Language of document : ECLI:EU:C:2020:552

Provisional text

OPINION OF ADVOCATE GENERAL

BOBEK

delivered on 9 July 2020(1)

Case C526/19

Entoma SAS

v

Ministre de l’Économie et des Finances,

Ministre de l’Agriculture et de l’Alimentation

(Request for a preliminary ruling from the Conseil d’État (Council of State, France))

(Reference for a preliminary ruling — Food safety — Novel foods and food ingredients — Regulation (EC) No 258/97 — Article 1(2) — Food ingredients isolated from animals — Whole insects intended for human consumption — Interpretation of the material scope of the regulation)






I.      Introduction

1.        Are insects novel foods? Over the course of human history, they are certainly not. However, in the eyes of EU law, the answer is said to be less clear. It may perhaps be assumed that, up until 15 May 1997, the relevant date foreseen by Regulation (EC) No 258/97, (2) insects had not ‘been used for human consumption to a significant degree within the Community’. Nevertheless, is it possible for whole mealworms, locusts and crickets to be subsumed equally under the second part of the definition of novel foods, enquired about in the present case, namely that they are ‘food ingredients isolated from animals’?

2.        In my view, they cannot. There is nonetheless the explicit invitation issued by the French and Italian Governments urging the Court to fill in judicially what they consider to be a gap left by the EU legislature back in 1997. One cannot ignore an invitation, a fortiori when it concerns such culinary delights as those addressed by the present case. However, one can, and in the present case one must, politely decline that invitation, pointing out the limits of what might still be called ‘judicial interpretation’ of a clear provision of secondary legislation, and what becomes its ex post rewriting.

II.    Legal framework

A.      EU law

1.      Regulation No 258/97

3.        Recital 1 of Regulation No 258/97 of the European Parliament and of the Council concerning novel food and novel food ingredients stated that: ‘… differences between national laws relating to novel food or food ingredients may hinder the free movement of foodstuffs … [and] create conditions of unfair competition, thereby directly affecting the functioning of the internal market’.

4.        Recital 2 read as follows: ‘…in order to protect public health, it is necessary to ensure that novel foods and novel food ingredients are subject to a single safety assessment through a Community procedure before they are placed on the market within the Community …’

5.        Pursuant to the original version (3) of Article 1 of Regulation No 258/97:

‘1.      This Regulation concerns the placing on the market within the Community of novel foods or novel food ingredients.

2.      This Regulation shall apply to the placing on the market within the Community of foods and food ingredients which have not hitherto been used for human consumption to a significant degree within the Community and which fall under the following categories:

(a)      foods and food ingredients containing or consisting of genetically modified organisms within the meaning of Directive 90/220/EEC;

(b)      foods and food ingredients produced from, but not containing, genetically modified organisms;

(c)      foods and food ingredients with a new or intentionally modified primary molecular structure;

(d)      foods and food ingredients consisting of or isolated from micro-organisms, fungi or algae;

(e)      foods and food ingredients consisting of or isolated from plants and food ingredients isolated from animals, except for foods and food ingredients obtained by traditional propagating or breeding practices and having a history of safe food use;

(f)      foods and food ingredients to which has been applied a production process not currently used, where that process gives rises to significant changes in the composition of structure of the foods or food ingredients which affect their nutritional value, metabolism or level of undesirable substances.

3.      Where necessary, it may be determined in accordance with the procedure laid down in Article 13 whether a type of food or food ingredient falls within the scope of paragraph 2 of this Article’.

6.        Under Article 3(1) of Regulation No 258/97:

‘Foods and food ingredients falling within the scope of this Regulation must not:

–        present a danger for the consumer,

–        mislead the consumer,

–        differ from foods or food ingredients which they are intended to replace to such an extent that their normal consumption would be nutritionally disadvantageous for the consumer.’

7.        Pursuant to Article 12 of the same regulation:

‘1.      Where a Member State, as a result of new information or a reassessment of existing information, has detailed grounds for considering that the use of a food or a food ingredient complying with this Regulation endangers human health or the environment, that Member State may either temporarily restrict or suspend the trade in and use of the food or food ingredient in question in its territory. It shall immediately inform the other Member States and the Commission thereof, giving the grounds for its decision.

2.      The Commission shall examine the grounds referred to in paragraph 1 as soon as possible within the Standing Committee for Foodstuffs; it shall take the appropriate measures in accordance with the procedure laid down in Article 13. The Member State which took the decision referred to in paragraph 1 may maintain it until the measures have entered into force.’

2.      Regulation 2015/2283

8.        Regulation (EU) 2015/2283 of the European Parliament and of the Council of 25 November 2015 on novel foods has repealed Regulation No 258/97 as from 1 January 2018.

9.        Recital 6 of Regulation 2015/2283 reads as follows:

‘The existing definition of novel food in Regulation (EC) No 258/97 should be clarified and updated with a reference to the general definition of food provided for in Regulation (EC) No 178/2002 of the European Parliament and of the Council.’

10.      Recital 8 of Regulation 2015/2283 states:

‘The scope of this Regulation should, in principle, remain the same as the scope of Regulation (EC) No 258/97. However, on the basis of scientific and technological developments that have occurred since 1997, it is appropriate to review, clarify and update the categories of food which constitute novel foods. Those categories should cover whole insects and their parts …’.

