1.Introduction

Dictionaries are the products of practical and theoretical lexicographic work and come in many varieties. At the same time, some scholars claim that both linguistics and terminology deal, at least partially, with reference tools that are generally called dictionaries. An examination of the relevant literature indicates that the theoretical and practical work with dictionaries in these two disciplines belong to different paradigms. Some scholars argue that lexicography deals with descriptions of general-language words, works with alphabetical organization structures, is descriptive, has laypersons as target groups, and focuses on help to decode texts. This may be a result of lexicographers following a semasiological line based on linguistic theories and principles, which often underlie general-language dictionaries. In contrast, terminologists/terminographers often claim that their lexicographic products deal with terms from specialized domains, work with systematic organization structures, are normative or prescriptive, have specialists as their target groups, and focus on help to encode texts. This may be a result of terminologists/terminographers following an onomasiological line based on concepts, which often underlies special-language dictionaries (Riggs 1989, 89–90; Bergenholtz and Tarp 2010, 28; Kageura 2015, 57). Such positions make you wonder where legal lexicography and legal dictionaries fit in.

At first glance, these divergent views of lexicographic products seem difficult to reconcile. However, the contributions in Mac Aodha (2014), an edited volume on legal lexicography, demonstrate that the views share some features as the contributions prove that legal dictionaries can be used for encoding as well as decoding that lexicographers can use terminological principles in making dictionaries, and that lexicographers can work with printed and digitized dictionaries. In light of the somewhat inconsistent views of lexicography and by implication legal dictionaries, this paper proposes an answer to the following question: What is this thing called legal lexicography? The below attempt to answer this question involves an examination of the ontological position and object of legal lexicography, and a discussion of the interdisciplinary nature of dictionary work. This is followed by a study of user profiling to identify user needs, and a description of methods for selecting and presenting legal terms and other data with particular reference to dictionary functions.

2.Legal lexicography and its object

One way in which to discover the nature of legal lexicography is to begin with a description of the object of legal lexicography. The object is an information tool, the dictionary, which provides a platform for needs-adapted information search. Authors and scholars have called these information tools by different names such as legal dictionaries, legal lexicons, legal encyclopedias, law dictionaries and judicial dictionaries (Greenberg 2014, 59–62), but those are designations of specific types of dictionary. Legal lexicography needs a general definition of its object, called the legal dictionary in this chapter, and the following general definition, based on Nielsen (2018a, 72–73), provides a relevant ontological position. Legal dictionaries have a number of surface features, which consist of their printed and digital user interfaces. These interfaces include prefaces, user guides, wordlists, and appendices, which are components found in many printed and digital dictionaries. In addition, a legal dictionary has three underlying features: (1) it has been designed to fulfil one or more functions; (2) it contains data that have been selected because they help to fulfil its function(s); and (3) it has data organization and presentation structures that marshal its data into the task of fulfilling its function(s). According to the functional theory of lexicography, a dictionary function is a specific type of help a dictionary provides to satisfy specific types of lexicographically relevant need of specific types of potential user in specific types of extra-lexicographic situation (for a discussion of the fundamentals of specialized lexicography, see Nielsen 2018a). For example, dictionaries may: (1) provide help to understand legal texts in the native or a foreign language; (2) provide help to produce legal texts in the native or a foreign language; (3) provide help to translate legal texts into or from a foreign language; (4) provide help to acquire general or specific knowledge about factual and linguistic matters concerning law and legal language. Examples (1)–(3) are referred to as communicative functions, because help is provided to users engaged in communicative activities, and Example (4) concerns what is often referred to as cognitive functions, because help is provided to users engaged in knowledge building.

The object of legal lexicography should be seen in the following light. Firstly, the theoretical and practical framework focuses on the type of help legal dictionaries intend to provide when consulted by users. Secondly, legal lexicography works with any type of data, whether linguistic or encyclopedic, which is in line with the conclusions drawn by Haiman (1980, 331) as well as Harris and Hutton (2007, 155), who state that “The boundaries between a general dictionary, an encyclopedia, a law dictionary and a legal encyclopedia are inevitably unclear, both in theory and as a matter of practice.” Thirdly, the above lexicographic basis prioritizes users and their needs for information when they consult legal dictionaries. These considerations support the position advocated in this contribution namely that legal lexicography deals with terms and general-language words, works with both systematic and alphabetic organization structures, is descriptive as well as prescriptive, addresses laypersons as well as experts, and prepares dictionaries for both decoding and encoding. On the basis of the above ontological position, the object of legal lexicography is information tools that function as utility products, which–through their surface and underlying features–provide specific types of help to specific types of user in specific types of user situation related to the field of law and its language. Even though legal dictionaries cover law and legal language, lexicographic projects require a plurality of skills and competences.

3.The interdisciplinary nature of legal lexicography

A study of available dictionaries and dictionary projects shows that legal lexicography is characterized by interdisciplinarity. As we have seen, legal lexicography is not a sub-discipline of linguistics or terminology with linguistic units or terms as its object but rather a discipline, the object of which is the dictionary. If the aim of a lexicographic project is to make a dictionary intended to provide help with linguistic aspects of legal language, then linguistic knowledge is required and (juri)linguists should be invited to assist. Similarly, if a project aims to make a dictionary that provides help with terminological aspects of law, then knowledge of terminology is required and terminologists/terminographers should be invited to assist. As pointed out by Bajčić (2017, 142–143), linguistics and terminology are two approaches to legal lexicography, which means that work with legal dictionaries is an exercise in lexicography centred on the dictionary and supported by, for instance, a linguistic or a terminological approach. Legal lexicographers may thus have to call in aid from specialists working in disciplines other than law and lexicography.

