Linking Signifié with Signifiant: The Court of Justice of European Union as a Product-Defining Authority

Saussurean concept of signifie-signifiant agnation can be used to explain the EU law product-nomenclature referrals. The CJEU has an especially important role in developing detailed rules of product nomenclature interpretation in cases where ambiguities emerge. In its jurisprudence, the CJ pursues preservation of the predominant intuitive model of that interpretation even in cases involving composite products. Only in cases where the composition can easily be identified, the Court relies on the concept of the defining element—i.e. it takes the dominant element of the product as a product nomenclature determining one. With respect to novel products, the CJ applies the same general penchant. Therefore, with respect to such products, the Court attempts to establish their signifié-signifiant match by referring to analogous features and characteristics of already existing products. Such an approach is an evidence of the CJ’s self-constraint of its otherwise Demiurgic power concerning the product nomenclature. This argumentation implies that even the CN classification offers a catalogue of relatively rigid designators and that their rigidity is respected by the CJ. The CN designators are not perfectly rigid because the CJ takes into account not only the logical values of respective CN positions, but also the practical aspects (including transactional costs) of the CN classification, as well as the general objectives of EU rules underlying the CN position identification.


Introduction
Any process of communication of legal rules can be interpreted as the one which is based on semiosis. Semiosis is an interactive process in which meanings conveyed by intentionally used signs are decoded to produce their understanding. In the so called binary concept of semiosis formulated by Ferdinand Saussure, there are two opposing factors involved in this process: signifiant i signifié. They denote, respectively, that what signifies (in an utterance or text) and what is signified, i.e. represented as a concrete designate of a name used to denote it [7]. Such an approach corroborates with the Wittgensteinian concept of language ability of referral in which the only relevant communication relationship is that between utterances and intentions of persons who produce them to refer to something the utterances are meant to mean. According to Ludwig Wittgenstein [9]: Der Satz ist ein Bild der Wirklichkeit. Der Satz ist ein Modell der Wirklichkeit, so wie wir sie uns denken. 1 There is a significant body of evidence that language, including legal language, is not apt to perfectly grasp that relationship between sentences and intersubjectively conceived reality. This implies that it neither is completely fit to produce a proper referral to highly abstract or strongly culturally-contextualized constructs, not to physical objects of reality, such as e.g. products. This imperfection is caused by idiomatic character of any natural language arising mostly from its open texture (Germ. Porosität der Begriffe) which makes denoting a gradable and therefore logically not strict operation [10]. The gradability of signifié-related reference features potentially resulting in a change in its signifiant can be referred to as ''soriticality''.
In a multi-lingual regulatory setting (such as the EU one), where legal provisions are meant to produce the same meaning in different languages, the said inability to achieve a complete, strict and logically equivalent referral in any natural language concerned can also be attributed to the differentiation of languages with regard to their openness. Openness is defined by Kazimierz Ajdukiewicz [1] exactly as language's ability to produce direct and strict semantic equivalents of words or phrases present in another languages. What we know is that official languages of the EU differ with this respect [4].
Also well-known Sapir-Whorf hypothesis defy the possibility to achieve the said equivalence. In its relative version, it holds that language of utterance has an impact on the mode in which thoughts are formed. If it is true, it will apply also to law. Most likely, the Sapir-Whorf hypothesis is indicative of the different intellectual paths employed by users of different languages in semiosis, i.e. to decoding legal norms from texts in which these norms are expressed. Its obvious consequence is that with increasing number of different language cross-referrals, the probability of different outcomes of semiosis will increase. This problem is only aggravated by different pattern of compositionality legal texts in different languages have.
Compositionality can be understood to be determination of meaning (itself being a product of contextualized thought) of complex syntagmas (i.e. sets of intentionally located words meant to produce some meaning) by their structure and the meaning of respective parts such syntagmas are composed of [3].