11.      Article 2(1) of Regulation 2015/2283 provides that ‘this Regulation applies to the placing of novel foods on the market within the Union.’

12.      Among the definitions laid down in Article 3(2) of Regulation 2015/2283, ‘novel food’ is defined, in paragraph (a), as follows:

‘any food that was not used for human consumption to a significant degree within the Union before 15 May 1997, irrespective of the dates of accession of Member States to the Union, and that falls under at least one of the following categories:

(v)      food consisting of, isolated from or produced from animals or their parts, except for animals obtained by traditional breeding practices which have been used for food production within the Union before 15 May 1997 and the food from those animals has a history of safe food use within the Union;

…’

13.      Article 35(2) of Regulation 2015/2283, which is entitled ‘transitional measures’ provides that:

‘Foods not falling within the scope of Regulation (EC) No 258/97, which are lawfully placed on the market by 1 January 2018 and which fall within the scope of this Regulation may continue to be placed on the market until a decision is taken in accordance with Articles 10 to 12 or Articles 14 to 19 of this Regulation following an application for authorisation of a novel food or a notification of a traditional food from a third country submitted by the date specified in the implementing rules adopted in accordance with Article 13 to 20 of this Regulation respectively, but no later than 2 January 2020.’

III. Facts, national proceedings and the question referred

14.      Entoma (‘the appellant’) is an undertaking marketing products consisting of mealworms, locusts and crickets intended for human consumption in the form of whole insects.

15.      By order of 27 January 2016, the préfet de police de Paris (Prefect of Police, Paris, France) suspended the placing on the market of those products and ordered their withdrawal until authorisation to place them on the market was obtained, following an assessment intended to demonstrate that they do not present any danger for the consumer.

16.      The appellant introduced an annulment action against that order before the tribunal administratif de Paris (First Instance Administrative Court, Paris, France). By judgment of 9 November 2017, the latter dismissed the action. On 22 March 2018, the Cour administrative d’appel de Paris (Second Instance Administrative Court, Paris, France) rejected the appeal brought by the appellant.

17.      Before the referring court, the Conseil d’État (Council of State, France), seised by appeal on points of law, the appellant notably argued that the second instance court erred in law in holding that the marketing of its products was subject to Regulation No 258/97. However, whole insects consumed as such were excluded from the latter’s scope of application. Whole insects benefit from the transitional measures foreseen under Article 35(2) of Regulation 2015/2283. For its part, the Ministry for Economy and Finance claimed that Regulation No 258/97, which pursued a public health objective, also applied to whole insects since the consumption of the latter is as risky as the consumption of food ingredients isolated from animals.

18.      In this factual and legal context, the Conseil d’État (Council of State, France) decided to stay proceedings and to refer the following question to the Court for a preliminary ruling:

‘Is Article 1(2)(e) of the Regulation of 27 January 1997 to be interpreted as including within its scope foods consisting of whole animals intended to be consumed as such or does it apply only to food ingredients isolated from insects?’

19.      Written observations have been submitted by the appellant, the French and Italian Governments, as well as the European Commission.

IV.    Assessment

20.      This Opinion is structured as follows. While there is little doubt that whole animals, including insects, are indeed now covered by the new Regulation 2015/2283 (A), that was, on its text, clearly not the case with regard to Regulation No 258/97 (B). Moreover, the purposive enlargement of the scope of a secondary law instrument against its clear wording, as essentially suggested by the French and Italian Governments, has, in my view, little to do with interpreting an existing text, but pertains to effectively writing a new one (C).

A.      Article 3(2)(a)(v) of Regulation 2015/2283

21.      In a somewhat unusual way, I shall start with legislation that is not applicable in the present case: Regulation 2015/2283, which pursuant to its Article 36 entered into force on 1 January 2018, and the definition of novel foods contained therein.

22.      The new definition of ‘novel food’ contained in Article 3(2)(a) of Regulation 2015/2283 sets two cumulative criteria: (i) any food that was not used for human consumption to a significant degree within the Union before 15 May 1997, and (ii) that falls under at least one of the ten categories listed in Article 3(2)(a).

23.      From that list, number (v) is relevant for the present case. That category includes ‘food consisting of, isolated from or produced from animals or their parts, except for animals obtained by traditional breeding practices which have been used for food production within the Union before 15 May 1997 and the food from those animals has a history of safe food use within the Union.’ (4)

24.      Insects are perhaps not the obvious choice for human consumption. However, they are (invertebrate) animals. Whole mealworms, locusts and crickets are thus clearly food consisting of, or food produced from, animals. Furthermore, jokes about (non)intentional consumption aside, insects were admittedly not used for human consumption to a significant degree within the Union before 15 May 1997. Moreover, recital 8 explicitly states that ‘the categories of food which constitute novel foods … should cover whole insects and their parts …’.

25.      Thus, under the new regime, whole insects destined for human consumption are included under Regulation 2015/2283. The reason why that regime, which was not applicable in the present case, is mentioned first, is twofold.

