately make findings of fact. Just as in a
jury trial, such rulings on issues of law may constrain the judge’s discretion
In addition to these types of rulings,
a presiding authority must decide important issues of law concerning the
evidence proffered by the parties. Some evidentiary rules are process rules
–such as the rules governing the discovery of evidence (e.g., rules governing
the taking of depositions, the production of documents, and the conduct of
Other evidentiary rules are exclusionary rules, concerning which proffered
items of evidence are inadmissible in the proceeding. A judicial example is
Federal Rule of Evidence 702, which states the conditions under which expert
testimony may be admitted in federal courts, and which the U.S. Supreme Court
interpreted in its decision in Daubert.
In deciding whether to admit an expert’s opinion that (for example) the
plaintiff’s exposure in utero to a certain drug caused the plaintiff’s injury,
a federal trial judge must first decide whether the empirical and theoretical
basis for that opinion is sufficiently “reliable” and “relevant” to the case at
hand, and may decide whether its probative value is “substantially outweighed”
by the risk of misleading the factfinder.
If the trial judge decides to exclude the testimony, then the jury will not
hear that opinion and the judge as factfinder may not rely on it as evidence.
In administrative proceedings, the rules governing discovery and admissibility
of evidence generally are quite different than those in judicial proceedings.
These differences are often due to differences in institutional structure and
in the blend of epistemic and non-epistemic goals.
Another important type of ruling on
evidence assesses (as a matter of law) the “legal sufficiency” of the totality
of admissible evidence introduced into the case, and allocates to particular
parties the burden of producing that evidence. On every proposition at issue in
a judicial proceeding, there is a rule of law assigning to some party the
burden of producing evidence sufficient for a reasonable factfinder to find the
issue in that party’s favor.
The party that has the burden of producing or coming forward with evidence (for
short, the “burden of production”) must lose the contest over the proposition
if that party fails to produce enough evidence – in the form of real evidence,
documents, and testimony. The totality of admitted evidence that is relevant to
any proposition at issue must meet minimal sufficiency requirements before the
presiding judge will present the proposition to the factfinder. Rulings on
sufficiency create therefore a threshold of reasonableness before the exercise
of the factfinder’s discretion. For example, in a tort complaint alleging that
the defendant negligently caused the plaintiff’s injury, the plaintiff has the
burden of production on a number of propositions, including the propositions
that the defendant in fact engaged in negligent conduct and that the
defendant’s negligence in fact caused the plaintiff to suffer some injury. If
the plaintiff fails to produce what the courts consider legally sufficient
evidence to support findings for the plaintiff on these issues, then the trial
judge should decide those issues against the plaintiff “as a matter of law.”
Although this example is from a judicial proceeding, administrative
adjudications may have similar rules about sufficiency.
In judicial civil cases, parties can
raise the issue of the legal sufficiency of the evidence at various times,
through a number of different motions. First, a party can obtain such a ruling
in a motion for summary judgment, before the trial begins.
Second, a moving party can obtain a ruling at trial, after the non-moving party
has had an opportunity to produce the evidence. The issue may be raised then by
a motion for directed verdict or for judgment as a matter of law.
Third, the motion can be renewed if the factfinder reaches an unfavorable
finding, in a motion for judgment notwithstanding the verdict or a renewed
motion for judgment as a matter of law.
Each motion, brought at a different time in the proceeding, may require the
moving party to establish slightly different factual predicates, but the
essence of the argument is the same: that as a matter of law the non-moving
party will fail, or has failed, to satisfy its burden of producing evidence
that the law considers minimally sufficient.
It is also possible for a court to
decide an issue as a matter of law for
the party who has the burden of
production, if the party successfully argues that the evidence produced is so
overwhelming that any reasonable jury would have to find for that party. It is
far more common, however, for a court to enter judgment as a matter of law against the party with the burden of
production, because of a deficiency in the evidence. This is primarily because
it is easier to devise and apply rules of law about when evidence is missing or
deficient, than about when produced evidence compels an inference.
Province of the Reviewing Authority
Just as the presiding authority
oversees the work of the factfinder, the reviewing authority oversees the work
of the presiding authority. In the case of judicial proceedings, the reviewing
authority is an appellate court.
In the case of an administrative proceeding, the reviewing authority may be
either an administrative body or a court conducting judicial review of the
final administrative action. The general rules governing the relative authority
of the reviewer over the decisions of the presider are called the “scope” or
standard of review.
When it comes to reviewing issues of
fact, the reviewing authority is required by law to be extremely deferential.