Moreover, evidence exists that the natural language, especially the legal one, can be significantly elliptic in the sense, that when interpreted in a natural (i.e. non-legal context, or the context which is non-specific for a given branch of law) language, the text can have hidden meanings. Elliptic nature of a text (regardless of whether legal or non-legal) in a given language implies that it is not able to be a perfect vehicle of thought (or-in other words-reproducers of meaning)-themselves never being vague/elliptic [2]. Fodor's concept is relevant for the delineation of theoretical framework of analysis of EU law concerning the nomenclature of goods. As this law is multilingual, it raises the question of procedural and material equivalence of its nomenclature rules. The former is concerned with the mode employed in respective languages to obtain a legal norm (i.e. the binding, unambiguous, indication of nomenclature assignment to respective objects it is meant to denote) from EU legal provisions. The material aspect of reference concerns whether, irrespective of the procedural aspect, the nomenclature rules guarantee the same outcome in each and every EU official language. It is conceivable that with plain language interpretation of legal texts reproduced in two or more different languages, those of them which strive to rely on the same interpretation pattern may produce different outcomes, whereas it is also possible to produce the same outcome with the application of differentiated procedural rules determined by the very texts. Such a presupposition is legitimate whenever these texts are meant to be interpreted in different cultural contexts, with domestic legal systems being inherent components of these contexts.
In the EU, both the procedural and material aspect of interpretation has special importance because legal texts are reproduced in 24 official languages. These reproductions are considered equally authentic. As such, they must be relied upon (also for the purposes of cross-referencing) to produce meaning which, against the odds, is intended by the EU legislators to be uniform.
From this perspective, it is interesting to verify whether the CN classification offers a catalogue of rigid designators or not. According to Kripke [5], rigid designators are signifiants (i.e. ''designators'' in Kripke's terms) which designate the same item (product) with respect to every possible context in which this item exists, whereas non rigid designators do not. This means that with respect to rigid designators, universal petrification signifié-signifiant relationship is achieved by virtue of some legitimized rules. Positive verification that CN classification indeed is a catalogue of rigid designators would create a strong argument for retroactive effect of both the EU Court of Justice's judgments in this area and the European Commission's tariff classification regulations [6] issued on the basis of the Commission Regulation (EEC) No. 2658/87 of 23 July 1987 on the tariff and statistical nomenclature and on the Common Customs Tariff. 2 If they are indeed rigid, they will have to be treated as objective; therefore, their interpretation and, consequently, understanding, should be immune from any contingent, contextual influence. Non-rigid notions do not have such an immunity.

Defining in the Context of Combined Nomenclature
Combined nomenclature (CN) represents a specially viable legal method of referring. 3 It creates structural semantic hierarchy of referents based on the Harmonised Description and Coding System (HDCS, most often referred to as the Harmonised System, HS). The Harmonised System is intended to be a complete catalogue of names of products being subject to international trade. The HS is being developed under the auspices of the World Customs Organisation (WCO). European Union is member of WCO on the basis of the Council Decision 87/369/EEC of 7 April 1987 concerning the conclusion of the International Convention on the Harmonized Commodity Description and Coding System and of the Protocol of Amendment thereto. 4 The CN itself represents a system emulating signifié-signifiant relationships worked out within the HS. This relationship can be identified on the basis of a common language or plain text, i.e. signification experience people have with regard to naming things which appear around them. As a result, the CN systemic interpretation is based on ordinary meaning, i.e. the narrow-context meaning based on semantic meaning, as contrasted to widely-contextualised communicative meaning (Slocum [8], pp. 270-281). Thus, the CN rules are construed by referring signifiants to names deemed semantically proper for the purposes of this nomenclature. Once this relationship is established, it is meant to make it possible for the CN users to reverse the path of the logical order adopted by the CN standardsetters, i.e. it is meant to reconstruct it having signifié at the beginning, i.e. the object somehow named according to the CN rules. By specifying the reversible relationship between signifiant and signifié in the process of CN attribution, the users normally employ their own, largely intuitive yet based on their understanding of general product-differentiation rules, knowledge of goods. This implies that these persons should be aware of all the important features of the good subject to the CN and of the most essential (or rudimentary) characteristics of their manufacturing.