26.      First, the French and the Italian Government maintain essentially that the new regime and the new definitions simply clarify what was already the case before under Regulation No 258/97. Therefore, a comparison between the wording of the two is called for.

27.      Second, the relationship between the material scopes of both regulations is also of further temporal relevance. It follows from Article 35(2) of Regulation 2015/2283 that products falling within the material scope of the new regulation, but outside that of the previous regulation, which were lawfully on the market before 1 January 2018, could remain on the market until 2 January 2020, unless any other hypothesis referred to in that provision materialised. Thus, should it be established that the products at issue were not materially covered by the previous regulation but were nevertheless lawfully placed on the market at that time, it would follow that the marketing of those products could provisionally continue up until 2 January 2020.

B.      Article 1(2)(e) of Regulation No 258/97

28.      The referring court asks whether whole animals, in particular whole insects intended for human consumption, were already included within the material scope of Regulation No 258/97 in order to determine if, at the date of the facts in the main proceedings when Regulation No 258/97 was still in force, the marketing of the products at issue was subject to authorisation under Regulation No 258/97.

29.      According to the appellant and the Commission, it follows from the wording of that provision that foods made of whole animals were not covered by Article 1(2) of Regulation No 258/97. Therefore, whole insects did not fall under either the category of food ingredients isolated from animals (Article 1(2)(e) of the regulation), nor any other category listed in Article 1(2). The French and Italian Governments contest that conclusion on the basis of the general scheme and purpose of Article 1(2) of Regulation No 258/97.

1.      Text

30.      Article 1(2) of Regulation No 258/97 contained two cumulative conditions for foods or food ingredients to be qualified as novel and fall under that regulatory regime. First, the temporal condition: Regulation No 258/97 applied to the placing on the market, within the then Community, of foods and food ingredients which had not previously (i.e. before 1997) been used for human consumption to a significant degree within the Community. Second, there was the substantive condition: food and food ingredients needed to fall under one of the categories listed in Article 1(2).

31.      The first condition, the temporal one, is not questioned by the referring court. Although the evidential issues associated with who shall establish and how the ‘significant degree of consumption’ within the Community, which would include the Union as a whole, not just one Member State or its part, would no doubt be intriguing, they are not the subject matter of the present reference.

32.      Therefore, moving immediately to the substantive condition, what is to be understood by ‘food ingredients isolated from animals’?

33.      Neither the notions of ‘ingredients’ nor ‘isolated from’ were defined in the regulation. It is, however, settled case-law that the meaning and scope of terms for which EU law provides no definition must be determined by reference to their usual meaning in everyday language, while also taking into account the context in which they occur and the purposes of the rules of which they form part. (5)

34.      The Oxford English Dictionary defines ingredients as ‘any of the foods or substances that are combined to make a particular dish’ and isolate as ‘identifying something and dealing with it separately or, in chemistry biology, obtaining or extracting (for instance a compound) in a pure form.’ Furthermore, the Commission noted that the notion of ingredient has been defined in another EU measure as ‘any substance or product, including flavourings, food additives and food enzymes, and any constituent of a compound ingredient, used in the manufacture or preparation of a food and still present in the finished product, even if in an altered form’. (6)

35.      Thus, a food ingredient is, it might be noted in whichever of the official languages used, (7) a component of a larger, composite end product. It is normally not a product to be consumed in and by itself, but to be added in order to create other food or a particular dish. Certainly, I acknowledge that with regard to some substances, that line might be a bit blurry. There are a number of ingredients which can be eaten as such (for example, honey or sugar).

36.      That is, however, not the case for whole animals. In that sense, whole animals are unlikely to be an ingredient. At least for carnivore humans, they are the food, not a food ingredient. Since Article 1(2)(e) of Regulation No 258/97 clearly distinguished between food (products) and food ingredients and only covered the latter in relation to animals, it must be concluded that animals eaten as such, in their entirety or in some of their parts, could not be a food ‘ingredient’ within the meaning of Regulation No 258/97.

37.      Next, there is isolated from animals. (8) In contrast to, for example, ‘consisting of’, or ‘produced from’, (9) ‘isolated from’ refers to a process of extraction from the animal, be it the whole animal or its parts, hence further limiting the scope of that regulation in relation to animals. ‘Isolated from’ could be read in two ways: first, as a chemical, biological, or mechanical process, whereby essences, substrates, powders, in that sense indeed any type of ingredients, are extracted from animals. That would certainly be the more natural reading. Second, it might arguably also be possible, at a stretch, to read the process of isolation from as simply referring to the mechanical extraction of an ingredient from the body of an animal. In that latter sense, ‘isolated from’ could perhaps also mean being removed from an animal, thus referring effectively to a distinct part of an animal or an organ.

38.      However, on no possible reading of ‘isolated from’ would that refer to the whole animal, unless one were to create a tautology, in which whole animals are ‘isolated from’ whole animals. (10)

39.      Finally, combining the individual ingredients of the definition contained in Article 1(2)(e) of Regulation No 258/97 into one dish, food ingredients isolated from animals would, on its natural reading, refer to components deriving from animals that are added to other food. Put differently: (i) whole animals eaten as such were thus not included; (ii) parts of animals eaten as such were not included; (iii) whole animals used as an ingredient for the preparation of a dish, as far as that is possible, were not included; (iv) only specific parts or elements of animals used as an ingredient could be  included.

