Frame approach to legal terminology: What may be gained from seeing terminology as manifestation of legal knowledge?
Introduction
Meaning as knowledge held by users11.Thank you to Cristina Farroni, Università di Macerata and an anonymous reviewer for valuable comments on a previous version.Meaning in language has been conceptualized in many different ways in the course of its linguistic description and investigation, including, for instance the Principle of Compositionality (i.e., “the meaning of an expression is a function of, and only of, the meanings of its parts together with the method by which those parts are combined”, Pelletier 1994) and similar approaches, depending upon logical operators as the engine of meaning generation. Such approaches share the characteristic that they intend to conceptualize (word) semantics as a system that works independently of textual and contextual factors. A different approach to meaning which is gaining ground today is that of the broad approach to the study of language termed ‘cognitive linguistics’ (Croft and Cruse 2004). This approach relies on the idea that it is not relevant to distinguish sharply between syntactical structure and lexical meaning. Instead, all linguistic regularities, syntactical as well as lexical, are seen as contributing to the emergence of meaning. So, instead of using logical operators to describe lexical semantics independently from syntax (like in truth-conditional semantics), all linguistic regularities are basically seen as semantic or at least semanticized in the way that basic conceptual models like Figure-Ground or Path are used as basic categories (e.g. Lakoff 1987; Langacker 1986; Talmy 2000). The meaning of a word and its syntactical behaviour are then both explained through the contribution of semantic and syntactical characteristics to the cognitive emergence of the conceptual structure (Faber and L’Homme 2014, 144).
Concentrating again on the meaning of words in concrete texts which is central to the task of describing legal terminology, word meaning may be conceptualized as the result of applying knowledge that is learnt through previous experiences and instruction and is stored in long-term memory in an ascription process. Hence, the meaning of words emerging in a specific context is dependent upon the knowledge held by the people involved in the communicative process and their conceptualizations of the context in which the words are used.
The frame approach central to this chapter is an example of such cognitive linguistic approaches. Taking a frame approach to terminology means that the (generalized) meaning of a term is seen as the knowledge that is (generally) evoked by the term when used in a communicative context. Within this conceptualization, the study of meaning zooms in upon the stored knowledge typically referred to and seen as relevant in specific communicative settings, so-called ‘understanding-relevant knowledge’ (verstehensrelevantes Wissen, Busse 1997, 15), U-relevant knowledge after Ziem (2014, 132). Hence, the cognitive approach means a shift in the basic approach from a focus on a context-independent meaning towards a context-dependent understanding with the human conceptualizer at the centre of attention (Kerremans, Temmerman, and de Baer 2008, 178–179). Such an approach allows for the explanation of the inherent vagueness of legal concepts by way of models of prototype semantics (e.g. Heylen and Steurs 2014). A further and more general consequence of this shift is that it makes it easier to explain consistently why the same word may carry different meanings in different fields of specialization, due to differences in interest and perspective (Kerremans, Temmerman, and de Baer 2008, 180). This challenge to understanding each other across disciplines is often described in studies of project communication involving experts of different fields (e.g. Holste 2019; Janich and Zakharova 2014).
The traditional approach to terminology in theory as well as in the construction of terminological resources has been leaning towards the context-free approach (Faber and Araúz 2016, 3). Taking a frame approach to the study of legal terminology instead has at least two central consequences:
-
The process of understanding terminology in actual communicative settings is seen not as mere decoding, but as an inference process, i.e., a process in which participants enrich underspecified linguistic input based on their stored conceptual knowledge and their experience with the specific type of communicative settings.
-
The models of the meaning of terminological units should be structured in accordance with what is known about the basic and general structural characteristics of storing knowledge in long-term memory.
In this chapter, the second consequence will be at the centre of attention, as it is used as an argument for choosing the frame as the structural model for meaning.