Legal lexicography involves theoretical and practical principles for the production of dictionaries, so several disciplines are likely to be involved in any project on legal dictionaries. Firstly, IT specialists contribute to developing databases and user interfaces, and legal lexicographers should be able to communicate effectively with those specialists (or acquire the relevant skills and competences for doing that work themselves). Secondly, modern legal dictionaries use electronic corpora as sources; therefore, knowledge about corpus building and corpus analysis is relevant. Thirdly, legal dictionaries contain terms and general-language words from the general field of law or one or more of its sub-fields, so domain-specific knowledge and cooperation with legal experts are necessary in relation to both the internal structure of the field of law and the language used. Finally, legal dictionaries are information tools designed to fulfil the needs of users in specific types of extra-lexicographic user situations and knowledge about and cooperation with specialists of, for instance, translation, knowledge management, and text production are necessary (for a discussion of the relationship between lexicography and interdisciplinarity, see Nielsen 2018b). When they have the necessary (inter)disciplinary competences and skills available for a particular project, lexicographers should link those to the lexicographically relevant needs of intended users.

4.Users and their lexicographic needs

When you want to discuss how lexicography can help with legal terminology and term management, a good place to start is to study the intended dictionary users, as dictionaries are compiled to provide help to satisfy the needs users have. The various types of competence of intended dictionary users either help users or place them at a disadvantage in a specific extra-lexicographic situation, for example understanding and writing legal texts. The levels of competence indicate which lexicographically relevant information needs the intended users have and provide guidance to lexicographers for selecting the data necessary for responding to such needs. Lexicographers distinguish between three general types of user: subject-field experts, semi-experts, and laypersons (Bergenholtz and Kaufmann 1997, 98–99). These three types are rather broad but in respect of legal dictionaries, subject-field experts may be described as users with specialist legal knowledge, semi-experts as users with legal knowledge below the level of specialists and above the level of basic, general knowledge of law possessed by people with a general higher education. Laypersons also constitute a heterogeneous group ranging from users with no legal knowledge to the general legal knowledge of those with a higher education. This typology may be applied to users who look for legal knowledge in dictionaries, but does not take into account users who have other needs, such as legal translators and teachers of a foreign language, who will often possess translation and language skills relating to legal language that may be relevant for planning and making legal dictionaries (see further below).

The members of these diverse groups have different cultural, legal, linguistic, translation, writing, etc., competences, and dictionaries ideally contain data that help users where competences are inadequate. The existing literature discusses different ways in which to profile dictionary users, and their competences and skills can, for instance, be identified by answering a list of questions like the following (adapted from Nielsen 2014, 159–160):

1. What is the level of their general cultural and factual knowledge?

2. At what level do they master the field of law in their own jurisdiction?

3. At what level do they master the field of law in a foreign jurisdiction?

4. Which language is their native language?

5. At what level do they master their native language?

6. At what level do they master a foreign language?

7. How extensive is their experience in translating between the languages in question?

8. At what level do they master legal language in their native language?

9. At what level do they master legal language in the foreign language?

10. At what level do they master writing legal texts and genre conventions in their native language?

11. At what level do they master writing legal texts and genre conventions in the foreign language?

The answers to these, or similar dictionary-specific, questions will indicate which general types and levels of competence the intended users of a dictionary have and enable lexicographers to put data into the dictionary that help users where competences are insufficient. It is often practically impossible to identify all relevant competences of each potential dictionary user, so the social science concept of ideal types of users will normally suffice. According to Allen (2004, 76–78), ideal types are not the product of the impulsive urge of researchers but are logically constructed concepts, which can be applied to lexicographic projects. However, it is possible to make legal dictionaries dedicated to one or a very limited number of users, for example, for specific in-house terminology management processes in companies and for legal translation service providers. It is reasonable to assume that, as ideal types, legal experts and semi-experts generally have considerable factual knowledge of their own legal system but little to medium knowledge of facts in a foreign legal system. Similarly, legal experts and semi-experts will likely have little to medium competence in relation to a foreign (legal) language, including writing skills. Professional translators may have a considerable general-language competence and a medium to high competence in legal language. They are also likely to have factual knowledge of law ranging from little to in-depth, considerable translation competence, and considerable competence in writing general and legal texts in their native language as well as a foreign language. Consequently, laypersons and professional translators will need more, and perhaps different, data on factual legal matters than legal experts and semi-experts, whereas legal experts and laypersons will need more data that can help them with legal language, writing and translation issues. All user groups will need data on factual differences between terms from one jurisdiction and their counterparts in another jurisdiction as well as data on legal specifics in a foreign jurisdiction, as only few users will be familiar with these types of particulars. As a result, legal dictionaries need different types of data to satisfy the needs of their users, who may have asymmetric competence and knowledge levels.

The lexicographic needs of users are directly linked to the range of coverage of dictionaries. Nielsen (1994, 39–43) proposes the following typology of legal dictionaries. Single-field dictionaries cover one subject-field, such as a dictionary of law and a dictionary of mechanical engineering, whereas multi-field dictionaries cover two or more subject-fields, such as a typical business dictionary and a dictionary of science and technology. Single-field dictionaries come in two types. General-field dictionaries attempt to cover an entire subject-field, such as a dictionary of law and a dictionary of finance, while sub-field dictionaries attempt to cover one sub-field within a general field, such as a dictionary of contract law and a dictionary of land law. These types of dictionary have advantages and disadvantages. Multi-field dictionaries are likely to treat relatively few terms and general-language words from each subject-field, whereas sub-field dictionaries are likely to give in-depth treatment of relatively many terms and words from a sub-field. Taking a different perspective, professional translation service providers may use multi-field dictionaries for either one specific client or for two or more clients in their terminology management process. A translation service provider that translates many different types of text for a specific client can benefit from a multi-field dictionary that provides help to translate, for instance, legal documents, technical documents and marketing documents for that client. A translation service provider that specializes in legal translation may choose to use a general-field dictionary or several sub-field dictionaries covering law and legal language.

The above typology of legal dictionaries is also related to user types. In respect of single-field dictionaries, experts will be users with specialist legal knowledge and for sub-field dictionaries, experts will be users with specialist legal knowledge within their sub-field, for example land law specialists. In relation to sub-field dictionaries, semi-experts will be users who are specialists in legal sub-fields other than that treated by the dictionary concerned. The user category of laypersons may include learners of law and legal language, novice and seasoned legal translators, as well as jurilinguist translators, and these user types may relate to any one of the three types of legal dictionaries specified above.