40.      Moving away from the abstract and using instead a specific example: what about the consumption of frogs’ legs? From my interpretation of what is meant by ‘food ingredients isolated from animals’, (11) the human consumption of frogs’ legs would not be caught by Article 1(2)(e) of Regulation No 258/97. Frogs’ legs are parts of animals to be eaten as such, not an ingredient isolated from animals. By contrast, a hypothetical frogs’ legs powder, or frog flour, if they indeed existed as ingredients isolated from frogs, could indeed be covered.

41.      The same must then be applicable to parts of insects. It is, a fortiori,  true of whole insects. Certainly, Regulation No 258/97 could have potentially applied to ingredients isolated from insects if ever used as components in other products. (12) However, on its text, it clearly did not apply to whole insects to be eaten as such, just as it was not supposed to apply to any other whole animals.

2.      Context

42.      Two elements of legislative context are worth noting: the internal system and logic of Regulation No 258/97 (1) and, as far as it can be ascertained, the historical context and legislative intent (2). Those two points provide an answer to the structural arguments raised by the French and Italian Governments, concerning the internal logic and coherence of the regulatory choices operated by the then Community legislature (3).

(1)    The internal logic: micro changes in structure

43.      A glance at the other specific categories of Article 1(2) that were supposed to compose the substantive element of the definition of novel food is quite telling. The original six categories (13) had a rather clear common denominator: the food or food ingredients were being modified either genetically (letters (a) and (b)), or at the molecular level (c), or the micro(biological) level (d), or were created by a new production process not currently used, giving rise to significant changes in the composition or structure of the food (f). In sum and simplified, it was clearly the micro changes in the organisms to be ultimately consumed by humans which were supposed to be caught by that instrument.

44.      Such an internal context and logic only confirms that the textual reading of Article 1(2)(e) of Regulation No 258/97 is correct: if the logic of Article 1(2) categories were to cover micro changes, it would be rather surprising if suddenly a category within such a list would be applicable to whole animals without those being in any way altered at that micro level. Thus, the overall structure of Article 1(2) explains rather well what the EU legislature might have sought after in letter (e), namely aiming precisely at what that latter states: food ingredients isolated (indeed, in that sense primarily biologically or chemically, or in any other way inducing changes at the micro-biological level) from animals.

(2)    The historical context and legislative intent

45.      As far as the historical context and legislative intent are concerned, the Commission submits that it is uncertain whether the EU legislature specifically wanted to target insects. It is even more uncertain whether it was aware of the risks that the consumption of such products could entail. It is likely that the EU legislature decided to regulate only those products of which it anticipated the placing on the market in 1997.

46.      The legislative history of Article 1(2) of Regulation No 258/97 indicates that the Commission initially suggested a rather comprehensive approach. In its initial draft regulation in 1992, the Commission indeed proposed to include products ‘produced from, or consisting of, or containing an organism, or part of an organism, with no established history of food use’. (14) That indeed broad definition disappeared in the Commission’s amended proposal following the European Parliament’s first reading.

47.      Thus, as far as any emerging intent may be established, it is rather that the original sweeping definition was considerably cut back. It was not the intention of the EU legislature to comprehensively include novel foods relating to animals, with the exception of what remained in Article 1(2)(e), in the form of the limited subcategory ‘food ingredients isolated from animals’. That is in a way understandable, since back in 1997, the available animal-based foods seemed to have had a long history of food use at the time. There were no novel foods of that type in Europe as far as whole animals were concerned, with the traditional ones already being covered by other pieces of EU legislation. (15)

48.      Finally, as far as insects as foods were concerned, it appears that the EU legislature had no particular intent in their regard. After all, why should they: eating whole insects or parts thereof was not really on the menu in Europe at that time.

(3)    Interpreting versus justifying legislative choices

49.      Despite the clear wording and structural arguments just outlined, the French and Italian Governments defend the view that whole insects were included within Regulation No 258/97 on the basis of the latter’s general scheme and purpose. The arguments raised by those Governments, in relation to the structure of Regulation No 258/97, are essentially twofold.

50.      First, within Article 1(2)(e) of Regulation No 258/97, there was a clear distinction based on whether foods and food ingredients come from plants or from animals. Foods deriving from plants were comprehensively covered by Regulation No 258/97, with the latter covering ‘food and food ingredients consisting of or isolated from plants’. Thus, Article 1(2)(e) indeed included (i) food consisting of plants, (ii) food isolated from plants, (iii) food ingredients consisting of plants and (iv) food ingredients isolated from plants. (16) By contrast, as far as animal-related food was concerned, Article 1(2)(e) covered only one sub-category, namely food ingredients that were isolated from animals. On that account, the French Government maintains that even if expressions used with regard to animals, on the one hand, and plants, on the other, are indeed different as to their text, they should be read in the same way.