Accounts of terms and their meaning generally emphasize the idea of terms as reflections of expert knowledge. Terms are special in the way that they are used to reflect aspects of conceptual knowledge where boundaries play a role: using a term means indicating a pointed difference between a concept and one or more neighbouring concepts (ten Hacken 2015). According to ten Hacken (2015), this way of defining terms is relevant in order to differentiate terminology from specialized vocabulary: Specialized vocabulary lacks the clear differentiation intent and is thus closer to the more prototype-based meaning of everyday communication. In this chapter, however, we will concentrate upon terminology. Terms are used when the context requires a sharp instead of a blurred distinction, for instance when, in cases of legal adjudication, it is necessary to find out whether an act by a person should count as bribery or not. In traditional approaches to (legal) terminology, this often leads to a focus on one-dimensional so-called hyper-hyponymic relations, i.e., a relation between a more general and some more specified concepts along optimally one, but always some few dimensions (Rackevičienė, Janulevičienė, and Mockiene 2019, 3). Work of this kind tends to give special importance to relations between concepts in logically structured systems, visualized in the form of tree diagrams. As an example, Rackevičienė, Janulevičienė, and Mockiene (2019) look for the positions of concepts on bribery in the conceptual systems underlying an international convention and three national legislations in order to find functional equivalents for translations of texts belonging to the different systems, concentrating upon the logical relations between the concepts. In their analysis, functional equivalence is thus given when two concepts from different legal systems show similar logical relations and distinctions (private vs. public, active vs. passive) to neighbouring concepts. The position in the system and the differences between concepts are the main meaning-carrying characteristics. In a frame approach, on the other hand, although the hyper-hyponymic relations may also play a role, they are not the only decisive criterion for distinguishing and ordering terms in conceptual systems. In the sense of L’Homme (2005, 1122), terms are taken seriously in their full lexico-semantic nature and not reduced to labels of knowledge units in a system. Instead of seeing the hierarchical system as the main source for meaning, frame semantics approaches rely more on the idea of multidimensional networks inspired by the way knowledge is actually structured in the human mind when modelling meaning. Hence, frame approaches are open to different and potentially unrelated dimensions constituting the concept related to a term. Put somewhat pointedly: frame approaches intend to be closer to human ways of thinking in associations along networks, whereas traditional logically oriented approaches to terminology intend to be as logically stringent as possible. Behind this also lies a difference between seeing domain-specific knowledge as person-external common knowledge to be drawn upon by experts in communication, in the traditional approach, and seeing domain-specific knowledge as carried by individual members of the relevant community in the form of shared knowledge. This distinction will be elaborated upon in the following sections.
The present chapter consists of an overview of different strands of frame-based research into legal meaning as knowledge, followed by a presentation of the central aspects of this kind of approach, emphasizing the general characteristics that these approaches to legal terminology share.
Frame approaches
General aspects of frame approaches in connection with terms
The central characteristic of frame approaches is the attempt to coordinate the description of (lexical) meaning with what we know about the mind and its structures. In other words, the idea is to blur the boundaries between the traditional concept of lexical meaning and more encompassing knowledge structures actually held by language users: “Frame Semantics … is based on the assumption that the meanings of lexical units … are constructed in relation to background knowledge, whose structure can be analysed in terms of semantic frames” (L’Homme, Robichaud, and Subirats 2014, 1365). Through blurring this boundary, connections between more knowledge-driven approaches (concentrating upon the structure of the knowledge of a specialist field) and more lexicon-driven approaches to the description of terms and their meaning are possible (L’Homme 2018, 4). In this way, bridges may be built between what is traditionally termed lexical meaning, on the one hand, and the description of background knowledge relevant for textual understanding be it linguistic or extra-linguistic (U-relevant knowledge, cf. above), on the other. The idea is to avoid the concentration upon one perspective and thus to strike a balance between a focus on providing answers to lexicon-driven questions as well as to knowledge-driven questions (L’Homme 2018, 6). Lexicon-driven questions depart from the utterance perspective (e.g. what does the word ‘equity’ mean in an English legal context, what words collocate with it?), whereas knowledge-driven questions depart from the need for insight into the discipline (e.g. what are the characteristics for applying the legal figure ‘equity’ in English law in adjudication?).
That the frame approach is looking for a balance between different approaches is for instance visible in the fact that the lexical perspective is not privileged in the meaning description. Searches may depart from words (what does the word mean?) or from disciplinary concepts (what words are connected to the same frame?). Hence, analysers may still be looking for the meaning related to a word. However, they do not presuppose that this meaning is stored in isolation in the knowledge reservoir of individuals. Instead, it may be generated by collecting the knowledge units evoked by a lexical unit, and different lexical units may evoke (elements of) the same frame.