As a matter of routine, the CN rules concerning the CN coding attribution to any product refer to the qualification of that product's basic material or substance of which it is made. That is to say, that the process of attribution should start with an analysis of evident features of a product, yet it should also reflect on the process of its manufacturing. If, as a result of such an analysis, it appears that two or more CN items of nomenclature can apply to the product, the most specific item (i.e. the item which provide the narrowest and most detailed semantic reference to the product considered) should be its CN code. Thus, the CN classification is based on lexical priority rules rather than purely material ones.
In some cases, however, the identification based on the reference to basic material or substance is not possible. This impossibility arises from the fact that some products appear to be fitting to more than one CN category or no CN category altogether, e.g. because they are made of a mixture of two or more component parts or materials, and some products can be new. In such cases CN classification rules require that the substance or material which gives the product considered its essential character should be taken as a decisive factor for the semantic name selection from the CN catalogue of names. If a product was manufactured in a form of a functionally arranged set of machinery, then the decisive for the assignment should be that component piece of machinery which performs the fundamental function of the whole set.
Regardless of their apparent interpretation-guiding effectiveness, these rules are far from being clear and thus effective with regard to cases which fall short of being typical. Thus, there is a significant jurisprudence of the EU Courts of Justice with regard to the assignment problem.
One of the earliest, fundamental for the assignment problem, ECJ's judgments was that rendered in the case 38/76 Industriemetall LUMA GmbH v. Hauptzollamt Duisburg. 5 In this judgment, the Court outlined the principal procedure applicable to the assignment problem. The Court, unsurprisingly, indicated that the CN-based classification analysis should start with the analysis of essential characteristics of a product to attributing. According to the CJ, this procedure should commence with the physical examination of the product's composition, including unintended additions. After that, the procedure should involve semantic interpretation of the nomenclature applicable to the product considered with due regard to the product's intended use (i.e. its intended functions). The final stage of this procedure is the application of a EU customs tariff or any other law relevant rules consequential to the CN product classification. If necessary, in the classification procedure recourse should be made to the European Commission's CN-classification decisions, its clearing notes or to the classification opinions formulated by the Committee of Nomenclature-the two latter not being binding legal acts but still considered ''extremely important factors in the interpretation of the Common Customs Tariff''. 6 The CJ noted that such opinions were important in forming classification criteria applicable to products which are alike, yet a material enough difference exists between them. The Court, therefore, implied that a comparison would always be involved in the process of classification between the ''typical'' product construed with intuitive or informed reference to the CN logic (i.e. the reference pattern) and the product being subject to classification which had to be compared to this pattern. Whenever the pattern and the product compared with it do not match to each other, new analysis is needed to presumably produce a new, different, classification.
In its judgment in LUMA the Court emphasized that assignment of a given nomenclature position to a good could result from a simultaneous application (as ''material conditions of classification'') of many different criteria which, at least apparently, may have equivalent analytical status. It is, however, of utmost importance for the assessor to identify these features which are decisive from the point of view of deciding upon tariff position-in particular, this should involve looking for the closest CN equivalent of a product considered if there is any direct link between any CN position and that product. According to the CJ, this operation should involve identification and assessment of the essential criteria of classification.
In LUMA, the CJ also focused on the identification of a substance of the product at stake: the iron alloy ingots, the stage of their processing, and the final form in which it appears in the market. Yet, after having indulged itself with such an analysis, the Court admitted that it was apt to produce a technical definition of a product. Yet, somewhat astonishingly, the CJ admitted that such a definition might, in some cases, be different from the definition needed for CN classification purposes. These purposes rather require that the product manifests ''objective characteristics which constitute the necessary and sufficient condition'' for the classification of a product under a given nomenclature position.
The CJ adopted a similar approach to the CN classification in its judgement in the case 36/71 Günter Henck v. Hauptzollamt Emden. 7 The Court again upheld here that certainty of law and ease of verification of a classification outcome required relying on objective characteristics and objective features of products considered-as they are specified in specific positions of the Community Customs Tariff. In the judgment 40/88 Paul F. Weber (in liquidation) v. Milchwerke Paderborn-Rimbeck e. G 8 the Court re-confirmed-as a general rule-that: …in the interests of legal certainty and ease of verification, the decisive criterion for the classification of goods for customs purposes is in general to be sought in their characteristics and objective properties as defined in the wording of the relevant heading of the Common Customs Tariff and of the notes to the sections or chapters.