51.      Second, there is the issue of internal logic within the category of animals itself, which boils down to the question of why only ingredients isolated from animals should be covered and not whole animals. In that regard, the French Government maintains that it would be illogical to distinguish between foods and food ingredients, since all of them will eventually be eaten by consumers. It would also be meaningless to apply the rules on novel foods to food ingredients containing parts of insects, but to exclude whole insects, as the Italian Government also contends.

52.      Moreover, the French Government considers that holding that whole insects and their parts were not included within the scope of Regulation No 258/97 would breach the principle of non-discrimination between undertakings marketing foods containing insects and those marketing whole insects for human consumption. Since both are in a comparable situation, both should be subject to the same legislation.

53.      To my mind, the arguments raised by the French Government under the heading of ‘general scheme’ belong to the category of ‘why not also’ arguments, which, at the most basic level, and in the realm of interpretation of a legal text, can simply be answered with ‘because it does not say so’. Such arguments question the choices and categories established by the EU legislature, suggesting that perhaps something else should have been included as well. They already start hinting at what is essentially the main argument advanced by the French Government, fully developed with regard to the aim and purpose of Regulation No 258/97: since the purpose of the measure is to protect public health, and that Government believes that whole insects may also pose problems in terms of public health, they should be included as well, no matter what the text says.

54.      I shall deal with those arguments in the next section, concerning the purpose of Regulation No 258/97 and what it may imply for the interpretation of its notions. However, to my mind and in any case, the structural answer to those questions has already been given under letters (a) and (b) of this section, to the extent that any such answer in fact needs to be given when interpreting a clear provision of secondary law (when the question to be answered normally concerns the meaning of what is in the legislation), as opposed to when faced with a challenge as to its validity (when the legislator may indeed be asked what motivated its choice of (non)inclusion of some other categories, thus having to explain and to justify the underlying legislative logic).

55.      Regulation No 258/97 appears to have aimed at ingredients from animals and not whole animals because the overall intention of the regulation was at micro, not macro, level alterations to the food. Broader capture with regard to (whole) plants, when compared with (whole) animals, seems to be the choice because, at the time of its adoption, although plants had been modified for decades, (17) Europeans had not yet started to change their consumption habits with regard to animals. Thus, there was presumably no need for any such inclusion of whole animals, with the exception of those already covered by other letters of Article 1(2).

C.      The text of Regulation No 258/97 ‘reinterpreted’ in the light of (one of) its purpose(s)?

56.      The key argument advanced by the French Government, bolstered by the references to a number of studies and reports provided by its national agencies, relates to the protective purpose of Regulation No 258/97. It runs as follows. The stated aim of Regulation No 258/97 is the protection of public health. In pursuance of that aim, Regulation No 258/97 would cover food ingredients isolated from animals. Thus, ingredients isolated from insects would equally be covered. If the EU legislature thus acknowledged that parts of insects could present a risk for health and should therefore be covered by the regulation, whole insects should a fortiori also be covered by the regulation, since they present the same, if not even greater, risks.

57.      According to the appellant, the purpose of the previous regulation to protect public health has no impact on the fact that whole insects were not included in its material scope. The wording of that regulation is sufficiently clear. There is thus no need to resort to a purposive interpretation of Regulation No 258/97.

58.      According to the Commission, the exclusion of whole insects was in line with the other objective of Regulation No 258/97, namely to contribute to establishing the internal market. If the EU legislature can rely on Article 114 TFEU to prevent future obstacles to trade resulting from divergences between national laws, the emergence of such obstacles must be likely and the measure at issue must aim at avoiding them.

59.      What the French Government, supported by the Italian Government, is effectively proposing is a teleological expansion of the scope of application of Regulation No 258/97. Something that was previously not included should now be, since it presents the same type of danger.

60.      There indeed exists, in the judicial arsenal, an interpretative technique called teleological reduction: something that, in the normal interpretation of the notions used in the legislation at issue, is at first sight included, will ultimately be judicially excluded because, in view of the purpose and aim of the legislation, it should not have been included in the first place. It simply fell into the basket because legislators normally use broad and open notions. It may thus happen that their application without a teleological reduction will be unduly overbroad.

61.      However, to take the stated purpose of a legislative measure, or indeed as the Commission correctly points out just one of them, and on its basis, in the silence or even against the clear wording of that instrument, start pencilling in new categories not previously foreseen by that legislation, may indeed be called ‘teleological expansion of the material scope’ of a legislative instrument. Apart from that euphemism, however, it is also more commonly known under a different name: legislating.

62.      It would indeed be somewhat hypocritical to now start reproaching the French Government for not respecting that limit in its submissions. It is fair to admit that the record of this Court in respecting that limit of the judicial function in the past has been far from impeccable. In what follows, after having dealt with the argument that Regulation 2015/2283 is a mere clarification of what had always been included in Regulation No 258/97 (1), I shall try to convince the Court, yet again, not to engage in such types of ex post interpretative readjustments, or rather rewriting, of what is otherwise a clear secondary law provision (2).

1.      Regulation 2015/2283: an amendment or a codification?

63.      The French Government suggests interpreting Regulation No 258/97 in the light of Regulation 2015/2283 in order to establish the scope of the former. It considers that the scope of Regulation 2015/2283 was not intended to be broader than that of Regulation No 258/97. The new regulation has only clarified that scope given the scientific and technological evolution since 1997. The fact that whole insects and their parts are now expressly included within the new rules does not mean that it was not already the case within the previous rules.