Conceptually coordinating meaning with insights into the structuring of the mind implies that a multi-dimensional structure is relevant, as knowledge is organized in associative networks in memory based on spreading activation (McClelland and Rumelhart 1985). In such networks, word meanings are combinations of knowledge from different dimensions with relevance for the use of the word in a specific context. Seen from the point of view of the content of the frame, the frame is also a unit of organization that binds together lexical units connected to each other by experience or by their belonging to the same part of for instance specialized legal knowledge or expert action (Engberg 2009b; Pimentel 2015, 431). Since Bartlett (1932), such structures presumed to be the building blocks of memory have been called schemas. A frame may be seen as a type of schema (Kann and Inderelst 2018, 60), containing knowledge elements describing situations evoked by lexical units in an abstract format (L’Homme 2018, 8). Furthermore, frames may be joined into scenarios, thus modelling larger parts of the actual knowledge of disciplinary specialists (L’Homme 2018, 9). Following the concept of knowledge as organized in schemas, the general frame approach to structuring knowledge is to work with a combination of slots and fillers (Barsalou 1992; Busse 2012; Fillmore 1982; Ziem 2014). Slots (called ‘Frame Elements’ in Frame Semantics) are dimensions, according to which the meaning of a term is organized, i.e., the building blocks of the structure of the knowledge which the word evokes. These slots are typically related to the predicate structure(s) to which the investigated terminological units contribute when used in texts (actors, objects, instruments …). In this way, the slot structure does not only tell us something about the core declarative knowledge traditionally contained in terminological definitions. It also includes elements with relevance for the combination of words in texts in order to depict special domain situations. In accordance with the tenets of cognitive linguistics, i.e., what was above called semanticizing linguistic relations, this is the way in which a frame approach to terminology blurs the boundaries between knowledge-driven and lexicon-driven interests in terms. Fillers, on their part, are the standard knowledge chunks used to characterize the term from the point of view of a specific dimension. Hence, the fillers represent the full U-relevant knowledge (not only the declarative knowledge) typically connected to a term by the members of the legal community, whereas the slots represent the standard way of structuring this knowledge. For instance, the frame related to the concept of criminal liability of corporations in US law can be modelled using slots like characteristic elements, relevance for society, related concepts, and specific theories, to name some (Engberg 2009b, 132). Fillers would then be the content typically connected to the slots, like deterrence as an aspect of the slot relevance for society that is often mentioned when the concept of criminal liability appears in disciplinary communication (Engberg 2009b, 134).
Frame approaches to (legal) terminology and terminography
In connection with legal terminology and knowledge, frame approaches have been used especially for two types of purposes: as a basis for constructing legal terminological resources and as a way of representing the knowledge behind legal texts for more discourse analytical purposes. The two approaches are not radically different, but for ease of explanation they are distinguished here.
Frame approaches for legal terminological resources
In the terms introduced in the previous section, the traditional approach to constructing legal terminological resources has been more knowledge driven than lexicon driven, although at least translators would actually benefit from a greater focus on information about the more linguistic aspects of this knowledge (Pimentel 2015, 430). Consequently, there has been a drive to develop the theoretical basis of legal terminology towards being more lexically driven. An important step was to abandon the idea of terms as mere pointers to knowledge and instead see them as linguistic units with all derived characteristics (Cabré 1999). This has led to three high-profiled approaches, which are presented in what follows.
In the first approach, the so-called lexical semantic approach to terminology (L’Homme 2019), the idea is to create terminological resources based on Fillmore’s FrameNet approach (Fillmore, Johnson, and Petruck 2003). The aim is to gather and present information in a systematic way with a view to creating resources that are optimally lexicon oriented. In the field of legal communication, Janine Pimentel has been the central researcher working with this approach (Pimentel 2012, 2013, 2015). Terms are basically seen as lexical elements that are used in specialized settings and with restrictions on their interpretation. Hence, the representation of their meaning may depart from descriptions in FrameNet, which intend to give frame-structured representations of lexical meaning in general language, i.e., representing the non-specialized meaning. Lexical units to be included in the resource are described by attributing to them a role in existing frames that are then, if necessary, specified in accordance with the specificity of the legal discipline (Pimentel 2015, 432–433). The basis of the attribution is studies of relevant corpora, hence allowing the terminologist to not only concentrate on nouns, but also to give relevant attention to the verbs. Exactly the frame as an instrument of representation with its aim of representing elements involved in actions allows verbs to acquire their relevant role in understanding and formulating legal texts.
The second high-profiled approach to using cognitive linguistic insights to creating terminological resources is Frame-Based Terminology (Faber 2012, 2015). Specific applications to the field of law are found in Peruzzo (2014) and Faber and Reimerink (2019). Here the idea of predications as the basis not only of textual utterances, but also of structuring representations of knowledge is found again. The approach is a specification and further development of the general approach to the linguistic description of language called Lexical Grammar Model (Faber and Mairal 1999). The frame type ‘event templates’ is central, similar to the approach by L’Homme (Faber and Reimerink 2019, 16; Peruzzo 2014, 157). This choice of frame type supports a focus in the knowledge representations on verbs, which are seen as the central terms in the legal field with its emphasis upon actions in the real world (Faber and Reimerink 2019, 20). In this approach, the ambition to present detailed descriptions in accordance with general-language theories plays a prominent role. Hence, the approach works with a distinction between the internal structuring of concepts in accordance with a general system (Pustejovsky’s qualia roles that represent the semantic structure of the lexical meaning in order to understand the emergence of compositional default meaning when words are used in sentences and texts, Pustejovsky 1995), and an external structuring according to general types of relations that represent the relations of a concept to other concepts. Apart from bringing Frame-Based Terminology closer to general linguistics beyond the field of lexical semantics, it also reveals an ambition to make representations of language-specific legal terms as little language-specific as possible. This is a difference not in kind, but in degree when comparing the approach to the lexical semantic approach to terminology, but it appears to play a major role in Frame-Based Terminology, probably in order to contribute not only to research in terminology, but also in general language. Furthermore, the orientation towards a broader linguistic theory is also connected to the ambition of creating knowledge bases that may be used not only by humans, but also by computers.
