Summary
explaining that the true measure of a court's willingness to defer may depend, in the final analysis, on the persuasiveness of the agency's interpretation, given all the attendant circumstances
Summary of this case from Stowell v. Secretary of Health Human ServicesOpinion
No. 92-1945.
Heard February 1, 1993.
Decided March 3, 1993.
Robert D. Kamenshine, Atty., Civil Div., U.S. Dept. of Justice, with whom Stuart M. Gerson, Asst. Atty. Gen., Washington, DC, A. John Pappalardo, U.S. Atty., Boston, MA, Douglas N. Letter, Atty., Civil Div., Margaret J. Porter, Chief Counsel, and Leslie Kux, Associate Chief Counsel, U.S. Food Drug Admin., Washington, DC, were on brief, for appellant.
Robert Ullman, with whom Jacob Laufer, Steven Shapiro, and Bass Ullman, New York City, were on brief, for appellee.
Appeal from the United States District Court for the District of Massachusetts.
Before SELYA, Circuit Judge, ALDRICH, Senior Circuit Judge, and CYR, Circuit Judge.
The government seized, and seeks to condemn, twenty-nine cartons of undiluted black currant oil (BCO), in capsule form, owned by claimant-appellee Oakmont Investment Co. (Oakmont), alleging that BCO is a food additive of questionable safety. Because we believe that encapsulated BCO, intended to be ingested as purchased, cannot properly be termed a food additive as defined in the Federal Food, Drug, and Cosmetic Act (the Act), as amended, 21 U.S.C. §§ 301 et seq. (1988), we affirm the district court's dismissal of the government's in rem complaint.
I. BACKGROUND
On October 11, 1988, the United States Food and Drug Administration (FDA) seized 200 bottles of encapsulated BCO, packed in twenty-nine cartons, and brought an in rem action contending that, under 21 U.S.C. § 342(a)(2)(C), the capsules should be condemned as "adulterated" food because they contain a "food additive," the BCO, that Oakmont had not proven to be safe.
At the ensuing bench trial, certain facts were uncontradicted. BCO is a liquid obtained by squeezing black currant berry seeds. It is composed of polyunsaturated fatty acids. In its pure liquid form, it can be ingested by the spoonful as a dietary supplement. However, Oakmont markets BCO in capsules which are to be swallowed whole. The capsules contain pure BCO — nothing more. They are made from gelatin and glycerin (or an equivalent plasticizer) and have no independent nutritional value. Rather, a capsule serves a dual purpose as a container (enabling consumers to ingest predetermined quantities of BCO in solid form) and as a prophylactic (protecting the BCO from rancidity).
On these and other facts, the district court dismissed the government's complaint and ordered the capsules released. See United States v. 29 Cartons, Etc., 792 F. Supp. 139, 142 (D.Mass. 1992). The court reasoned that when, as in this case, BCO comprises the only active ingredient within a gelatin capsule, it can properly be classified as a "food," but not as a "food additive." See id. at 141-42. Accordingly, the FDA erred in seizing the bottles on the ground that they "allegedly contain an unsafe food additive." Id. at 142.
When the FDA appealed, the district court stayed its release order.
II. THE REGULATORY LANDSCAPE
To put this case into workable perspective, we first review the relevant statutory provisions. The Act defines "food" as:
(1) articles used for food or drink for man or other animals, (2) chewing gum, and (3) articles used for components of any such article.
21 U.S.C. § 321(f). The FDA concedes that pure BCO (sold, say, as a bottled liquid) falls within section 321(f)(1) and is, therefore, "food." Substances classified as "food" are presumed safe. Thus, the FDA can prevent sale of bottled BCO or any other "food" only if it proves by a preponderance of the evidence that the food is "injurious to health." 21 U.S.C. § 342(a)(1); see, e.g., United States v. Lexington Mill Elevator Co., 232 U.S. 399, 411, 34 S.Ct. 337, 340, 58 L.Ed. 658 (1914); United States v. An Article of Food [FoodScience Labs., Inc.], 678 F.2d 735, 741 n. 3 (7th Cir. 1982) (Cudahy, J., concurring). Although the FDA suspects that BCO may be unhealthful, it is unable at the present time to translate this suspicion into legally competent proof.