64.      In defending that line, the French Government relies on recitals 6 and 8 of Regulation No 2015/2283. Recital 6 states, ‘the existing definition of novel food in Regulation (EC) No 258/97 should be clarified and updated’. (18) Under recital 8, ‘the scope of this Regulation should, in principle, remain the same as the scope of Regulation No 258/97. However, on the basis of scientific and technological developments that have occurred since 1997, it is appropriate to review, clarify and update the categories of food which constitute novel foods. Those categories should cover whole insects and their parts …’. (19)

65.      Conversely, those two recitals are interpreted differently by the appellant and the Commission. In particular, according to the appellant, the fact that whole insects now fall within the scope of Regulation No 2015/2283 does not necessarily entail that they also did so under the former rules. The inclusion of whole insects is not a mere clarification, but an addition to the previous definition of novel food.

66.      In view of the clear wording of both provisions, discussed in detail in the previous sections of this Opinion, I find the argument of the French Government entirely untenable. First, it suffices to simply contrast the wordings of both conditions in the respective regulations: ‘food ingredients isolated from animals’, on the one hand, and ‘food consisting of, isolated from or produced from animals or their parts’, on the other. Second, at the level of the quoted recitals, the French Government appears to focus only on the word ‘clarify’, while omitting the equally present and clearly articulated ‘update’ and ‘review’.

67.      Third, at the structural level, the very existence of a provision like Article 35(2) of Regulation 2015/2283 demonstrates that the EU legislature was well aware that the material scope of the regulation is in general much broader than that of its 1997 predecessor. That is also why there was the need to provide for a transitional period to accommodate those products that were lawfully placed on the market at the time of the entry into force of Regulation 2015/2283 but which were not covered by Regulation No 258/97. According to the Commission, Article 35(2) of Regulation 2015/2283 was indeed intended for whole insects in order to ensure that the latter could still freely circulate, albeit for a limited period of time, after the entry into force of Regulation 2015/2283.

68.      Thus, the scope of definition concerning foods from animals contained in Article 3(2)(a)(v) of Regulation 2015/2283, when compared with Article 1(2)(e) of Regulation No 258/97, is very clearly an amendment, considerably expanding the scope of that definition.

2.      Dynamic (re)interpretation of the scope of Regulation No 258/97 via its purpose (in a changed social context)?

69.      Finally, there remains the argument of the need for a ‘judicial update’ of what is said to be an outdated piece of EU legislation. Even if one were to agree that there was no gap back then since insects were not really à la carte in 1997, there apparently was a gap in 2016, when the préfet de police de Paris (Prefect of Police, Paris) ordered Entoma to withdraw them from the market. Should it therefore not be possible to resort to a ‘dynamic interpretation’ of Article 1(2)(e) of Regulation No 258/97 in order to conclude that whole animals fell within the scope of the latter regulation as a consequence of the subsequent evolution relating to eating habits and the emergence of new risks attached thereto?

70.      Indeed, the interpretation of (indeterminate) legal notions should never be static. It must react to the societal evolution, both technical and social. (20) Moral categories evolve over time. (21) The same goes for technical definitions, such as that of ‘food’ for instance. The interpretation of such notions cannot be frozen in time.

71.      That being said, there are limits to such dynamism in law if being carried out by a court. Three general ones, which are applicable transversally, might be mentioned, as well as an additional one, which is of particular relevance in highly technical fields.

72.      First and above all, the text itself is the limit. The proposed dynamic interpretation must be compatible with the natural meaning of the words, indeed interpreted in their evolving context. Thus, in the unlikely event that a text from 1850 were to refer to ‘liability for vehicles’, interpreted today, such a liability would include a (motor) car, as well as an electric car. If the text were to refer to ‘liability for coaches’, squeezing a (motor) car into that notion might still be possible under certain conditions. However, if the text were to read ‘liability for small two-wheeled carriages for public hire propelled by one horse’, then subsuming a (motor) car under that notion would simply no longer be possible.

73.      It therefore matters what may plausibly be accommodated under a given text, within its reasonably conceivable semantic vagueness, which ought to form the natural limit to any judicial expanse. In the present case, whole animals to be consumed as such simply cannot be subsumed under ‘food ingredients isolated from animals’. It is established case-law of the Court, at least with regard to the limits of conform interpretation applicable to national courts, (22) that interpretation cannot be contra legem. (23)

74.      Second, there is the imperative of legal certainty and foreseeability of the law, in particular for individuals, who must be able to predict at least to some degree what the applicable legal regime is and to adapt their behaviour accordingly. (24) Unexpected and thus unforeseeable departures from the natural meaning of words makes navigating any legal system equal to walking on quicksand. The lack of legal stability in interpretation fosters rule cynicism and disregard for the law: why bother caring about the law if something is likely to mean something else from one day to the next?

75.      The same limits are a fortiori  applicable when the EU measure at issue imposes obligations or sanctions. (25) While the present case does not concern sanctions, it pertains to setting obligations for market operators, which, on any normal construction of the previously applicable rules, they would not have had.