Finally, a third approach is relevant here, although it does not work directly with the concept of frame, namely, the so-called ‘Sociocognitive Terminology’ linked to the method of ‘Termontography’ as a way of representing its result (Kerremans, Temmerman, and De Baer 2008; Temmerman 2000, 2007). Like in the approaches by L’Homme and Faber, the basic motivation is to construct efficient and relevantly structured knowledge representations in the form of resources for understanding and for translating specialized texts. The approach should be mentioned here, because it is also based upon a cognitive view to meaning and meaning generation. In this case, the basis is especially Vygotsky (1986) and his ideas of meaning and meaning acquisition as centrally influenced by the context of the interpreter. The traditional idea of objective terminological meaning is substituted by the concept of unit of understanding. The substitution allows to represent meaning as dependent upon different disciplinary and linguistic contexts, i.e., the meaning of the same term may contain different knowledge elements depending on whether it is used, e.g., in a legal or an accounting context. Overlaps are possible, but the representation can model different understandings, and the conceptualization easily allows to orient knowledge resources towards the needs of different disciplinary understandings. Similarly, resources may be profiled towards differing levels of presupposed knowledge on the side of the user. Instead of working with frames as the basic model for representing knowledge and meaning, the approach works with the idea of prototype and of units of understanding composing larger networks of knowledge structured in a kind of language-oriented ontologies. Hence, ontologies as suggested in this approach play the role of basic structuring principle in the multilingual description of terms played by the frame in the other two approaches mentioned here, thus enabling comparisons across languages.
Frame approaches representing knowledge underlying legal communication
The second group of approaches concentrates upon representing knowledge underlying legal texts in order to understand the individual concepts or make relevant comparisons of such concepts, and less upon creating tools in the form of knowledge bases or terminological databases.
One line of research in this strand is interested in applying a frame semantics approach to model the actual knowledge structures underlying expert communication in the legal field. In this case, too, the basic assumption is that of frames as models of knowledge stored in long-term memory as suggested by Bartlett (1932). However, the attention is upon the knowledge relevant for communicative understanding (U-relevant knowledge, cf. above) and upon frame semantics as a tool for legal conceptual work. Busse, Felden, and Wulf (2018), as a very thorough example, present three different major studies in which they demonstrate the power of frame semantics as a method for assessing the shared disciplinary knowledge constituting legal concepts. One study focuses upon the special content and structure of a German statutory concept from criminal law in connection with its realization in legal practice (Diebstahl, i.e., the German version of the concept of theft). One study looks at finding the differences in knowledge between two related German legal concepts from private law (Eigentum and Besitz, two concepts from the field of property law). The last study uses frame semantics as a tool for demonstrating developments in the interpretation of a German criminal law concept (Gewalt, i.e., force). In Zarco-Tejada and Lazari (2017), a similar frame semantic approach is used for the purposes of comparative law, studying the national variants of the concept of state responsibility in a Spanish, French, English and Italian context. This line of research thus concentrates upon meeting the needs of legal experts for conceptual work when understanding national legal concepts or when securing comparability of concepts in comparative work. In both cases, the approach offers a broader conceptual description as in traditional legal work, which tends to concentrate upon functional aspects (cf. the functionalist approach to comparative law, Zweigert and Kötz 1996, 35)
Finally, as the second line of research in this strand a version of frame semantics interested more in describing selected text-based elements in their relation to world knowledge presented by Konerding (1993) has been used for more discourse analytic studies of legal concepts. The original method was developed to construct meaning descriptions in dictionaries. The purpose of the method is to function as a descriptive instrument for qualitative textual content analysis inspired by frame semantics. However, unlike the first approach presented in this subsection, there is no ambition of assessing the full knowledge relevant for communicative understanding and the actual knowledge structures in their entirety (cf. Busse, Felden, and Wulf 2018, 35–36 for a critical view on this purpose in connection with the concept of frame). On this basis, the so-called Knowledge Collection and Categorization Method (Engberg 2009b, 129) has been developed for studying aspects of legal concepts as represented in a text or in different texts in a corpus. Hence, there are overlaps in interest with Textual Terminology (cf. Condamines and Picton 2022). However, where Textual Terminology aims at building large resources for purposes like Natural Language Processing, the Knowledge Collection and Categorization Method aims at analysing texts for discourse analytical purposes. The method has for instance been used for assessing differences in the way different authors emphasize elements of a concept in their writing, thus giving access to diverging positions to the content of a concept that may help us understand conceptual development (Engberg 2009b). Furthermore, the method has been used as a tool for systematic conceptual comparisons relevant for decision-making in legal translation (Engberg 2018, 2020). These last-mentioned types of analyses will be presented in more detail below.