In addition to regulating the sale of food per se, the Act contains provisions anent food additives. These provisions are designed to protect consumers against the introduction of untested and potentially unsafe substances, such as flavor, texture, or preservative agents, into food. A gloss was added to the treatment of food additives in 1958. See Pub.L. No. 85-929, 72 Stat. 1784 (1958) (codified in scattered sections of 21 U.S.C.). Unlike section 342(a)(1), which places the burden of proving injuriousness upon the government in respect to foods, the food additives amendment allocates the burden quite differently: the FDA can prevent the sale of products containing a food additive unless and until the processor shows that the substance, when added to food, is generally recognized as safe (in the vernacular, "GRAS"). See S.Rep. No. 2422, 85th Cong., 2d Sess. (1958), reprinted in 1958 U.S.C.C.A.N. 5300, 5301-02 (explaining the processor's burden "of proving that a newly discovered substance which . . . [is] add[ed] to the food we eat is safe"). Thus, in contrast to the Act's treatment of "food," any substance that meets the Act's definition of a "food additive" is presumed to be "unsafe" under 21 U.S.C. § 348 until the FDA, or more particularly, the Commissioner of Food and Drugs, has promulgated a regulation prescribing conditions assuring safe use. See 21 U.S.C. § 348(a)(2); 21 CFR § 5.10(a)(1) (1992).
The 1958 amendment defines a food additive in pertinent part as:
any substance the intended use of which results or may reasonably be expected to result, directly or indirectly, in its becoming a component or otherwise affecting the characteristics of any food (including any substance intended for use in producing, manufacturing, packing, processing, preparing, treating, packaging, transporting, or holding food; and including any source of radiation intended for any such use), if such substance is not generally recognized, among experts qualified by scientific training and experience to evaluate its safety, as having been adequately shown through scientific procedures . . . to be safe under the conditions of its intended use. . . .
21 U.S.C. § 321(s). To be labeled a food additive, then, a substance must (1) be intended, or reasonably expected, to become a component of food or to otherwise affect the characteristics of food, and (2) not be GRAS.
The Act thus creates a distinction between foods and food additives which has meaningful consequences for purveyors and for the public. The distinction also significantly affects the ease with which the FDA may regulate a substance's sale.
III. THE ISSUE
This appeal revolves around the question of whether the FDA or Oakmont must carry out the research necessary to show that BCO is, or is not, GRAS. The issue reduces to whether pure BCO, when sold in encapsulated form, must be regulated as a "food" within the meaning of section 321(f) or as a "food additive" within the meaning of section 321(s).
The meat of the parties' disagreement lies in their differing interpretations of that portion of the Act which states that a substance can be a food additive if its intended use results, or may be expected to result, "in its becoming a component or otherwise affecting the characteristics of any food." 21 U.S.C. § 321(s). The FDA reads the quoted language as creating two independent and disjunctive standards: to satisfy the first prong of the food additive definition, a substance must either (1) be a component of food, or (2) otherwise affect the characteristics of food. Because each constituent part or element of a food (that is, each "component") necessarily affects the food's characteristics, the FDA considers every component, at least potentially, see infra note 3, to be a food additive. Drawing on this interpretation, the FDA asserts that the seized capsules are composed of three consumable components — BCO, gelatin, and glycerin — and that, therefore, each of these three ingredients is subject to potential regulation as a food additive.
The district court bifurcated the trial and, during the initial phase, determined only that BCO does not meet the first prong of the bipartite food additive definition. Thus, the district court had no occasion to reach the second prong, viz., whether BCO is GRAS. Hence, that issue is not before us.
In the FDA's view, the second of the two independent standards confers potential food additive status on substances that, while they are not constituent parts of a food, may nevertheless have deleterious effects on food. One example might be chemicals used in packaging food.
We use the adjectival modifier "potential" because gelatin and glycerin are concededly GRAS. Hence, these components cannot be classified as food additives because neither can fulfill the definition's second prong.
As Oakmont parses the statute, it creates only a single, unitary food additive standard. The phrase "or otherwise affecting the characteristics of any food" signals that a component is potentially a food additive only if it affects the characteristics of some food to which it is added. Unlike the FDA's interpretation, Oakmont's interpretation attaches no significance to a substance's mere presence as a component of a whole. It focuses instead on the substance's affirmative use in a way that affects food.