The disciplinary boundaries of the subject field of law and its terms are easily established–or so it would seem. At one point in time, law may have been easy to define as a pure discipline but especially after the industrial revolution, other disciplines have affected law in ways that are relevant to legal lexicography and the treatment of terms. For example, securities law concerns legal rules pertaining to non-legal subject matter such as various types of actors in financial markets, and terms from information technology and its subject matter affect IT law and its terminology. In sum, the above discussion shows that legal dictionaries can be general-field and sub-field dictionaries and either type of dictionary may provide help in the form of communicative functions and cognitive functions, or a combination of the two. Finally, these dictionaries may have legal experts, semi-experts or laypersons (or a combination thereof) as their target groups. For legal dictionaries to work in practice, it is relevant to look at how the above considerations may affect the selection of terms and other types of data.

5.Selection of terms

The selection of terms and general-language words that function as search entries in printed and electronic dictionaries depends on dictionary functions, range of coverage, and intended user groups. General-field dictionaries have a wider range of coverage than sub-field dictionaries, and dictionaries providing help to produce and translate legal texts will likely include terms as well as general-language words (e.g. nouns, verbs, adjectives, adverbs) in order to facilitate production and translation of texts. Lexicographers often refer to search entries as lemmas, because they are not always words in a grammatical sense. Some multi-word terms contain different parts of speech (e.g. alphanumeric terms) and can therefore not easily be classified as words. Lemmas in a UK English legal dictionary (general-field) may include the following examples: ABH, bind over, construe, contract for work done and materials supplied, -ee, Part 20 claim, payment into court, quasi-judicial, royalty, and strike out. Whether a term or word is to be included in a legal dictionary depends on whether the term or word is relevant for providing the help that the intended user group need for successfully carrying out particular types of communicative or cognitive activity.

Terms are important parts of legal vocabulary and may be classified in various ways. A traditional classification of legal vocabulary can be found in, for instance, Alcaraz and Hughes (2002, 16–18), who distinguish between legal terms of art, which are terms that are exclusively found within the domain of law and which are monosemic, such as barrister, tort, and estoppel from Anglo-American law. Semi-legal terms (or mixed terms) are words and phrases from the general vocabulary that have acquired additional special meanings within the field of law in narrowly defined legal contexts. These terms are therefore polysemic and include the term(s) issue from Anglo-American law, which has the legal meaning “children” within the law of wills and succession, and the meaning “disputed point” in procedural law. Finally, everyday vocabulary with everyday meanings is found in legal communication. A similar classification distinguishes between pure law terms, which correspond to legal terms of art, and law terms found in everyday communication. The latter seem to correspond to semi-technical legal terms, such as the Anglo-American term land, which is used in everyday communication by someone buying or selling a house, while the legal meaning is a delineated piece of landed property with buildings, plants, supply lines and rights attached. A third type is made up of everyday words assigned a special connotation in a given legal context, such as words or phrases regularly and primarily used in everyday communication that “become a part of the subject-matter of a statute thus acquiring new semantic dimensions either expanding or narrowing their original meaning” (Chromá 2004, 15–16). When they have decided on which types of legal vocabulary to include in their dictionaries, lexicographers will start the process of selecting the relevant terms.

Specialized lexicography has long embraced the use of systematic classifications for delineating subject fields and selecting lemmas, which is also relevant for making legal dictionaries. Petersen (1995a, 83–84) suggests the use of three types of systematic classification. An external subject classification delimits the legal subject field in relation to neighbouring disciplines and this classification is useful for compiling corpora of representative texts in which terms can be found. The next step is to prepare an internal subject classification that helps lexicographers to structure the legal field so they can work systematically when making their dictionaries as it helps lexicographers to present this structure in textual, diagrammatical or pictorial form to dictionary users. Finally, a terminological classification provides lexicographers with a list of the terms used in the legal field and helps lexicographers to extract the terms.

This three-step selection process may be refined for legal dictionaries. Nielsen (1994, 134–137) suggests that the general field of law should be divided into all the sub-fields that can be identified within a given jurisdiction, for example, contract law, environmental law, family law, insolvency law, land law, tort law, and jurisprudence. A specification like this will provide a proper basis for selecting the lemmas to be included in legal dictionaries. Furthermore, this method provides a detailed account of the terms and general-language words of a legal sub-field and at the same time makes it easy for lexicographers to identify collocations and phrases that may be useful for some dictionary functions.

This method may be extended to legal sub-field dictionaries. It may be sufficient to examine a sub-field in a general way, but may be helpful to break down a sub-field into constitutive thematic elements in order to obtain an overview of the legal sub-field. For instance, the sub-field of contract law may be divided into the following thematic elements (Nielsen 1994, 136):

1. Classification of contracts

2. Formation and validity of contracts

3. Legal effects of contracts

4. Judicial implication and construction of contracts

5. Performance of contracts

6. Non-performance and remedies

7. Assignment of contractual rights and liabilities

8. Discharge of contracts

This type of classification provides a workable basis for the selection of lemmas. The number of constitutive thematic elements may vary from one sub-field to another and may even have to be specified further by adding thematic sub-elements to one or more of the general elements, for example:

When selecting data for legal dictionaries, lexicographers should treat law as a jurisdiction-dependent domain. This approach is advocated by de Groot (1990, 122), who claims that “Die Fachsprache der Juristen ist extrem systemgebunden. Rechtssysteme sind von Staat zu Staat unterschiedlich. Jeder Staat hat seine eigene selbständige juristische Terminologie.” In other words, each jurisdiction structures its legal system in a way that suits only that jurisdiction and the legal language used reflects the structure of that system. However, adjectives such as English, German, and Spanish used in legal contexts may not only refer to the legal framework of a particular country, such as US English, UK English, and EU English, but may refer to varieties of English, German, and Spanish used in different legal systems. As pointed out by Nielsen (2014, 161), varieties of English are used in international treaties and conventions, and these texts aim to describe regulatory frameworks in a language that can be understood by and is acceptable to as many as possible no matter which legal system they belong to. Furthermore, labels such as US English and UK English refer to countries that have multiple jurisdictions, such as a federal jurisdiction and 50 state jurisdictions in the USA, and the jurisdictions of England and Wales, Scotland, and Northern Ireland in the UK. To the extent relevant, legal dictionaries should specify any geographical and jurisdictional constraints relating to terms and substantive issues.