When the factfinder is a jury, strictly speaking the appellate court does not
review the verdict directly. Rather, any challenge to a jury verdict generally
must arise first as a motion to the trial judge, and then any appeal challenges
the trial judge’s ruling on that motion. This is not a matter of mere
semantics. The scope of review that is due to the trial court’s ruling depends
upon the nature of the motion made to the trial judge – which may be reviewed
for “abuse of discretion” or reviewed “de novo,” as discussed below.
Direct appellate review of a finding of
fact does occur when the presiding authority is also the factfinder. For
example, in civil bench trials in federal court, the trial judge must make
findings of fact on the basis of the evidentiary record.
Appellate review of those findings of fact uses the “clearly erroneous”
That is, the appellate court is not permitted to substitute its own assessment
of the evidence for that of the trier of fact, so long as it is not clear that
the trier of fact has committed a mistake.
Judicial review of administrative
factfinding is similarly deferential. In general, the reviewing court must
respect the findings of fact of the administrative agency unless those findings
are “arbitrary [or] capricious” or “unsupported by substantial evidence” in the
In practice, these two standards may come to much the same thing.
The scope or standard of review for
many other rulings of the presider is the “abuse-of-discretion” standard, which
is also very deferential. As long as the presiding authority’s ruling is not
“manifestly erroneous,” the reviewing authority has the obligation to let the
ruling stand. Often the type of ruling at issue involves the appropriateness of
applying a correctly stated rule of law to the particular circumstances of a
specific case, and the reviewing authority, which does not observe those
circumstances first-hand, is not the best decider. Abuse-of-discretion review
is applicable, for example, to trial court decisions to exclude proffered
evidence because it is potentially prejudicial or needlessly cumulative.
Not all abuse-of-discretion review,
however, can be justified on the rationale that the presider is in a better
position to assess the circumstances than the reviewer is. The
abuse-of-discretion standard of review applies to federal trial court rulings
to exclude expert testimony for failure to satisfy Federal Rule of Evidence
In Daubert matters, however,
appellate courts are often in as good a position as the trial court to decide
questions about the reliability and relevance of proffered scientific
testimony. In such situations, a more complicated justification, in terms of
non-epistemic as well as epistemic policies, would be needed to justify
abuse-of-discretion review instead of de novo review. Arguing the merits of
such a justification, however, must remain beyond the scope of this essay.
With respect to “pure” issues of law,
appellate review of trial court decisions or judicial review of administrative
decisions is generally “de novo”: that is, the presiding authority’s rulings
are subject to review without any deference to the presiding authority’s
If the presiding authority stated a rule of law incorrectly, or applied it in
such a way that it is clear that the presiding authority misunderstood the law,
then the reviewing authority simply reverses such a ruling and orders an
appropriate remedy (which may include sending the case back to the trial court
for re-trial or remanding the matter to the administrative agency). An example
is a ruling that the evidence produced by a party who has the burden of
production is legally insufficient. The argument that a party has failed to
meet its burden of production raises an issue of law on which the moving party
is entitled to de novo review by the appellate court. One justification for
this in a judicial setting is that otherwise a trial judge would have undue
power to deprive a plaintiff of her right to a trial by jury. A broader
justification is to enable appellate courts to create rules about minimal
sufficiency of evidence that they can then enforce uniformly across all trial
courts in the jurisdiction.
In general, therefore, reviewing
authorities do not have the power to make decisions that are reserved to
factfinders, and have precisely defined powers to set aside decisions made by
presiding officers. Unless a proposition at issue can be resolved as a matter
of law, if an appellate court decides to disregard a verdict or finding that is
essential to the case, then the case must be sent back to the trial court to be
re-tried. Similarly, a reviewing authority generally does not have the power to
make administrative decisions, in the place of the agency. If the reviewing
court vacates an administrative agency’s action because it was arbitrary or
capricious, then the matter must be remanded to the agency for further action.
Factfinding in a legal context is a
highly structured process in which there are distinct decision-making roles.