How does a frame-oriented conceptualization of legal terms influence the description of terms?
The previous presentations of the frame idea, the reasons for its adoption in the description of legal terminology, and some of the central instantiations of it as a tool for analytical and knowledge representation purposes function as a backdrop for an overview of the consequences of taking a frame approach to legal terminology in this section. I will concentrate upon how a frame approach enables researchers and practitioners to grasp the dynamic of legal knowledge due to two related characteristics of knowledge and work in the discipline of law: the fact that legal knowledge is distributed among participants in specialized communication (unintentional dynamics), and the fact that the content of legal concepts is interpersonally challenged among legal experts at any point in time (open synchronic variation). It is my claim that frame approaches to legal terminology are conceptually well-suited for grasping these aspects.
Dynamic nature of legal knowledge due to distributed character
Semantic change over time is a characteristic of all language use, dependent especially on the fact that language as such (and hence also lexical meaning) is distributed among the language users and does not refer to any ‘objective’ instance outside the context of the language users apart from accepted convention. However, due to the ideal of at least Western thinking that a legal system should be characterized by the rule of law, traditionally phrased as a government of laws, not of men, legal thinking has tended to ‘forget’ or tone down this aspect, because it works against the ideal. Instead, the ideal of literal interpretation of legal texts has been championed, although it is very difficult to theoretically explain what the literal meaning of words in a text actually is.
Expert communication, to which legal communication involving lawyers belongs, is even more prone to dynamics than everyday communication (see e.g. Temmerman and Van Campenhoudt 2014 for studies of this aspect for different disciplines). Kerremans, Temmerman, and De Baer (2008, 180–181) have isolated the following factors as drivers of constant development of the shared disciplinary knowledge underlying meaning in context:
-
Urge for more and better understanding of the world as part of any discipline;
-
Meaning as created in processes involving different language users with potentially different knowledge bases;
-
General prototype structure of the knowledge underlying meaning-making in concrete situations.
Two factors are of special importance in this context: the distributed character of meaning making, on the one hand, and the prototype structure of human knowledge, which means that mental categories normally do not have sharp edges concerning instances belonging to the categories, on the other hand: individuals must not be aware of the fact that they do not fully agree on categorization, as long as they agree upon a sufficient part of the core of the category. This allows for differences arising ungoverned and in a ‘hidden’ fashion (see e.g. Heylen and Steurs 2014).
By looking at knowledge through a frame semantic lens, and especially by focusing on the distributed character of knowledge in frame semantic analyses, it is possible to understand the actual dynamics in a consistent theoretical context. Furthermore, law is a culturally constituted sign-system, which is “arbitrary as it depends on space and time” (Wagner, Matulewska, and Cheng 2020, 240). Hence, the knowledge system of individuals is in a constant process of development due to influence from developments outside and inside the society (Wagner, Matulewska, and Cheng 2020, 242–244). In traditional approaches to terminology, this aspect has mainly been studied under the heading of ‘variation’. An example is the analysis by Peruzzo (2017) on the influence of national legal systems on the development of terms on the victims of crimes. Attention in this analysis (although not in general in Peruzzo’s work, see e.g. Peruzzo 2014) is on assessing the variation of terms about basically the same concepts and on categorizing the observed phenomena. In other words, formulations and regularities for their variation are at the centre of attention. Frame approaches, on the other hand, concentrate more upon developments in the content with a special interest in what parts of a concept (= which slots) are subject to developments, and how these changes may be explained as developments in the content of the frame. As indicated above, one of the studies presented in Busse, Felden, and Wulf (2018) is an example of such a frame-based investigation of conceptual development, looking at how the statutory concept of Gewalt has changed in interpretations by courts in the German legal system due to changes in conceptual positions over time. Kerremans, Andries, and Temmerman (2016) investigate the use of denominations for types of mothers in statutory proposals in Dutch and French in the Belgian Parliament, with a focus on neologisms, i.e., words used for concepts that have not yet been included in statutory law. By comparing proposals in Dutch and French, it is possible to assess differences in what aspects and distinctions are highlighted in proposals in the various languages. On this basis, the authors find differences that signal differences in the knowledge used in understanding concepts of the field in the various language communities, differences that are prone to generate dynamics in terminological meaning.