Selection of terms pertains not only to lemmas but also to equivalents, and the selection of equivalents involves some of the methods and considerations described above. Legal lexicographers often compile text corpora in one or more languages and these should be up-to-date and organized in a systematic or thematic way concerning both substantive content and genres. Petersen (1995b, 106–110) describes a five-step procedure for selecting equivalents that is relevant for the selection of equivalents in legal dictionaries. The first step is introspection, where legal lexicographers and legal experts use their prior knowledge of correct equivalents. The second step involves searching the corpora to find confirmation of potential equivalents and to discover distinctions between possible equivalents, such as variants and conceptual hierarchies. The third step involves searching corpora for thematic and genre parallelism in order to make relevant substantive and language comparisons. For instance, legal textbook excerpts in one language should be compared with excerpts from legal textbooks in the other language dealing with the same substantive issues. Similarly, contracts in one language should be compared with contracts in the other language on the same subject matter. The same applies when corpora contain different language versions of EU legal acts, such as regulations and directives. The fourth step concerns the search for terms of abstract concepts, which are often difficult to find because of the differences in legal systems (e.g. the concept of corporate veil in Anglo-American law). Law is a culture-dependent subject field and it may be problematic to say that a term in one language is actually an equivalent of a term in another language. Not only does this make it difficult to present equivalents in dictionaries but it also makes it difficult to search for equivalents in corpora. This emphasizes the value of having fully systematic classifications for each language and jurisdiction concerned.

The application of the above method may discover instances of various degrees of equivalence between terms from different jurisdictions. The existing literature generally distinguishes between three types of equivalence in respect of lemmas and equivalents (see e.g. Šarčević 1989, 279–281; de Groot 1990, 122–127; Nielsen 1994, 162–178). Full equivalence means that a term in one jurisdiction has exactly the same semantic, pragmatic and substantive properties as a term in another jurisdiction. There is seldom full equivalence between terms from different jurisdictions, but legal concepts and terms defined in EU regulations would appear to be fully equivalent in all official EU languages because, strictly speaking, they belong to the same supranational jurisdiction. EU member states may adopt defined concepts and terms in EU directives without changes in their national law and these may be examples of concepts and terms with full equivalence. In this connection, Bajčić (2017, 112) argues that the concept of equivalence is difficult to define and that “equivalence is difficult to achieve and legal equivalence well-nigh impossible to achieve.”

As a result, incongruence between legal systems leads to partial equivalence between lemmas and equivalents, i.e. terms from two jurisdictions do not have the exact same semantic, pragmatic and substantive properties. Partial equivalence may result in there being more than one possible equivalent to a lemma and in those cases dictionary users should be informed of the semantic, pragmatic and substantive properties of each equivalent adopted.

Various approaches are available to legal lexicographers dealing with partial equivalence. An equivalent may have properties that constitute a subset of the semantic, pragmatic and substantive properties of a lemma (inclusion). Alternatively, part of the set of semantic, pragmatic and substantive properties of an equivalent may be equal to part of the set of the same properties of a lemma (intersection). A supplementary approach is to examine essential and accidental characteristics of terms in order to determine the degree of partial equivalence. However, it may be difficult to distinguish between essential and accidental characteristics in a meaningful manner. For instance, it may be debatable how many essential and how many accidental characteristics are required to determine the degree of partial equivalence, especially since terms may have from one to many essential characteristics and none to many accidental characteristics within or across jurisdictions.

Owing to the terminological and conceptual differences between legal systems, lexicographers may find that legal concepts in one jurisdiction have no corresponding concepts in another jurisdiction (zero equivalence). In such cases, lexicographers may decide either to exclude terms as lemmas or to provide suggested equivalents such as explanatory phrases (see Section 7 below). In cases of partial equivalence, lexicographers need to decide which partially equivalent terms to select. This challenge may be solved by dividing partial equivalents into two overall groups. Acceptable partial equivalents are equivalents that have the properties necessary to function satisfactorily as translation equivalents of lemmas. Such equivalents may be described as near or approximate equivalents. The other group is unacceptable partial equivalents, which are terms that do not have the properties to function satisfactorily as translation equivalents, and may be described as remote equivalents.

Lexicographers face the problem of how to select translation equivalents, or rather surrogate equivalents, in cases of zero or remote equivalence. One solution is a lexical borrowing (also referred to as transcription) where a term from one jurisdiction is taken over as an equivalent (loan word). However, a lexical borrowing does not result in the presentation of actual legal equivalents, because they do not normally exist in the target jurisdiction/language unless such loan words (i.e. terms) are established terms in the legal language of a given jurisdiction. Lexicographers may also use loan translations (also referred to as calque and word-for-word translation), which means that the morphemes of lemmas are translated item by item (i.e. directly) into equivalent morphemes in another language. The main problem with this method is that dictionary users are presented with equivalents that do not form part of the legal vocabulary of their own or another jurisdiction. Furthermore, lexicographers may decide to use explanatory equivalents. These are descriptive equivalents having the purpose of explaining the semantic, pragmatic and substantive properties of lemmas, and which can function satisfactorily as translation equivalents (see Section 7 below). Finally, lexicographers may invent new terms (also referred to as coinage and neologism). This may be equivalents lexicographers invent themselves or equivalents suggested by legal experts, but there is always the risk that such equivalents will be misunderstood or not used because they are unknown to the legal community.

The selection of source-language terms and target-language terms for legal dictionaries requires systematic and thematically based methods concerning the structure of the field of law, its language, and the different text types involved. Furthermore, selection is an ongoing quantitative and qualitative process in which persons with lexicographic, legal, terminological, linguistic and use-related knowledge collaborate in the process of selecting terms. For example, Nielsen (2015) describes cases where lexicographers failed to identify and incorporate legal terms despite using comprehensive, up-to-date electronic corpora. However, legal dictionaries contain much more than terms.