Applying its interpretation of the statute to the facts at bar, Oakmont argued below, as it does here, that the BCO contained in the seized capsules is itself a food and not a component of some other food, that it is intended so to serve, and that its sale in a convenient carrier medium does not transmogrify it into a food additive. In holding that food is defined "by its 'use for food,'" 29 Cartons, 792 F. Supp. at 141 (quoting 21 U.S.C. § 321(f)), whereas a food additive is defined by its effect on another substance, see id., the district court substantially adopted Oakmont's reading of the law and its focus on a substance's intended function.
In specific terms, then, we must determine whether, as the FDA would have it, any element of any substance that has more than one component may be branded a food additive, or, rather, whether, as Oakmont urges and the court below believed, such treatment should be reserved for elements which, when so added, effect a change (or, at least, could be expected to effect a change) in some other active ingredient.
IV. FOOD FOR THOUGHT
The Seventh Circuit has recently grappled with a factually similar case presenting this very issue. See United States v. Two Plastic Drums, Etc., 984 F.2d 814 (7th Cir. 1993). Employing a perspicacious analysis of the Act's text and legislative history, the court rejected the FDA's notion that all components of a substance are necessarily food additives. The court observed that the "'or otherwise'" phrase contained in the statutory definition of a food additive targets only those components that "have the purpose or effect of altering a food's characteristics." Id. at 818. The subsequent enumeration of sample food additives, describing each substance by its "function or by [its] effect on food," makes it clear that an additive must stimulate some change in a food to which it is added. Id. at 818. Turning to the legislative history, the court observed that the FDA's broad definition of a food additive, which would apply to all components, even a substance which comprises the only active ingredient of the whole, subverts congressional purpose. Blurring the distinction between food additives and food in this way would permit the agency to tilt a delicately balanced statutory scheme that allocates the burden of proving an additive's safety to the processors while leaving the burden of establishing a food's safety with the FDA. See id. at 819.
The Seventh Circuit also recognized the incongruity of categorizing a food's single active component as an additive. Because "that single component does not affect the characteristics of the food in question — rather, it constitutes the food," id. at 818, it has no place within "the common understanding of an additive, defined by Webster as 'a substance added to another . . . to impart or improve desirable properties or suppress undesirable properties.'" Id. at 818 n. 3 (citation omitted). Thus, in order to qualify as a food additive, a component must be added to a food in order to change that food's properties. See id. at 819. On that basis, pure BCO, in capsule form, is not a food additive. See id. at 820.
Judges should hesitate to write lengthy opinions merely for the sake of committing their own prose to posterity. Given the existence of a cogent, well-reasoned, eminently correct opinion closely on point, we embrace it. We will, therefore, affirm the judgment below for substantially the reasons elucidated in Two Plastic Drums. We pause, nevertheless, to essay a few additional observations.
First: We are reluctant to believe that Congress traffics in absurdities. Since it defies common sense to say that a substance can be a "food additive" when there is no (other) food to which it is added, we think that the FDA's reading of the Act is nonsensical, and, hence, must be incorrect. Moreover, classifying BCO as a "component" merely because it is combined with two totally inert substances serving collectively as a carrier medium would itself create a bizarre paradox: as the Seventh Circuit noted, "to hold that BCO is a component of the dietary supplement would be to find that BCO is a component of itself." Two Plastic Drums, 984 F.2d at 817.
Second: In the FDA's estimation, a processor's "subjective intent" that only one of a product's components constitutes the food is irrelevant because "it is the objective intended use, i.e., the intent to combine two or more components, that counts." Appellant's Brief at 11. But, this harangue misses the mark. We fully agree that a processor's subjective determination of what constitutes a food is not determinative in cases of this stripe — but neither is the naked fact that more than one component has been combined. In the final analysis, what counts is the use of an ingredient for its effect on food. Here, from an objective standpoint, BCO is not being used for its effect on gelatin and glycerine. Thus, contrary to the FDA's loudly expressed fears, eschewing its rendition of the statutory text will not supplant objectivity with subjectivity.
Moreover, if the FDA worries that processors may muck the statutory classifications with convenient recitals of subjective intent, we question the agency's espousal of a rule that would "arbitrarily classify a substance as either food or food additive by how it is marketed rather than by the nature and use of the substance itself." Two Plastic Drums, 984 F.2d at 819. In the words of Sir Francis Bacon, the FDA's suggested "remedy is worse than the disease."