Another driver of unintentional semantic dynamics in legal meaning is the inherent vagueness of legal concepts (Engberg and Heller 2008; Wagner, Matulewska, and Cheng 2020, 251; 257–260). Again, the distributed nature of meaning, which a frame semantic approach is apt to grasp theoretically, is a central characteristic. A vague concept is one where borderline cases exist in which it is not possible to decide whether communicative instances belong to the category indicated by a concept or not. An example is the concept of reasonable doubt from private law. What is found here is less dynamics at system level and rather differences in the interpretation by readers in accordance with their conceptualization of the context in which the concept is used. This type of dynamics is especially visible in translations. Heylen and Steurs (2014) have developed a methodology consisting of an automated term extraction process for source text analysis combined with a corpus linguistic approach to the distribution of lexical units in a target text functioning as equivalents of the terms isolated in the source text. By assessing the range of different renderings in the target language of the same source language term it is possible to find the level of generality and vagueness of the investigated terms. From the point of view of traditional prescriptive terminology, these characteristics are systematic flaws. However, a frame approach with its acceptance of dynamics from the process of meaning making based on meaning potentials rather than as the process of decoding a fully delineated conceptual meaning allows us to work with a way of understanding the role of terms and the disciplinary concepts, they are connected with in accordance with the actual rather than idealistic characteristics of expert legal communication.
Non-monolithic character of legal knowledge, subject to contemporary challenge
Supplementary to the characteristic that (even) legal meaning is dynamic over time due to the fact that knowledge behind processes of meaning making is distributed among individuals, as discussed above, conceptual dynamics over time is also caused by the fact that concepts as building blocks of expert legal knowledge are typically openly non-monolithic at any one time. This characteristic is related to the distributed character of knowledge described above. However, where the development due to distribution is more of a coincidence, as language users are normally not aware of the differences between their knowledge bases and thus drive the development unconsciously, the non-monolithic characteristic hints at open and explicit differences that are even discussed in expert communication. Hence, where the dynamics due to distribution is a consequence of accepting terminology as human language (Cabré 1999), the non-monolithic characteristic relies upon the fact that legal experts constantly argue over the correct meaning of legal concepts, i.e., the first of the three drivers of Kerremans, Temmerman, and De Baer (2008) mentioned in the previous section. Thus, open synchronous differences in meaning between different legal experts is frequent. Again, this is different from the conceptualization of traditional terminology oriented towards standardization, in which the basic assumption is that the discipline agrees upon definitions of its concepts and that the task of terminological work is then to clarify the concepts via defining explicitly and assigning one and only one designation, a term, to this definition. In other words, traditional standardization-oriented terminology is interested in the agreed knowledge of the discipline, as in Simonnæs (2013), for instance, who assesses the translations offered by students for the culture-bound Norwegian concept of med-moder (co-mother). However, as already mentioned above, disciplinary knowledge is characterized pragmatically by a constant push towards achieving new knowledge and hence develop concepts. In law this is especially the case because the rationale behind legal concepts is to solve problems in society stemming from differing interests among groups of citizens (Engberg 2016). For instance, rules on speeding balance the interests of car drivers to go as quickly as possible against the interests of other car drivers, bike riders and pedestrians of being safe in traffic, to mention a simple example. Already this aspect of compromise between interests carries in it the germ for the open co-existence of different conceptualizations, as it is highly likely that the holders of the different interests will consciously try to push the compromise over time, for instance intending to move the speed limits up or down. Furthermore, the fact that the meaning of legal concepts depends upon culturally constituted interpretation schemas that change with societal trends (Wagner, Matulewska, and Cheng 2020, 240) means that one of the expert activities of lawyers is to venture novel ways of interpreting basic legal concepts in the light of developments in the norms and values of a society. On these grounds, it is more likely in the field of law than in many other disciplines that competing versions of a concept will exist at the same time, which may share more or less the slot structure, but differ in the accepted fillers (cf. examples below). This is what I mean by stating that legal concepts are non-monolithic.
This state of affairs is difficult to grasp in traditional terminological approaches because they tend to focus on the common rather than on the individual. Frame-based approaches, on the other hand, have an advantage here, because they have a space for modelling the knowledge of humans rather than a decontextualized form of meaning. Hence, it is no theoretical problem in a frame approach to accommodate the (differing) knowledge of individuals (as the knowledge to be included in the model is seen as always distributed among individuals). Furthermore, it is unproblematic to accommodate the fact that it is even part of expert knowledge to know that differing and competing conceptualizations exist and to know how I as an individual position myself towards the different variants. For humans are able to know also what is in our view not correct but held by others to be true.
As examples, these characteristics have been investigated and modelled empirically in frame-based approaches in my own work. Engberg (2010) studies the simultaneous existence of different approaches to the conceptualization of murder in Swiss criminal law22.Schweizer Strafgesetzbuch Article 112. at different points in the development of the concept between two statutory definitions (1942 vs. 1990) as a factor in driving the diachronic dynamics. At any of the investigated points in time, different players in the development process prefer one of two opposing concepts (murderers to be convicted due to reproachable personal characteristics vs. murderers to be convicted due to their reproachable actions). This is a case of the same slot (the criterion for being classified as having committed a ‘murder’) being filled differently by different participants in the communication. What changes over time is the value given to the different variants by the (shifting) majority: both positions are present all the time, but what is the filler not accepted by the majority at the beginning of the development turns to being the preferred filler in the end without fully eliminating the previously preferred variant.