6.Selection of other types of data

Legal dictionaries contain different types of data selected because they support dictionary functions and the treatment of legal terms. Some data types support several functions whereas others support only one, and legal lexicographers should select those data types that match user competences, user situations and user needs. Bergenholtz and Tarp (1995, 22–28) discuss data types in specialized dictionaries and these types of data are also relevant for legal lexicography. Partly based on their discussion, the following list indicates which data types legal lexicographers should consider including in their dictionaries, with due regard to dictionary functions:

1. For producing legal texts in their native language, users typically need data on orthography, part of speech, gender, pronunciation, grammatical regularity and irregularity, definition, collocations, phrases, and usage.

2. For understanding legal texts in their native language, users typically need data on part of speech, gender, pronunciation, grammatical irregularity, definition, collocations, phrases, synonyms, and antonyms.

3. For producing legal texts in a foreign language, users typically need data on orthography, gender, pronunciation, derivation, syntax, grammatical regularity and irregularity, definition, translation equivalent, translation of collocations, translation of phrases, pragmatic and cultural restrictions, as well as data contrasting structure of own and foreign legal subject field.

4. For understanding legal texts in a foreign language, users typically need data on part of speech, gender, pronunciation, grammatical irregularity, definition, translation equivalent, translation of collocations, translation of phrases, synonyms, and antonyms, as well as data contrasting structure of own and foreign legal subject field.

5. For translating legal texts into a foreign language, users typically need data on the native language on part of speech, gender, pronunciation, grammatical regularity and irregularity, definition, collocations, and phrases. On the foreign language: orthography, part of speech, gender, grammatical regularity and irregularity, translation equivalent, degree of equivalence, translation of collocations, translation of phrases, pragmatic and cultural restrictions, and data contrasting structure of own and foreign legal subject field.

6. For translating legal texts from a foreign language, users typically need data on the foreign language on part of speech, gender, pronunciation, grammatical irregularity, definition, collocations, and phrases. On the native language: orthography, part of speech, gender, grammatical regularity and irregularity, translation equivalent, degree of equivalence, translation of collocations, translation of phrases, pragmatic and cultural restrictions, and data contrasting structure of own and foreign legal subject field.

7. For acquiring knowledge about their own legal language, users typically need data on orthography, part of speech, grammatical regularity and irregularity, syntax, derivation, definition, and usage.

8. For acquiring knowledge about foreign legal language, users typically need data on orthography, part of speech, grammatical regularity and irregularity, syntax, derivation, definition, usage, data contrasting native and foreign legal language, and data contrasting structure of own and foreign legal subject field.

9. For acquiring knowledge about their own field of law, users typically need data on the structure of the legal subject field, definitions, and conceptual relationships.

10. For acquiring knowledge about a foreign field of law, users typically need data on the structure of the legal subject field, definitions, conceptual relationships, and data contrasting structure of own and foreign legal subject field.

This list needs a few clarifying comments. Firstly, some of the above data may come from specific sources of information, such as statutes, case law and books, and lexicographers should consider including data on sources (citations) for some communicative and cognitive functions (see Figure 1). Secondly, the main reason for including the data type pronunciation is that a wide understanding of text includes both written and oral communication and the optimal legal dictionary would provide help with both types of communication. Thirdly, several data types support more than one function, for instance, equivalents are a data type that supports foreign-language production as well as translation activities, but may also help understanding foreign-language legal texts and knowledge acquisition, in particular for laypersons and semi-experts. Collocations and phrases are useful because they show legal terms in context and may thus clarify the meaning of those terms by specifying relevant referential foci. As explained by Biel (2014, 46–47), collocations embedding legal terms “establish links between terms and elements of conceptual frames,” and collocations may disambiguate “near-synonyms and polysemous senses of a term, because they tend to combine with different collocates.” Finally, definitions appear to be a data type used across the board and may thus be regarded as an indispensable data type in legal dictionaries. However, legal lexicographers should heed the words of Harris and Hutton (2007, 49), “The notion of a definition adequate to all occasions and all demands is a semantic ignis fatuus.” This implies that different definitions are required for different users and functions.

Legal dictionaries usually contain definitions presented in various ways. The existing literature discusses several types of definition based on various approaches, such as semantic definitions, encyclopedic definitions, synonym definitions, extensional definitions, intensional definitions, partitive definitions, relational definitions, and categorial definitions (see e.g. Harris and Hutton 2007, 111–123; Fuertes-Olivera and Arribas-Baño 2008, 47–51). From a functional perspective, definitions of legal terms should contain data that help dictionary users satisfy their information needs in specific types of extra-lexicographic situations. This applies to the substantive or linguistic content of definitions as well as the way and the language in which the data are presented. Accordingly, the competences identified in the user profile linked to the activities in which users are engaged play a role in writing definitions.

Dictionary definitions should be adapted to those user situations in which they can satisfy user needs. This concerns the understanding of concepts, and several scholars have attempted to explain what it takes to understand concepts (and terms). Peirce (1998, 138) offers one explanation: “Consider what effects that might conceivably have practical bearings we conceive the object of our conception to have: then, our conception of those effects is the whole of our conception of the object.” In other words, you have not fully understood a concept until you have identified all the practical consequences the concept may possibly have. Writing definitions that meet Peirce’s requirements is a tall order but from a practical point of view, definitions in legal dictionaries that provide help in various types of situation should answer the following questions: (1) What is this concept (meaning explanation or term explication)? (2) What does this concept do (function or effect)? (3) How is this concept related to other concepts (relation)? Definitions based on these three elements provide users with a basis for understanding legal concepts and acquiring the knowledge needed to understand, write and translate legal texts (see also Chapter 2 in this volume on definitions in law across legal cultures and jurisdictions).

Dictionaries designed to provide help to translate legal texts contain various types of data related to the definition of terms. A multilingual judicial dictionary on the law of procedure (Hjelmblink 1991) shows how a terminological approach can be applied. The dictionary is intended for use by Danish lawyers and non-lawyers, translators, civil servants, and businesspeople, and primarily covers terms from England and Wales, Scotland, France, and Germany. Each jurisdiction is treated in separate sections and the terms are translated into Danish. Figure 1 shows a dictionary article (sometimes referred to as an entry) from the German-Danish section.