Each of the three principal roles (factfinder, presiding authority, and
reviewing authority) has a significant zone of discretion in which to operate,
yet decisions within that zone are constrained by decisions within the other
two provinces. The factfinder’s zone of discretion centers on making findings
about issues of fact, but rules of law often take propositional issues away
from the factfinder and give them to the presiding or reviewing authority. The
reviewing authority enjoys the greatest discretion in deciding what the rules
of law are, but this discretion is constrained by rules identifying issues of
fact reserved for the factfinder to decide. Of course, the discretion of the
reviewing authority is also constrained by the legislature (through statutes),
by administrative rulemaking, and by higher reviewing authority. Between the
factfinder and the reviewing authority lies the zone of discretion of the
presiding authority, which is least constrained in decisions about how to apply
the established legal rules to the particular circumstances of the individual
The strategy behind factfinding in law
is not so much to establish a methodology or logical model of inference to use
in assessing the quality of factfinding outcomes, but rather to maintain a
dynamic process of rule-governed decision-making, through which (it is hoped)
reasonable decision makers will come close enough to achieving the epistemic
objective over time. In deciding either to establish or to apply each of these
rules, there is a persistent need to balance epistemic and non-epistemic
aspects of the factfinding process, so that the process adequately serves its
pragmatic function within the governmental institution. There is a great deal
of theoretical work to be done in deciding the proper balance for each type of
decision, by each type of decision maker, and for each type of propositional
 This article
was originally published in APA Newsletter on Philosophy and Law, Fall 2003. Professor Walker wishes to acknowledge
the research support of Hofstra University in preparing this essay. Copyright
2003 Vern R. Walker.
 Science is
perhaps the leading contender to law in this regard, although scientists might
spend fewer resources than government does in overseeing the communal
enterprise of scientific factfinding. Indeed, much of the oversight of
scientific factfinding is performed by government – through such institutions
as the National Institutes of Health, the Centers for Disease Control and
Prevention, the Environmental Protection Agency, the Food and Drug
Administration, and so forth.
 See the Federal Insecticide, Fungicide,
and Rodenticide Act, 7 U.S.C. 136d (2000), and the Federal Food, Drug, and
Cosmetic Act, 21 U.S.C. §§ 342(a)(2)(B), 346a(a) (2000), as well as
 E.g., Dan
B. Dobbs, The Law of Torts § 143, p. 334 (West Group; St. Paul, Minn.
 See, e.g., New York Pattern Jury
Instructions – Civil 2:10 (2002).
 See Vern R. Walker, Preponderance, Probability and Warranted Factfinding, 62 Brooklyn L. Rev. 1075, 1076 (1996).
 See Fleming James, Jr., Geoffrey C.
Hazard, Jr., & John Leubsdorf, Civil
Procedure § 7.5, pp. 323-24 (Little, Brown & Co.; Boston, 4th
 E.g., Ahn v. Kim, 678 A.2d 1073 (N.J.
 See, e.g., Restatement Second of Torts §§ 291-93 (1965).
 On hearings
on preliminary questions, see Federal Rule of Evidence 104; Christopher B. Mueller & Laird C.
Kirkpatrick, Evidence §§ 1.10-1.13, pp. 31-49 (Aspen; New York, 3d ed.
 See generally
Federal Rules of Civil Procedure 26 et seq.; James et al., supra note 6, Chapter 5, pp. 231-91.
 Daubert v.
Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). For more discussion of
Federal Rule of Evidence 702 and the admissibility of expert testimony after Daubert, see [cite to other articles in
 Daubert, supra note 11, 509 U.S. at 589-95; Federal Rules of Evidence 403,
 For a general
discussion of sufficiency of the evidence, see James et al., supra note 6, at § 7.19, pp. 357-65.
 E.g., Federal Rule of Civil Procedure
 E.g., Federal Rule of Civil Procedure
 E.g., Federal Rule of Civil Procedure
 This essay
does not discuss the distinctions among different layers of appellate review,
between appeal as of right and by writ of certiorari, or between appeals and
certified questions. While such procedural distinctions add complexity to the
factfinding process, the general theme of this essay does not require examining
 Federal Rule
of Civil Procedure 52(a).
 James et al.,
supra note 6, at § 12.9, pp. 668-74.
Administrative Procedure Act, 5 U.S.C. § 706(2)(A), (E) (2000).
 See, e.g., Association of Data
Processing Service Organizations, Inc. v. Board of Governors of the Fed.
Reserve System, 745 F.2d 677, 683 (D.C. Cir. 1984).
 Federal Rule
of Evidence 403; Mueller & Kirkpatrick, supra
note 9, § 4.9, pp. 174-75.
Electric Company v. Joiner, 520 U.S. 114 (1997).
 I leave aside
here such complications as the U.S. Supreme Court’s doctrine announced in
Chevron, U.S.A. Inc. v. NRDC, 467 U.S. 837 (1984) (announcing a deferential
approach to an agency’s interpretation of a statute that Congress has entrusted
to the agency to administer).