The influence of balancing of interests on the concept is also the object of study in Engberg (2009b). The studied concept is ‘corporate criminal liability’ from the context of US law, which is mentioned also above. The concept balances the interest of corporations to freely perform their activities in a non-personalized way, on the one hand, and the interest of society in being able to punish wrongdoings as acts of the (non-personalized) corporation, despite the system characteristic that only natural persons and not corporations may be subject of criminal punishment. Based on a study of a corpus of journal articles, aspects of the frame of the concept are established and it is assessed in the framework of matrix frames as descriptive framework to what degree the investigated journal articles agree upon the filling of different slots. Thus, the non-monolithic character of the shared knowledge is demonstrated through an investigation of expert communication.
As indicated above, a frame approach can also easily accommodate the fact that even the concepts held by individuals are non-monolithic. In Engberg (2009a) this is demonstrated empirically by showing that individual articles represent the differing fillers shown in Engberg (2009b), discuss them and position themselves towards them. The descriptive advantage of the frame approach in this context is that it is not limited by logical rules, but only by the affordances of the human cognitive system.
How may a frame approach be practically useful for legal translation?
In the introductory section I mentioned that an important driver behind the development of frame-oriented approaches to terminology like Frame-Based Terminology or the lexical semantic approach to terminology has been to build more efficient resources for translators to use when making decisions in the translation process. An example of a resource derived from Frame-Based Terminology is Eco-Lexicon33. http://ecolexicon.ugr.es/en/index.htm whereas an example of a resource building on the lexical semantic approach is DicoInfo.44. http://olst.ling.umontreal.ca/?p=2550 The main idea in such resources is to present in a structured way knowledge that on the one hand takes the linguistic nature of terms seriously, and on the other hand incorporates non-linguistic modelling of the world reflected in textual expressions based on the basic tenets of cognitive linguistics. An advantage of following a frame approach in this context is that it enables the investigator to compare perspectives of different languages to comparable entities in the form of non-linguistic event models or event frames not bound to the individual language or to a small number of logically oriented structuring criteria.
In the field of law, a central source for building up relevant frames are the results from comparative law studies. Among the presently applied approaches to comparative law, conceptual comparisons are more useful for translators than functionalist comparisons (Brand 2007; Engberg 2013). In brief, the basis of a functionalist approach to comparative law is to use as tertium comparationis the functions of legal rules concerning the way they solve the societal problems that gave rise to their construction. Societal problems are often quite similar across jurisdictions, objects of study are then similarities and differences in the way the legal rules work. The conceptual approach to comparative law, on the other hand, uses similarities in the broader conceptual structure as tertium comparationis. Frames are well-suited as a presentational basis for such results and especially for making it easy for translators to find the information relevant for their purpose. An example of such a conceptual comparison relying upon a frame approach is the work by Zarco-Tejada and Lazari (2017) on the concept of state responsibility as represented in legal texts in Spanish, French, English and Italian mentioned above.
For legal translators, approaches like the ones developed by Zarco-Tejada and Lazari (2017) are interesting and relevant. However, as such approaches are developed to meet the requirements of legal experts, they may not be optimal for the purposes of translators. Where comparative lawyers are often centrally interested in similarities and differences between normative aspects of legal concepts, the task of translators is to solve problems of communication and text formulation across language borders. This means that translators are typically at least also interested in other, more culturally and linguistically oriented frame elements or slots. In previous work (Engberg 2016, 2020), I have proposed methodological approaches to the analysis of legal communication which may generate results especially relevant for legal translators and deliver input for representing the knowledge in a frame format. The frame format requires input for the slots as well as for the fillers. In order to get an overview over the slot structure, the idea is to depart from insights into the characteristics of law and legal communication. I suggest three perspectives from which legal texts may be studied, in order to generate translation-relevant input to frames modelling the knowledge of legal experts:
-
National culture (generating slots based on national characteristics of legal concepts)
-
Socio-functional systems (generating slots based on characteristics connected to the discourse-based system of legal experts, which is not bound by national borders)
-
Interpersonal communication (generating slots connected to linguistic interactions involving legal experts and their expertise).
Where the two first perspectives have to do with declarative legal knowledge, the third perspective is oriented towards linguistic aspects as well as aspects of different views on conceptual characteristics held by different individuals, i.e., the aspects described above on the distributed and especially the non-monolithic character of legal concepts.
In order to generate knowledge about the fillers for the slots, on the other hand, different general approaches to law are used as generators. One example of a suggested dimension is the normative approach to law, concentrating upon law as a normative system, looking at the provisions and qualifications indicating scope and prerequisites for the provisions; another example of a suggested dimension is the Law as Culture approach (Gephart 2006, 292–300), where aspects like law as organization or law as an act-oriented ritual play a role and widen the scope of content to be used to fill the slots generated by the three first-mentioned perspectives. The inclusion of the Law as Culture approach is a consequence of the broader interests of translators mentioned above, due to their specific use of comparative legal knowledge.