Figure 1 contains a Danish definition of the German term, which follows a label allocating the term to German civil procedure (indicated by the abbreviation ZP). The source of the definition is given in a reference to a specific place in a book on German civil procedure (Jauern. § 88), the full bibliographical data of which appear in the dictionary’s front matter. The indication of source precedes the Danish term that functions as translation equivalent written in bold. The second half of the dictionary article describes the conceptual hierarchy to which the German term can be assigned. Users are informed that the term Klagenhäufung is the superordinate term (indicated by the upward-pointing arrow), and that the term objektive Klagenhüfung is an antonym of subjektive Klagenhäufung (indicated by the symbol ≠ and of which Streitgenossenschaft is a synonym). Furthermore, the article tells users that there are three types of objektive Klagenhüfungen: alternative Klagenhäufung, eventuelle Klagenhäufung, and kumulative Klagenhäufung. The downward-pointing arrows indicate that the last three terms are at the hierarchical level immediately below that of objektive Klagenhäufung. The article ends by referring users to a diagrammatical representation of the conceptual hierarchy (a tree diagram) in a separate appendix.

In addition to the terminological approach to legal lexicography, this multilingual judicial dictionary shows how lexicographers can place data concerning legal facts and conceptual hierarchies in different places in the dictionary and through cross-reference structures make the differently located data work together to satisfy user needs. Finally, each of the four bilingual sections starts with a Danish description of the judicial system in the non-Danish jurisdiction concerned written in plain text and supported by a diagrammatical representation of the court system. These descriptions support and supplement the definitions provided in individual articles with both background and systematic legal knowledge that is helpful to the intended users.

The principles of legal lexicography also allow lexicographers to adopt a comparative-law approach to translation dictionaries. Two dictionaries, one on Norwegian-English contract law and one on Norwegian-English law (Craig 1992; Craig 1999), illustrate how a comparative approach may be applied. The dictionaries are intended to satisfy the information needs of, in particular, Norwegian translators, but also the business community, lawyers, state and administrative personnel and teachers. The dictionaries aim to build a bridge between the Anglo-American and the Scandinavian legal traditions in order to avoid misunderstandings by giving users a basic understanding of Anglo-American law. Figure 2 shows how the comparative approach may be used.

The dictionary article illustrated in Figure 2 contains definitional data in English even though the intended user groups are Norwegians. Furthermore, the definition of the Norwegian term is embedded in the comparative explanation, the main focus of which is Anglo-American law (AAL in Figure 2). Nevertheless, the contrastive data expand the definition of the Norwegian term and, being in English, the definition indirectly helps translators and other intended user groups to translate Norwegian texts in a proper way. In the dictionary’s user guide, the author explains that if there is no separate definition of the Norwegian term the English equivalent is a direct equivalent of the Norwegian term and the two terms thus have the same meaning (Craig 1999, viii). Figure 2 is a relatively short article with comparative-law features. The two Norwegian-English dictionaries contain considerably longer articles that give in-depth comparative descriptions of the differences between Norwegian and Anglo-American law. For example, the article treating the Norwegian term hevning (generally: termination of contract) is about 9 pages long in both dictionaries and compares the remedies available to contractual parties who seek to deprive contracts of their binding effect because of breach, invalidity or frustrating events.

Legal lexicographers have used the comparative-law approach in dictionary components complementing and supplementing definitions in articles. Such components are referred to as subject-field components, the purpose of which is to give a systematic presentation of the legal field(s) treated by a dictionary (for a discussion of subject-field components in dictionaries, see Bergenholtz and Nielsen 2006). A multilingual dictionary on contract law covering the jurisdictions of England and Wales, Denmark, Norway, and Sweden (Anderson 1977) has a subject-field component with several sections written in English and divided into numbered paragraphs. The dictionary’s target group is English and Scandinavian practising and academic lawyers as well as businesspeople and its primary function is to help users understand English terms. The first subject-field section contrasts general features of Anglo-Scandinavian law, and the second section describes some shared general features of Scandinavian law. Section three explains the meanings of common law and equity in English law, while section four provides an outline of the English judicial system at civil law. The last three subject-field sections provide outlines of the judicial systems at civil law in Denmark, Norway, and Sweden, respectively. To help users understand the terms included in the English-Scandinavian wordlist, several articles contain cross-references to specific, numbered paragraphs in the relevant subject-field sections where users can find additional information. The subject-field sections follow the Länderbericht method, which describes all the relevant legal issues in one jurisdiction before describing all the relevant issues in another jurisdiction (Zweigert and Kötz 1998, 32).

Lexicographers use methods other than the Länderbericht method when writing subject-field components in legal dictionaries. A Spanish-Danish legal dictionary intended for Danish companies, institutions and individuals who come into contact with the Spanish legal system contains a short contrastive description of selected aspects of the general field of law in the two jurisdictions (Henriksen 1991). The dictionary has two functions, namely to provide help to understand Spanish legal terms and concepts, and to provide help to translate Spanish legal texts into Danish, and the Danish subject-field component aims to provide information complementing and supplementing the data in the articles. The author opted for the analytical method, which compares one legal issue across jurisdictions before going on to compare another legal issue across jurisdictions. The subject-field component has numbered paragraphs dealing with issues including, but not limited to, the law of persons, family law, property law, and criminal law.

For some communicative functions, for example providing help to translate and produce legal texts in a foreign language, lexicographers may elect to present definitions in two languages. A Norwegian-English dictionary of law (Lind 2003) intends to provide help for translators, interpreters, lawyers and others who need to transpose Norwegian legal terminology to English. The author chose to provide definitions of central terms of Norwegian law in both Norwegian and English, and Figure 3 shows one example of this approach.

In a dictionary intended to provide help to translate and otherwise communicate in English aimed at informed laypersons as well as professional lawyers, who are native speakers of Norwegian, the inclusion of definitions written in the native language as well as translated into the relevant foreign language may benefit dictionary users in at least two ways. Firstly, users will know whether they have found the terms they need help with by reading the native-language definitions. Secondly, the translated definitions may help users in their foreign-language communication, whether users are informed laypersons or professionals. Such translated definitions are particularly helpful for users who have to explain legal concepts that are unfamiliar to the receiver of the communication because of structural differences in the legal systems known to the interlocutors. Definitions are only one type of data in legal dictionaries and are often linked directly to other data types and the way in which these are presented.