Additionally, the Court recalled its logic presented in LUMA, in which it required that essential characteristics criterion should take priority over any other criteria even in the situation, where the CCT invoked manufacturing process as a factor which had to be taken into account. 9 If at all considered, relevant manufacturing process should as a secondary source of CN-classification argumentation. Thus, the CJ upheld that: …whilst the Customs Tariff does indeed in certain cases contain references to manufacturing processes, it is generally preferred to employ criteria for classification based on the objective characteristics and properties of products which can be ascertained when customs clearance is obtained.
The Court of Justice's jurisprudence indicates that, in some more complex cases, the nomenclature can refer to manufacturing process of products concerned and to the intended use of these products. In its judgment in the case 120/75 Walter J. Riemer v. Hauptzollamt Lübeck-West 10 the Court upheld that the manufacturing process had to be taken into account as a classification factor whenever it results in an irreversible transformation of a product to such a degree that it does must be reflected in a shift within the applicable CN classification. In this particular case, such a shift concerned fresh blackberries versus deep-frozen blackberries. The Court adopted a similar approach in its judgment in case C-423/09 Staatssecretaris van Financiën v. X BV, 11 in which it upheld that manufacturing processes could result in a significant change in properties and objective characteristics of products, whichwhenever it is accounted for in the context of the CN-should result in a change in the CN assignment applicable to these products. The Court, however, emphasized that the mere method of preserving or storing the product should not be considered a criterion resulting in such a change unless, the CN classification rules legitimizes this circumstance to be a basis for change in the CN positions. 12 It the CN nomenclature-concerned cases, discrepancies in respective official language versions of this nomenclature are usually not so acute as in other cases. Even if they emerge, they do not cause significant ramification for legal interpretation. Above all, it should be noted that the CN signifiants are expressed in two languages which are perfect equivalents for each other: the natural and the artificial (numerical) one. Of course, the latter is universal-yet it is dependent on the referring value of the natural language. The numeric system can, to a significant extent, be relied with only limited recourse to the natural language system as it reflects its own systemic logic (e.g. the logic of increasing semantic specificity downwards the CN system of nomenclature, which normally reflects on the increasing level of transformation of a product and/or its narrowed down scope of intended use. This is not to say that referrals to the CN natural language system is not needed, but rather to emphasize that the two logic systems can be relied together to sort out language-related discrepancies via systemic methods of interpretation. As such, these systems create a mutually reinforcing reference framework gauging the CN interpretation in the matters of doubt. The lesser importance of this problem for the proper implementation of EU law (if compared to many other areas of EU law) arises most significantly from the fact that comparative analysis of respective language versions (performed under the so called CILFIT formula 13 ) is much less effective in this realm that the analysis considering HS own criteria invoking objective general characteristics of products 10  considered. 14 In fact, relying on the HS criteria is in line with more general rules applicable to EU law interpretation in cases where semantic discrepancies exist between different language versions of the same legal act. These rules require that, in the case of a discrepancy: (a) the interpretation rely not only on one language version of a EU act, (b) no language version be considered controlling (solely authentic), (c) the interpretation invoke the objective and the general scheme of rules of which a problematic legal provision is a part.
The system of CN interpretation should be considered unique in the entire system of EU interpretation, as-unlike in other areas-it can refer to some underlying external rules (i.e. the HS), it has its unique semantic coding mode (i.e. where the language coding system is interwoven with the numerical one), and it has developed its own, highly specialized, methodology of interpretation referring to the most pronounced characteristics of products considered, their dominant substance, their intended use, or the level of their transformation.