Taking a frame approach to legal terminology and its relation to legal knowledge supports a constructivist approach to the process of legal translation. As the frame used to represent a legal term is seen as a model of the legal knowledge which legal experts dispose of in their long-term memory, the frame may be seen as the basis on which legal experts construct mental representations based on textual input when receiving a text. This opens an avenue for evaluating suggested target text renderings by assessing the ways in which they intend to include elements of the frame in target representations through text formulation. I will end my deliberations on frames and legal terminology by presenting an approach that may help comparing suggested translations with the source text and thus finding arguments for assessing the quality of the translations in the target text situation suggested in Engberg (2018, 44–46). Here, different translations of a source text term (Rüge, English: Objection) embedded in a sentence are evaluated based on their relation to the underlying expert frame. Such an objection is part of the event frame of an appeal case and has the function to initiate the case. The point of departure is thus the legal perception of a real-world event. The relevant parts of the event frame constructed through legal sources are included in Figure 1:
The German source text shown in Example (1) draws upon this frame by introducing the typical fillers of the ‘Types’ slot (underlined):
Gegen das Urteil richtete sich die auf eine Verfahrens- und Sachrüge gestützte Revision des Angeklagten
[Against the court decision was directed the appeal of the defendant based on a procedural and material law objection (interlinear translation)]
The strategy for guiding the mental representation of the textual meaning is thus to use the terms and presuppose relevant knowledge on the part of the receivers.
Target texts are evaluated according to the textual clues they have chosen in order to include slots in the construction of mental representations on the side of the target text receiver. Underlying the evaluation is the idea that the emerging mental representations built by target text receivers should have a relevant overlap with the mental representations that source text receivers may be expected to construct. Different target texts drawing on different slots of the same source culture frame may all lead to the construction of a relevant target representation. The relevance of different target representations is evaluated in accordance with the Skopos of the translation. Examples (2) and (3) are different renderings of the source text in Example (1). The renderings of the terms from the source text are underlined:
Tiltaltes revisionsanke, som var støttet på en processuel og en materielretlig indsigelse, rettede sig mod denne dom
[Defendant’s revising appeal, which was based upon a procedural and a material law objection, was directed towards this court decision (interlinear translation)]
Den tiltaltes anke af dommen støtter sig på påstand om overtrædelse af den materielle ret samt påstand om procedurefejl
[The defendant’s appeal of the court decision is based on claim of infringement of material law and claim of procedural error (interlinear translation)]
Example (2) uses a strategy where the ‘Types’ slot is included introducing the conventional fillers in the form of the terms for the two types of objections. Example (3), on the other hand, applies a different strategy by including a part of the ‘Internal causal chain’ slot (the claims that carry the objection) and distinguishing them along the dimensions of the ‘Types’ slot, but without introducing the conventional fillers by the terms. The first strategy will be effective, if target text readers already know the source expert frame, or if the target expert frame is sufficiently similar and known to the readers. In that case, a relevant target representation may be built based on the target readers’ long-term memory. The second strategy is also effective, if target readers have a relevant frame in their long-term memory, although it does not use the conventional filler from the ‘Types’ slot. However, it would be more effective than the first strategy if target text readers do not already have the expert frame in their long-term memory. For the chosen textual clues reveal more about content of the frame by indicating an aspect that carries the objection and by spelling out the distinction underlying the ‘Types’ slot. Hence, it includes two slots in the representation, helping non-experts to build a frame and not just receive the names of types which they cannot accommodate with their existing knowledge. Whether the first or the second strategy is preferable thus depends upon the projected readers of the translation, the degree of similarity of the legal concept across source and target culture, and the skopos of the translation.
By way of conclusion, I would like to highlight two take-aways from my deliberations in this chapter. Firstly, I hope to have shown that frame approaches to legal terminology (to which I would, as indicated above, also reckon the Term-Ontology approach) have an advantage over traditional terminology approaches in that they intend to be closer to the associative structuring of human long-term memory as the basis for making mental representations of texts. This means that they are able to easily grasp not only normative aspects of legal knowledge and relate them to legal terminology, but also include cultural and linguistic aspects with relevance for the task of the legal translators and legal communicators. Secondly, I hope to have shown by the last example how the results of frame approaches to legal terminology enable legal translators and other legal communicators to choose consciously between different possible renderings of a message. This effect depends especially upon the broadness of the knowledge that may be represented by frame approaches and upon the links between frame approaches and constructivist approaches to textual understanding.
My deliberations have been concentrated on aspects related to descriptive terminology work. Whether the affordances of frame approaches to legal terminology are also relevant for improving prescriptive terminology work, will have to be the topic of future research.