7.The presentation of data

The data in dictionaries can be presented in many ways. Dictionaries–whether printed or digital–involve databases as well as interfaces and legal lexicographers can, and do, place the different data types in various places. For example, dictionary articles often contain a plurality of data types, such as lemmas, grammatical data, definitions, collocations, phrases and cross-references. The presentation of terms plays a major role in legal dictionaries and terms are lemmas and equivalents, which lexicographers can present in different ways.

Some terms are written identically (homographs) but relate to different concepts. Legal lexicographers therefore need to decide how to present identically looking lemmas and at first glance, the choice is between polysemy and homonymy (see e.g. Nielsen 1994, 206; 242–244). These options are based on linguistic principles and may be appropriate for legal dictionaries assisting text production and translation. Theory and practice suggest that lemmas should be presented in such a way that the terms and the related data allow users to process the data optimally, thereby finding the answers to their consultation queries. The English term charge refers to several concepts, including: (1) a criminal accusation, (2) a judge’s instruction to a jury, (3) an expense, and (4) an encumbrance. Lexicographers can treat charge in four separate articles as homonyms or in one article as polysemes, whichever is best for the intended users.

Dictionaries intended to provide help with understanding, producing or translating legal texts should include lemmas other than terms. The English word charge appears in different types of legal communication as a verb with various meanings, including: (1) to make a criminal accusation, (2) to instruct a jury, (3) to ask for payment, and (4) to encumber. In dictionaries containing homographs belonging to different word classes, lexicographers should treat lemmas as homonyms at the level of word class and treat charge as nouns and verbs in different articles. Lexicographers should treat homograph terms and general-language words belonging to the same word class but with different inflectional paradigms in a similar way. For example, the English term authority refers to at least two concepts, namely: (1) the power to act on behalf of and bind another person (in which case it is uncountable), and (2) an official government body (in which case it is countable). This approach is relevant for the database functioning as the basis of the dictionary and the data fields in which the relevant data are placed, because it allows lexicographers to offer users different search options in electronic dictionaries (see Section 8 below).

Lexicographers of bi- and multilingual dictionaries have various ways in which to present equivalents. In those cases where a lemma has two or more acceptable equivalents, lexicographers may opt for a descriptive or a prescriptive approach. A descriptive approach is often not helpful to users who want to use the correct term in communicative situations. This may be illustrated by the Danish term forarbejder, that is material that form the background to a legislative enactment, and some possible English equivalents. A descriptive approach may result in the presentation of lemma and equivalents found in Figure 4 (other data types are not shown).

One disadvantage of the descriptive presentation of equivalents is that it may let users down. Users who want to know which English term to use will have to select an equivalent more or less at random, especially non-experts, and they probably consult the dictionary to find clear and unequivocal guidance. A prescriptive approach may result in the presentation of the lemma and equivalents illustrated in Figure 5.

In Figure 5, the lexicographers decided to help users by recommending the first equivalent written in bold and by showing the other equivalent as a possible alternative (synonym) on a separate line. This way of presenting equivalents is normative but at the same time does not completely rule out the synonym (for a discussion of descriptive and prescriptive approaches in lexicography, see Bergenholtz and Gouws 2010). Lexicographers can also adopt a prescriptive approach in dictionaries used by, for example, translation service providers that have several clients who commission legal translations (for a discussion of the treatment of legal term variants in dictionaries, see Nielsen 2022). One way in which to use prescription in a legal dictionary for all clients is shown in Figure 6. Translation service providers may also prepare separate dictionaries for individual clients and in such a case, the approach in Figure 5 may be more appropriate.

Authors of bi- or multilingual dictionaries with communicative functions face a challenge when terms in the source language have no equivalent terms in the target language. This may occur when there is no equivalent conceptualization or classification in the legal system of the target-language jurisdiction, or when there is no equivalent term but an equivalent conceptualization or classification in the legal system of the target-language jurisdiction. In such cases, lexicographers can either exclude a term as a lemma in the dictionary, thereby providing no help to users, or include a term as a lemma and present data that will likely meet communicative and cognitive information needs (Nielsen 1994, 168). The latter approach was adopted by Craig (1999) in his Norwegian-English law dictionary and Figure 7 shows one example of how the author dealt with the challenge of presenting data relevant for the intended target group.

Figure 7 shows a Norwegian term that has no equivalent term in English, but Anglo-American law has roughly equivalent conceptualizations as Norwegian law. Rather than being a term, the suggested translation is an explanatory phrase indicating the substantive content of the lemma. The situation becomes more complicated when there are no equivalent conceptualizations or classifications in Norwegian and Anglo-American law, as illustrated in Figure 8.

Figures 7 and 8 show that the author had in mind the information needs of the intended target group in communicative situations. Norwegian lawyers and translators who have to produce texts involving Norwegian law need help to communicate in English even in situations where Anglo-American law has no equivalent conceptualizations or classifications. The dictionary provides help by suggesting English translations even though the translations are not established in Anglo-American usage. In Figure 8, the suggested translation is accompanied by an explanation of the Norwegian lemma and its relation to other Norwegian terms that together form a terminological hierarchy. Furthermore, explicit cross-references (indicated by asterisks) refer users to other articles in the dictionary where additional information is available. In doing so, the dictionary presents data that provide help for communicative as well as cognitive user situations.

The data presented in Figures 4–8 are shown as independent of medium. The articles could appear in printed as well as electronic dictionaries, but born-digital dictionaries offer a number of advantages for users to search for terms and other data.

8.Databases, search engines and interfaces

Databases are the practical foundation of online and born-digital legal dictionaries and the technological options available enable lexicographers to design legal dictionaries with different overall structures. Nielsen (2014, 155–156) explains that online legal dictionaries are complex lexicographic tools with three main components. Firstly, the database with specially selected and structured data ensure easy search for and retrieval of data. The database has discrete data fields each containing a specific type of data, such as fields that contain lemmas, fields that contain definitions, fields that contain grammatical data, and fields that contain collocations and phrases. These fields are connected through links establishing a relationship between the field containing a specific lemma and the field with the appropriate definition, the field with the appropriate grammatical data, etc. Such relational databases enable lexicographers to present users with different search options based on user information needs and search results that match those needs.