Third: The FDA also maintains that because "the ingredients of multi-ingredient food products, such as cake mixes," indisputably fall within the food additive definition, the statute could not possibly contain a "requirement that a substance must be added to a preexisting food, which it must be shown actually to affect." Appellant's Brief at 9. We disagree. Cake mixes are foods composed of many interacting food additives, each with its particular effect on the whole. Absent any one ingredient, the concoction remains a cake mix, albeit one that may be short on sweetness or lumpy in texture. In that sense, cake mixes and products of that ilk are a far cry from a dietary supplement composed of a single active ingredient. What differentiates this case is that, if the BCO is removed, one is left with nothing but an empty capsule.
We do not quarrel with those courts that have held, when confronted with multi-ingredient products containing two or more active ingredients, that each active ingredient is potentially a food additive. See, e.g., United States v. 45/194 Kg. Drums, Etc., 961 F.2d 808, 812 n. 3 (9th Cir.), cert. denied, ___ U.S. ___, 113 S.Ct. 375, 121 L.Ed.2d 287 (1992); FoodScience, 678 F.2d at 738; United States v. 41 Cases, Etc., 420 F.2d 1126, 1130 (5th Cir. 1970).
Fourth: We think it advisable to mention the FDA's insistence, citing Chevron U.S.A. Inc. v. NRDC, Inc., 467 U.S. 837, 843, 104 S.Ct. 2778, 2782, 81 L.Ed.2d 694 (1984), that we must obey its interpretation of the Act. In our estimation, the purely legal question facing us in this case presents no occasion for deference. In this realm of judicial expertise, the courts, not the agency, have the last word. See id. at 843 n. 9, 104 S.Ct. at 2782 n. 9 ("The judiciary is the final authority on issues of statutory construction . . . ."); BATF v. FLRA, 464 U.S. 89, 98 n. 8, 104 S.Ct. 439, 445 n. 8, 78 L.Ed.2d 195 (1983) (observing that "deciding what a statute means" is "the quintessential judicial function"); FTC v. Colgate-Palmolive Co., 380 U.S. 374, 385, 85 S.Ct. 1035, 1042, 13 L.Ed.2d 904 (1965) (holding that "legal standard[s] . . . must get their final meaning from judicial construction"); Wilcox v. Ives, 864 F.2d 915, 924 (1st Cir. 1988) (quoting BATF v. FLRA, supra).
At any rate, the true measure of a court's willingness to defer to an agency's interpretation of a statute "depends, in the last analysis, on the persuasiveness of the interpretation, given all the attendant circumstances." Massachusetts Dep't of Educ. v. United States Dep't of Educ., 837 F.2d 536, 541 (1st Cir. 1988). "The simple fact that the agency has a position, in and of itself, is of only marginal significance." Mayburg v. Secretary of HHS, 740 F.2d 100, 106 (1st Cir. 1984). When, as now, a court is persuaded neither by "the validity of [the agency's] reasoning," nor by the interpretive fit between the agency's rendition, on the one hand, and the language and structure of the statute, on the other hand, a court should not defer. Skidmore v. Swift Co., 323 U.S. 134, 140, 65 S.Ct. 161, 164, 89 L.Ed. 124 (1944).
The longevity of an agency's position is often significant in assaying the degree of deference owed to it. See Bowen v. Georgetown Univ. Hosp., 488 U.S. 204, 212, 109 S.Ct. 468, 473, 102 L.Ed.2d 493 (1988) (refusing to apply Chevron deference to "agency litigating positions that are wholly unsupported by regulations, rulings, or administrative practice"); Skidmore, 323 U.S. at 140, 65 S.Ct. at 164 (acknowledging the value of "consistency" in respect to gauging persuasiveness). Here, the FDA's position is of recent vintage. Indeed, the original complaint in this action pinned food additive status not on BCO but on gamma linolenic acid, BCO's fatty acid constituent. And, in a prior case involving blue-green algae in gelatin capsule form, the FDA argued that the blue-green algae was an additive because it was to be consumed with water or other foods or liquids, not because of its placement in gelatin capsules. See United States v. Articles of Food [Blue-Green Algae], No. 83-180-FR, 1984 WL 1981, at *3-*4 (D.Or. Nov. 8, 1984).
V. CONCLUSION
We need go no further. The proposition that placing a single-ingredient food product into an inert capsule as a convenient method of ingestion converts that food into a food additive perverts the statutory text, undermines legislative intent, and defenestrates common sense. We cannot accept such anfractuous reasoning.
Affirmed.