76.      Third, the argument concerning the separation of powers within the Union at the horizontal level, more frequently referred to as the institutional balance, might again, in view of its actual application practice, perhaps not be the strongest one. However, within the legal context of the Union, the same limits to interpretation have also vertical or diagonal implications: enlarging interpretatively the scope of application of an EU measure normally has as its consequence an effect on the distribution of powers between the Union and the Member States on that given matter.

77.      On that aspect, the present case is intriguing. The traditional roles appear to be somewhat reversed. The French and Italian Governments are not in fact seeking to reclaim regulatory territory that on the proper construction of the scope of an EU measure should have pertained to the Member States. They aim to achieve the opposite.

78.      However, if the natural interpretation of the scope of application of Regulation No 258/97 proposed in the previous sections of this Opinion were to be maintained, that would mean that, before the entry into force of Regulation 2015/2283, the Member States would have always been free to regulate, should they wish to, the placing on their market of whole insects. That matter was simply not covered by Regulation No 258/97. As rightly pointed out by the Commission, the non-inclusion of whole animals meant in practice that Member States retained the competence to adopt rules regarding the placing on the market of animal-related foods which did not fall within the scope of the regulation.

79.      It is in this context that the overall argument of the French Government sits somewhat uneasy. It does not appear that France would have adopted any such national rules, be it under its retained competence on that matter, or, if in doubt, under Article 12(1) of Regulation No 258/97. Furthermore, there is no mention anywhere in the case file that that Member State, if it believed that whole insects were indeed covered by Regulation No 258/97, in spite of its wording, would have asked for the matter to be determined as a matter of interpretation of the scope of that regulation pursuant to the mechanisms expressly foreseen therein (Article 1(3) in conjunction with Article 13 of Regulation No 258/97).

80.      That is not to say that, on merits, the substantive arguments advanced by the French Government as to the dangers associated with the human consumption of whole insects might not be correct. Rather, it is simply to point out that, if one wished to transform those concerns into binding rules by which market operators must abide, other procedural avenues would have been more appropriate than seeking to ex post facto expand the scope of an EU measure to matters to which it clearly was not applicable.

81.      Fourth and final, there is the argument urging judicial caution in specific areas of law, in particular those concerning highly technical matters, in which courts possess little expertise. Within such areas, the third general argument, concerning separation of powers and the ensuing democratic legitimacy,(26) reaches additional dimensions in terms of knowledge and expertise.

82.      In the legislative procedures, as a part of the political and deliberative process, both the public as well as expert opinions will be heard and their views hopefully reflected. By contrast, courts, especially those not collecting any expert opinions or not hearing expert witnesses, are simply ill-equipped for decisions on such technical matters, in particular those where there is little or no scientific knowledge or consensus. (27) Their role in such areas should thus best remain a minimalist one, focusing essentially on two elements: the verification of the existence of flexibility, safeguards, and avenues for ongoing adaptation and precaution embedded within the instrument at issue, i.e. the procedural dimension of managing risk and uncertainty, on the one hand, with only limited substantive intervention in exceptional cases of legislative irresponsiveness to radically changed social and technical circumstances, on the other hand. (28)

83.      However, in both cases, if any such legislation were found to be wanting, the more appropriate judicial reaction in such a context is to annul the contested measure or the severable parts thereof, thus forcing the (EU) legislature to reflect anew. It will only rarely be a good idea for a court, including this Court, to start ‘interpretatively’ pencilling in new categories which require advanced technical or scientific assessment and knowledge of the subject matter.

84.      In conclusion, without those elements being in fact explicitly raised, since such issues would indeed pertain to the validity of a measure, it suffices to mention that, on the one hand, as outlined in the previous points of this section, Regulation No 258/97 contained safeguard and review clauses and procedures, which were apparently not made use of. On the other hand, the EU legislature has indeed been responsive to social and scientific change, as far as novel foods consisting of animals are concerned, since the adoption of a new regulation, Regulation 2015/2283, rather clearly demonstrates. The latter regulation is nonetheless not to be given de facto retroactive application by questionable ‘judicial interpretation’ of its predecessor.

V.      Conclusion

85.      I suggest that the Court answer the question referred to it by the Conseil d’État (Council of State, France) as follows:

–        Whole animals to be consumed as such, including whole insects, did not fall within the scope of Article 1(2)(e) of Regulation (EC) No 258/97 of the European Parliament and of the Council of 27 January 1997 concerning novel foods and novel food ingredients.


1      Original language: English.


2      Regulation of the European Parliament and of the Council of 27 January 1997 concerning novel foods and novel food ingredients (OJ 1997 L 43, p. 1).


3      As adopted in 1997 and published in OJ 1997 L 43, p. 1. However, with successive amendments made to the regulation, letters (a) and (b) of paragraph 2 have been left out. Their content was effectively taken over by other secondary law measures.


4      My emphasis.


5      See, for example, judgments of 9 November 2016, Davitas (C‑448/14, EU:C:2016:839, paragraph 26), and of 26 October 2017, The English Bridge Union (C‑90/16, EU:C:2017:814, paragraph 18 and the case-law cited).