The second component is the dictionary’s user interface, i.e. what users consider the legal dictionary. Users will see a dictionary and a database as one, but lexicographers should treat them as two separate and interacting components. The dictionary is thus a user interface giving access to data and a database that functions as an electronic repository of structured data. Against this backdrop, a database may be used to make more than one legal dictionary, for example a dictionary, the function of which is to help users acquire substantive knowledge of legal matters, and a dictionary for producing legal texts. In addition, lexicographers can also extract data from the database to produce printed dictionaries.

The final component is a search engine operating as an intermediary between the user interface and the database. The search engine allows users to search data fields in the database, retrieves the relevant data, and presents them in pre-set systematic ways, i.e. lexicographers can use different data presentation formats for different users and different dictionary functions, thereby preventing information overload. Dictionaries, the function of which is to provide help to acquire substantive legal knowledge will show, e.g. lemmas and definitions, whereas dictionaries for producing legal texts will show, e.g. lemmas, definitions, inflectional paradigms, collocations, and phrases. A structural setup like this shows that the database, the search engine, and the user interface combine to form a whole, which is the legal dictionary.

The fact that a database may serve several legal dictionaries, each allowing users to make targeted searches in specific data fields, has both practical and theoretical implications. One is that online dictionaries may contain several independent lexicographic sections to which the search engine gives direct access. Such dictionary sections support the functions and use of dictionaries, such as user guides and subject-field components providing carefully arranged introductions to the field of law (see Section 6). As discussed in Nielsen (2010, 74–82), lexicographers may include sections providing help to translate special structures of legal language and legal genre conventions as well as sections with translated statutes, in those cases where such sections can help users satisfy their information needs.

Lexicographers working with bilingual legal dictionaries and bilingual databases may consider the possibility of making monolingual dictionaries available to users. Based on the data types described in Section 6 above, databases for bilingual dictionaries will contain many data in both languages concerned relating to law and legal language in the two jurisdictions involved. The article shown in Figure 3 from a Norwegian-English law dictionary may serve as an illustration. A database may contain all the data types shown in Figure 3 and located in separate data fields: Norwegian lemma, English equivalent, definition in Norwegian, and definition in English. If lexicographers use a triadic set-up as the one described above, the search engine can search for Norwegian data exclusively and present those in the user interface, which in this situation will be a monolingual dictionary. The dictionary will show the Norwegian lemma and the Norwegian definition to Norwegian users who have searched for the Norwegian term fremleie. The search engine may also search for both Norwegian and English data and thus function as a bilingual dictionary, in which case, the dictionary will show the Norwegian lemma, the definition in Norwegian, the English equivalent, and the definition in English. Depending on a dictionary function, lexicographers may decide that the bilingual dictionary will not automatically show the English translation of the Norwegian definition, so that the translated definition becomes an extra feature that requires user action. The translated definition may be available in a popup window or a “show more” menu to those Norwegian users who need that additional information.

Search options linked to user needs are those most likely to find data that satisfy those needs. If the database fields for inflectional data addressed to lemmas contain the full forms of the relevant inflectional paradigms, the search engine can find the lemmas typed into the search box and present the correct legal terms to users. This allows users to copy paste terms from texts into the search box no matter which inflectional variant they present, and users do not have to spend time thinking about canonical forms of legal terms as a requirement for getting help. This search option is particularly helpful for users who look for help with legal terms in a foreign language and who have limited substantive and language competences.

Dictionaries intended to provide help with translation into a foreign language or production of legal texts in a foreign language may also provide function-specific search options. One such option is to allow users to search in the database field for collocations, phrases or examples. Users who want to know how best to translate a collocation in their native language into the foreign language can type the collocation into the search box and if a match is found, this will be presented to users in a predetermined way. The dictionary may present the collocation and its translation either as a stand-alone result–only the source-language collocation and its translation are shown–or as an embedded result–part of an article in which the collocation and/or its translation are highlighted.

The distinction between database, search engine and user interface is relevant for data other than pure text. Legal dictionaries may provide pictures, graphs, and diagrammatical representations as supplements to definitions, either shown automatically as the result of a search, or shown when users click on a text unit or icon. In addition, lexicographers may decide to present certain data types in audiovisual form, for example audio files containing pronunciation of legal terms, and videos supplementing definitions. Finally, dictionaries may present links to dictionary-external sources allowing users to find additional information published in print or on the internet, for example statutes and judicial authorities.

As a final point, the triadic set-up allows lexicographers to give users access to the data in all data fields in the database, except fields reserved for internal communication among dictionary editors. In connection with dictionaries that intend to provide help to acquire knowledge about legal matters, users may need access to all the relevant data types described in Section 6. Lexicographers should, in addition to the search options described above, provide users with an option along the lines “if you want to know more” so that the search engine will show all data relevant for acquiring substantive (and language) knowledge about specific legal terms (see also Chiocchetti et al.’s contribution in this volume on multilingual legal terminology databases).

9.Concluding remarks

What is this thing called legal lexicography? Legal lexicography covers many different types of dictionaries based on various theoretical and methodological approaches. The printed and electronic dictionary contains several individual features, such as different types of data and text in various presentation forms, which together make up the legal dictionary. These features are directly related to the objective of legal lexicography, which is to develop principles and guidelines that help lexicographers design, evaluate, make and use legal dictionaries that fulfil specific types of need of specific types of user in specific types of communicative and cognitive situation in the real world concerning law, its terms, and its language.

Legal dictionaries can be used in many and varied situations, for example as part of term management processes, and legal lexicography deals with legal terms in both systematic and alphabetical set-up and presentation forms. Moreover, legal dictionaries can be both descriptive and prescriptive and can thus be used to clarify and standardize concepts and terms using methodological and systematic approaches. Finally, legal dictionaries can be used for both encoding and decoding purposes, and dictionaries can cater for the needs of lawyers as well as non-lawyers. Modern legal lexicography has adopted a broad scope of application so that dictionaries are regarded as information tools that deal with terms, words, facts and things related to law and legal language.