In practice coding-decoding operations which are performed in the CN system context may be complex because-as it has already been said-some products subject to the CN nomenclature appear to be ''borderline'' (or ''blurred'') cases for classification. Some newly developed products may not easily fall within the existing CN positions, either. Yet, only in extreme cases the CN system is not able to offer enough flexibility to allow all products, including the said ''borderline'' items or new products be classified under some of its existing headings. In such cases, the bottom line classification criterion is ''objective features and characteristics''. This criterion can be perceived as the one which at least is apt to address the soritical nature of some relevant features and characteristics of products subject to the CN classification. Soriticality means that these features and characteristics are gradual, which means that only in some products they can be exposed perfectly. The CN assessment thus involves a judgment as to what extent it is enough to identify them in order to decide that the product falls within one CN heading versus another. or pedicure or appear in a liquid form on the market (the requirement which should be met in order for the product to be classified under CN 3304 30 00 heading). 16 With regard to composite products, i.e. the ones consisting of different product or materials, or being mixtures of different substances, the CJ requires the application of a nomenclature heading which denotes or refers to the material or component substance decisive for the essential character of the product considered. For example, with respect to gloves for horse-riding made of a textile and having an artificial padding, the CJ pointed out that proper CN classification of such products required a general assessment of their appearance and all their features which grant them their essential character/nature-especially by taking into account ''the relative importance of the criteria used for their classification'' (fr. ''l'importance relative des critères utilisés pour leur classement''). 17 This could be reinterpreted to mean that in matching signifiant with signifié, the CN-classifying requires a reversal of coding operation used to grant a product it's CN heading: the most ''denominating'' product features should be identified. This implies that that all the product's features and characteristics should be prioritized. To help perform this operation, the CJ indicated that it was essential to identify the material, without which product subject to the CN assessment. 18 In other words, the CJ ascertained that the essential material should be identified in the context of functions which grant the product concerned its essential, intended by the producer and perceived as such by the intended user, function. As a result, in Pacific World Limited the CJ judged that the gloves for horse-riding should be classified under the heading CN 39262000 (''articles of apparel and clothing accessories'') falling under a more general heading CN 3926 (''other articles of plastics and articles of other materials…'') and not under the heading CN 611610 [''gloves (…) impregnated, coated or covered with rubber''].
In the case of items composed of two or more products (e.g. machinery/device being configurations of more than one technical device, such as office machines for simultaneous xerocopying and scanning), the CJ pursues the identification of the component which performs the function decisive for the product. This component should then prevail as a signifié-identifying element for CN classification purposes. Thus, in the joint cases C-362/07 and C-362/07 Kip Europe SA et al. and Hewlett Packard International SARL v. Administration des douanes-Direction Générale des douanes et droits indirects 19 the Court indicated that adequate CN classification of composite products required prior identification of functions performed by the respective component devices and relationships emerging among them. Such an identification should be used to assess whether these respective functions are at the same level of significance or not. If they are indeed not at the same level of significance, the product performing the prevailing (i.e. the most significant, defining) function should be the one which grants the composite product its essential character and, therefore, the relevant CN heading. Whenever there is no function which prevails in a composite product, the product should be classified under the most general CN heading which applies.
The resulting model of classification relies significantly on cross-referencing among various CN headings and agnation which represents a subsequent choice between competitive, cross-referenced CN signifiants. This kind of approach is even present in the CN-related regulations adopted by the European Commission on the basis of the Council Regulation (EEC) No 2658/87 of 23 July 1987 on the tariff and statistical nomenclature and on the Common Customs Tariff. 20 An example of the European Commission's cross-referencing and negative cross-referencing (the latter indicating product features which exclude some alternative classification decisions) is provided in the Commission Implementing Regulation (EU) No. 2017/182 of 27 January 2017 concerning the classification of certain goods in the Combined Nomenclature. 21 This legal act contains explanatory Annex concerning the thumb grips for a game console controller, i.e. special small rubber pads which make game console operation more comfortable. The thumb grips to which the Regulation (and thus the classification argumentation provided for in it) refers are shown in Fig. 1.
Consistently to what has already been said about the general approach worked out in the CJ's jurisprudence, the explanatory note focuses on the identification of the prevailing function of the product; it also makes positive as well as negative cross-references to functions considered less exposed. Verbatim, the note reads as in Table 1. Table 1 presents an explanation why the product presented in Fig. 1 should be classified under CN code 3926 90 97, i.e. under the CN heading ''other articles of plastics''). Both the description of the goods and the reasons for this classification are meant to explain the European Commission's stance. The former section of the Table 1 offers a general description of the product considered (i.e. its general presentation and some essential reference to the substance it is made of as well as details of its design), technical information about its intended regular use, and its intended function defined in a way which makes it possible to define consumer's value added.