6      Article 2(2)(f) of Regulation (EU) No 1169/2011 of the European Parliament and of the Council of 25 October 2011 on the provision of food information to consumers (OJ 2011 L 304, p. 18). Note, however, that Article 2(1) of Regulation No 258/97 expressly excluded from its scope flavourings, food additives and extraction solvents.


7      In, for example, French, ‘ingrédients alimentaires’; in German, ‘Lebensmittelzutaten’; in Italian, ‘ingredienti alimentari’; in Spanish, ‘ingredientes alimentarios’; in Polish, ‘składniki żywności’; in Czech, ‘složky potravin’; in Dutch, ‘voedselingrediënten’.


8      In, for example, French, ‘isolés à partir d‘animaux’; in German, ‘aus Tieren isolierte’; in Italian, ‘isolati a partire da animali’; in Spanish,‘obtenidos a partir de animales’; in Polish, ‘pochodzące od zwierząt’; in Czech, ‘izolované z živočichů’; in Dutch, ‘uit dieren zijn geïsoleerd’.


9      The notions now used in Regulation 2015/2283 (see above, points 12 and 23).


10      While taking due note of the factual clarification submitted by the French Government that all the insects sold on the French market for human consumption would have, at the stage of marketing, been treated in one way or another for their conservation and transport. They would therefore no longer be technically speaking whole when sold, like oysters or eggs, because they would be deprived at least of water. True as that argument technically is, it still appears to me that a dried whole cricket remains a cricket, even if sold as gourmet chips.


11      Leaving aside, again, the issue of whether in 1997, frogs’ legs had been used for human consumption to a significant degree within the Community. The consumption was certainly not negligible in France. As for the whole of the Union, it might be another story.


12      While acknowledging that there are not many examples which immediately jump to one’s mind under this heading, that is if one disregards the (post-Harry Potter, no doubt flourishing) market of magic potions.


13      Reproduced above at point 5 of this Opinion.


14      Annex I, second hyphen, of the draft regulation (COM/92/295FINAL — SYN 426  (OJ 1992 C 190, p. 3)).


15      ‘Traditional’ foods or food ingredients consisting of or isolated from animals parts would themselves be normally governed, among other pieces of EU legislation, by Regulation (EC) No 178/2002 of the European Parliament and of the Council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety (OJ 2002 L 31, p. 1).


16      Naturally again under the condition that they were not ‘foods and food ingredients obtained by traditional propagating or breeding practices and having a history of safe food use’.


17      As also explicitly acknowledged by the cross-reference made in recital 4 of Regulation No 258/97 to novel plant varieties and varieties of agricultural plant species.


18      My emphasis.


19      My emphasis.


20      See my Opinion in Confédération paysanne and Others (C‑528/16, EU:C:2018:20, point 100 et seq.).


21      See recently, for example, for the notion of ‘public morality’, judgment of 27 February 2020, Constantin Film Produktion v EUIPO (C‑240/18 P, EU:C:2020:118, paragraph 39).


22      See, for instance, judgments of 15 April 2008, Impact (C‑268/06, EU:C:2008:223, paragraph 100), or of 15 January 2014, Association de médiation sociale (C‑176/12, EU:C:2014:2, paragraph 39).


23      But see, for example, judgment of 27 October 2016, Commission v Germany (C‑220/15, EU:C:2016:815, paragraphs 33 to 48) in contrast to my Opinion in Commission v Germany (C‑220/15, EU:C:2016:534, points 23 to 50).


24      See, in this sense, for example, judgments of 18 February 1982, Zuckerfabrik Franken (77/81, EU:C:1982:70, paragraph 23); of 19 June 2012, Chartered Institute of Patent Attorneys (C‑307/10, EU:C:2012:361, paragraph 60); and of 2 March 2017, Glencore Céréales France (C‑584/15, EU:C:2017:160, paragraph 55). See also, in the specific context of Article 1(2) of Regulation No 258/97, Opinion of Advocate General Szpunar in Davitas (C‑448/14, EU:C:2016:39, point 32).


25      See, for example, judgments of 29 March 2011, ThyssenKrupp Nirosta v Commission (C‑352/09 P, EU:C:2011:191, paragraphs 80 and 81); of 5 December 2017, M.A.S. and M.B. (C‑42/17, EU:C:2017:936, paragraphs 51 to 57) and of 20 March 2018, Menci (C‑524/15, EU:C:2018:197, paragraphs 46 and 49).


26      See also, in this regard, the recent Opinion of Advocate General Hogan in Austria v Commission (C‑594/18 P, EU:C:2020:352, point 42).


27      For the realist and wise acknowledgement of the limits of judicial review in such matters, see the order of the Bundesverfassungsgericht (Federal Constitutional Court, Germany) of 23 October 2018 in case no 1 BvR 2523/13 (ECLI:DE:BVerfG:2018:rs20181023.1bvr252313).


28      In detail, my Opinion in Confédération paysanne and Others (C‑528/16, EU:C:2018:20, point 139 to 141), or my Opinion in Lidl (C‑134/15, EU:C:2016:169, point 90).