This information is somewhat corroborated by the Commission's explanation of the reasons for its classification opinion. In this section, soriticality problem is defined and then solved. The solution is formulated on the basis of the guidance given by the CJ in the case C-152/10 Unomedical A/S v. Skatteministeriet, 22 in which the Court upheld that a product whose functions are somewhat subordinated to the functions of another product (main product) had to be also assessed in terms of its relevance for the enhancement of the essential functions of that main product. As a result of the relevance assessment, the product concerned should be classified either as a ''part'' (whenever the product as a whole is essential for the operation of the main product) or ''accessory'' (whenever the item assessed is ''an interchangeable part designed to adapt a machine for a particular operation, or to increase its range of operations, or to perform a particular service relative to the main function of the machine'' 23 ) or as any less relevant item (whenever the enhancement is of a more general character. In its assessment, the European Commission followed that reasoning. Since thumb grips for game console controller performed a general enhancement function, the product was classified under a very generic CN 3926 90 97 heading ''other articles of plastics''). Having considered these basic differences arising from the diversity of language reproductions of the same EU legal acts, the Court concluded that no language version could be considered decisive with regard to the exact meaning of the relevant terms and, consequently, to the delineation between the meaning of the phrase ''medicinal product'' and ''classified substances''. The Court went further to remind of its well established line of reasoning according to which the difference should be interpreted with reference to the context in which respective legal provisions function and the objective of the regulation they are the part of. The Court concluded that-in the context of these interpretative principles-the key phrase for the regulation considered was the lexeme ''medicinal products''. 28 The said phrase has a legal definition provided for in the Directive 2001/83. The primary objective of this legal act is to stringently regulate production, distribution and use of medicinal products. According to this provision, any marketing of medicinal products can be commenced only after a consent of a competent market supervision state authority has been granted. On the other hand, the most pronounced objective of the Regulation No. 283/2004 is the prevention and prosecution of any improper use of classified substances. This objective is to be achieved also by a system of controls of production and distribution as well as of relevant criminal sanctions. It transpired for the Court that the two regulations set forth two different reglamentation systems, having two different objectives yet relating to the same functions with regard to the products considered. The Court also noticed the fact that there It is quite interesting to note that the CJ reconstructed the objective and the narrative context of the taxonomic rules. It is striking that it applied the originalistic approach to its reconstruction. This means that instead of construing the present logical status of key legal terms considered, the Court invoked their original sense, i.e. the taxonomical rules which had been applied by the court in the moment, in which an allegedly illegal trade was conducted-i.e. the moment in which the national court's doubts emerged with regard to the interpretation of the relevant EU law provisions.
An interesting aspect of EU law interpretation in which the CJ invoked semantic formulas related to the pragmatic context in which this law applies was presented in the CJ's judgement in the case C-453/13 Newby Foods Ltd v. Food Standards Agency. 33 In this case, the Court adjudicated on the classification of a food product resulting from an application of a novel formula for processing meat offals. Established technology made it possible to obtain well-known in the industry ''mechanically separated meat'' (MSM) out of such offals, whereas the novel formula was meant to produce a different food substance characterized by similar to MSM pattern of degradation of meat tissue, yet by different appearance and texture resembling minced meat. The new features were considered important by the manufacturers because, if the new product were not classified as MSM, it could be allowed in retail trade (in contrast to MSM which-generally speaking-cannot be retailed in its original form) and, therefore, fetch a higher market price. It is essential to note that MSM must be labelled as such whenever it becomes an ingredient of any final meat product. As a result, the national court and the CJ faced an alternative of classifying the new type of a product either as MSM or as a product which could be marketed under any already existing generic name (other than 29 Para. 59 of the judgment. In its decision, the CJ noted that the question of classification of the product obtained by the application of the new production formula in fact concerned the interpretation of relevant provisions of the Appendix to the Regulation No. 853/2004/EC. 34 This regulation set forth hygiene requirements for processing of animal-originated food products. Its underlying idea was that such products, especially when they are degraded in processing, gave rise to an elevated level of health-related microbiological and chemical risks. Therefore, such products should be subject to especially stringent regulation concerning their production, processing, conservation, packaging, presentation, advertising and labelling. In this context, Regulation No. 853/2004/EC offered the definition of MSM. Paragraph 20 of its preamble emphasized that: [t]he definition of mechanically separated meat (MSM) should be a generic one covering all methods of mechanical separation. Rapid technological developments in this area mean that a flexible definition is appropriate. The technical requirements for MSM should differ, however, depending on a risk assessment of the product resulting from different methods.
Annex I to Regulation 853/2004/EC contains (in its points 1.13 and 1.14 respectively) definitions of minced meat and MSM. According to the former one, ''minced meat'' means ''boned meat that has been minced into fragments and contains less than 1% salt''. In contrast, point 1.14 of Annex I defines MSM as ''the product obtained by removing meat from flesh-bearing bones after boning or from poultry carcasses, using mechanical means resulting in the loss or modification of the muscle fibre structure.'' The distinction between minced meat and MSM is also well pronounced in the applicable EU law concerning their marketing.
The CJ also took into account the (then in force) Directive 2000/13/EC. 35 Alike Regulation 863/2004/EC, this directive made fundamental distinction between ''meat'' and ''MSM''. This distinction was made on the basis of the degradation of meat tissue fibre in those products. With regard to meat, the directive required that the natural tissues fibre of animal origin be largely intact-the requirement which automatically excludes MSM from the scope of this definition because MSM does not contain intact animal tissue fibre. The CJ noted that this feature had an important bearing on the perception of average consumers and that this perception had to be considered in the definition of MSM: Mechanically recovered meat differs significantly from ''meat'' as perceived by consumers. It should therefore be excluded from the scope of the definition. In its argumentation concerning the classification issue, the CJ defined the problem it had to solve as a soritical one, i.e. concerning blurred delineation between two terms, which-because of the gradual nature of their signifiés, in the borderline area can potentially be classified to two distinctive name categories. 37 The CJ argued that classifying any product as MSM required that ''three cumulative criteria which (had to) be read in conjunction with one another'' (Fr. trois critères cumulatifs qui doivent être lus en combinaison les uns avec les autres): the use of bare bones from which the intact muscles have already been detached, or of poultry carcasses, to which meat remains attached, the use of methods of mechanical separation to recover that meat, and the loss or modification of the muscle fibre structure of the meat thus recovered by reason of the use of those processes. 38 With regard to the soriticality issue which may arise in the assessment of whether DSM is indeed MSM, the Court noted that the so formed definition did not ''make any distinction as regards the degree of loss or modification of the muscle fibre structure, with the result that any loss or modification of that structure is taken into consideration''. In other words, the CJ required that in the classification, three interrelated factors be considered: (a) the characteristics of the essential substance subject transformation, (b) the transformation of this substance understood to be an intentional process, (c) the result of this transformation.
The essential element of the CJ's analysis in Newby located the main investigation of an appropriate signifié-signifiant match under the Regulation 853/2004 in the broader systemic context, i.e. in the context of other EU legal acts which contain legal definitions of MSM and regulative references to this product. The CJ's manifest intention realized by such an approach was to reduce risk of inconsistency which otherwise might have emerged between all these regulations. It is, therefore, not surprising that, at the end of the day, the CJ relied on its concept of intended transformation leading to a significant change in the substance of a product considered. Since, in this case, the transformation indeed led to a significant modification (degradation) of fiber structure of meat (much more extensive than that resulting from meat mincing) and involved separation of meat remainders from bones or processing of poultry carcasses, the CJ had a good reason to conclude that the resulting substance was indeed MSM. Hence, it could neither be classified as raw meat product, nor might have been marketed under a newly coined name (i.e. DSM).

Conclusions
In the interpretation of the EU law provision referring to product nomenclature, i.e. in the establishing signifié-signifiant relationship, the point of departure is always the reconstruction of the EU legislator's Aristotelian concept of truth understood to be the corroboration between semantic name (and/or numerical CN referent) and its contextualized designate. This rather obvious approach renders it unavoidable to produce interpretative problems whenever such an interpretation involves semantic forms of unstable logic values (as it is the case with respect to highly soritical terms) or whenever new designates have not yet been properly accounted for in the system of referents used in agnation. Interestingly, because of its simplicity and doubleverifiability (possible because of the CN numerical coding applied simultaneously to semantic referrals) CN nomenclature does not give any rise to any interpretation problems arising from compositionality irregularities in different official languages of the EU.
Interpretation of law involves a kind of a non-coincided in time, contextualized, communication between the legislator and any person attempting to competently interpret legal provisions set forth by this legislator. In the area of legally relevant nomenclature (such as the CN), despite the said point of departure (e.g. truth) the EU legislator pursues a logical model in which attribution/agnation criteria invoke intuitive meaning semantics. Interestingly, even in such a case, the legislator's act of attribution/agnation has performative value in the sense that the choices he/she makes are coercive: they stabilize the relevant relationship between rules-relevant meaning and the objects to which they pertain. In other words, they stabilize products' definitions and delineate them in an agnatic format which allows every object ''fit its own slot''. The performative effect of this activity is reinforced by the fact that, in CN application context, semantic values are immediately given their equivalent name expressed in a non-natural language format, i.e. in the numbering system of CN (and, consequentially, tariff codes).
In the cases, where soriticality problem emerges, it is the matter for the courts to eliminate ambiguity in the nomenclature of products (also that falling in the realm of the CN application). The CJEU has an especially important role as it can produce interpretation rules which are binding in a given case, yet which have so strong persuasive value that they are normally followed by other courts and authorities in the EU Member States. The most fundamental premise of the interpretation rules of EU nomenclature classification is that there must be no area of soriticality (and as a result, vagueness) left, even if the mere semantic rules might have allowed it. An epitomic case for this an approach is indeed the CJ's Newby judgment where a new product was decided to fall within the existing classification framework and its features were considered to bear some analogy to an existing product whose name was ultimately extended onto the product considered in this case. This is not to say that the CJ does not enjoy its Demiurgic power over this issue, but rather to emphasize that it attempts to self-constrain itself by doing utmost to exercise the already existing semantic framework and by making recourse to per analogiam rules of interpretation. Moreover, the CJ tries to make its reasoning model as understandable as possible by invoking regular (most likely intuitive) rules which attempt to grasp the ''objective'' features of products/processes to be named, i.e. the rules which are well rooted in common sense.
Yet, sometimes, the ''objectivity'' of the model means ''quick and dirty'' assessment. This means that this model does not rely on any ontological, objectivized concept of truth (and obviously departs from the Aristotelian idea of it), as it declines complete scientific evidence concerning essential features of a product concerned. Instead, it allows intuitive judgement to prevail. Such a line of reasoning was presented by CJ in its judgment in LUMA case where the Court indicated that technically verified (meaning ''based on scientific evidence'') features of a product subject to CN classification can be a specie different from ''objective features'' which its rulings often invoke, as the latter may rely on the most evident characteristics, and in consequence may represent a type of intuitive assessment.
It appears that such an approach can be justified by taking into account transactional costs involved in CN classification and its underlying problem of assessment accessibility. Assessment accessibility is cognitive ability of an assessing individual(s) to understand the features of a product concerned (to specify signifié by a reference to available range of products conceived in CN) and, based on this, of matching the most likely CN signifiant. The approach adopted by the CJ allows dismissal of any such signifié which cannot be specified at sufficiently low transactional costs, i.e. which require disproportional in its scope and, therefore, transactional cost examination of the product features. This intuitive approach based on the idea of proportionality is justified as, in the context of the CN, the legitimized by the CJ signifié-signifiant matching model does not aspire to achieve scientific accuracy, but rather to correct any naturally emerging asymmetry of information between traders in tariff-subjected products and EU tariff administration-all at socially acceptable costs.
The argumentation presented in this article indicates that the CN classification offers a catalogue of relatively rigid designators and that their rigidity is respected by the CJ. The CN designators are not perfectly rigid because the CJ takes into account not only the logical values of respective CN positions, but also the practical aspects (including transactional costs) of the CN classification, as well as the general objectives of EU rules underlying the CN position identification.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.