1.Introduction

The term phraseology is generally associated with a number of concepts, each of them focusing on a specific element that contributes to its definition. Among these terms are, for example, formulaicity and patternedness: they stress the fact that phraseology has to do with a pervasive feature of both spoken and written language, namely, repetitiveness. As a matter of fact, we speak like other people, we say things that have been said before (Hopper 1998, 159). Formulaicity refers to utterances which are similar to previous utterances and to the consideration that “anything that is said has been said in something like that form before” (Hopper 1998, 165), which is also the idea at the basis of machine translation: any phrase that is written has probably been written before, and translated, and the original and the paired translations are likely to be somewhere on the Internet (Bellos 2011, 253–254 in MacKenzie and Kayman 2016, 1). Languages are a massive inventory of formulaic expressions.

Similar to this concept is the idea that phraseology has to do with prefabricated language which evokes terms, such as: petrification, fossilization, prefabrication, fixedness, standardization, frozenness, embeddedness. The idea behind the use of these words is that phraseology has to do with constructions that have become conventionalized and, due to their interrelated pairings of form and function, are now highly fixed. Other terms associated with phraseology are rituals, conventionalism, conservatism, which refer to the fact that its use may consist in a vast collection of hand-me downs that reach back in time to the beginnings of language (or of a Language for Special Purposes, LSP) (Hopper 1998, 159). This allows for the association of another term, namely frequency, to the conceptualization of phraseology: to be considered as such, a phraseological unit has to meet the criterion of being frequently used in a type of text or language variety.

Many phraseological units exist because they “sound good” (MacKenzie and Kayman 2016, 1) which is in line with the so-called “acoustic inertia of the language” (Parks 2014, 220) – i.e., the idea that translation is also driven to a degree by the inertia of style and convention, meaning that the sound is as decisive as the sense in determining what gets said – and with the need of using a certain register and style in specific genres. This is why formality and style may be also associated with the term phraseology. Phraseology adds flavour to a text and is a stylistic feature of many LSP texts. They are diagnostic, vital elements and “powerful indicators of register”, as Partington puts it (1998, 17, 20 in Biel 2014a, 181).

However, phraseology may also exist because fixed and standardized expressions may be useful: they require little encoding and they get things done in communication (MacKenzie and Kayman 2016, 1). Phraseology is also associated with the term combinability and this goes back to the famous Firthian principle of “you shall know a word by the company it keeps” (Firth 1957, 11). Words tend to co-occur together and the frequent combination of two words may generate a collocation or a phrase. Closely related to the combinability and the above-mentioned concepts, is the idea of predictability and expectancy: language users (and readers of texts) may be able to predict the presence of certain fixed expressions and, in some cases like in legal genres, they may even expect the use of certain patterns in the text. Lastly, from a semantic point of view, phraseology tends to show another important feature, that is (non)-compositionality or syntactic (in)-flexibility, i.e., the fact that the meaning of a phraseological unit is not distributed among its components, but the expression as a whole is mapped onto its meaning, making impossible to decompose its linguistic elements.

This overview of concepts associated with the term phraseology helps to set the scene of this chapter and represents the background against which legal phraseology will be dealt with in the following sections. After defining (Section 2) and classifying (Section 3) legal phraseology from different yet complementary standpoints, the chapter exemplifies main traditional approaches (Section 4) as well as current and new ones (Section 5). It also reviews a selection of methods adopted to carry out research in this field together with some practical applications (Section 6). The final remarks (Section 7) aim at showing the advantages or disadvantages of using this kind of patterning in the discursive construction of legal texts.

2.Defining legal phraseology

Scholars have understood and accordingly defined phraseology in legal linguistics studies in different ways, following a variety of traditions and perspectives, mainly due to the inner complexity of defining a multifaceted concept. Characterized by terminological fuzziness and overlaps, LSP phraseology – and legal phraseology in particular – has usually been placed at the periphery of the discipline of phraseology (Goźdź-Roszkowski and Pontrandolfo 2015, 2018; Pontrandolfo 2013, 100–104), as a special case or exception from the rule (Kjær 2007, 506). This is also due to the frequent overlaps with the terminological sphere: legal phraseological units are often described as multi-word terms and are commonly dismissed as not being phraseological at all or left undescribed (Kjær 2007, 507).

A preliminary distinction should therefore be made: the term can refer either to patterns which are recurrent in legal texts (phraseology in legal language) or, as put forward by Kjær (1990b), to patterns which have a legal effect in a text (legal phraseology) (see also Ruusila and Lindroos 2016 and the distinction between content vs. non-content phrases in Section 3 applied to legal lexical bundles). The following examples will clarify the distinction: without prejudice to is a complex preposition which is frequently used in legal texts but which does not have a specific legal meaning (it can be replaced with equivalent expression such as “without harming or affecting something”); on the other hand, to dismiss the appeal is a phraseological unit (more precisely a collocation) which has a legal and performative effect in the text (i.e., the court refuses to hear and try a case already decided in a lower court); its use is phraseological in the sense that the term appeal collocates with the verb dismiss and it is the standard, legal way of expressing this concept in judicial proceedings. However, this distinction is not always applied in legal linguistics studies where the two terms are frequently used as synonyms. Moreover, the former actually includes the latter, so that it can be considered a more general term which also includes those specific types of units having a legal meaning in legal texts.

The focus of the present chapter is on both types of units, so the term legal phraseology (and legal phraseological units/phraseologisms) will be used as an umbrella label to denote typical and frequent patterns found in legal texts which may or may not have legal effects in legal documents. Moreover, as in the case of the term legal terminology, it is worth stressing that it may also refer to the discipline which studies the word combinations in legal settings. In her 1990a paper, devoted to the state of the art of LSP phraseology, Kjær underlines the ambiguity of the term phraseology at three different levels (1990a, 5–6):

1. term phraseology denoting phraseology theory in the terminological sense. Its subject is primarily the combinability of terms (LSP words);

2. lexico-phraseology denoting phraseology theory in the lexicological word combinations in language for general purposes. It stands in contrast to the inventory of free Language for General Purposes (LGP) word combinations;

3. LSP phraseology denoting the inventory of phraseological word combinations in language for specific purposes. It stands in contrast to the inventory of free LSP word combinations.

Most of the attempts at defining a legal phraseological unit have been based on the definition of phraseology in LGP (see, among others, Biel 2014b, 30–31; Pontrandolfo 2013, 67–105; Pontrandolfo 2020; Wray 2002, 9), where a wide range of labels have been used to define a phraseological unit (see Table 1) (see Granger and Paquot 2008; Nikitina 2017, 42–46; Pontrandolfo 2013, 67–87). Each of these terms focuses on a particular aspect of the word combination, such as its formulaicity or the frequency/recurrence of certain patterns.

The situation is rather similar in LSP, as shown in Table 2.

All these LSP denominations describe the technical nature of the combinations by means of adjectives like “specialized” and “technical” and allow the identification of one of the key features of phraseology in LSP, namely, the presence of terms. As will be demonstrated in this chapter, legal phraseological units tend to cluster around terms which are the building blocks of any LSP.

Most of the definitions of legal phraseology found in the literature are adapted from LGP studies. From a more traditional perspective, related to the “structure, meaning and use of word combinations” (Cowie 1994, 3168), a phraseological unit can be defined as a construction – in the sense of multi-word lexical unit or fixed expression – of a different nature (e.g., idioms, collocations, formulae, proverbs, sequences) in which its composing elements acquire a meaning which is not predictable from the sum of meanings of its constituents (see, among others, Cowie 1994; Granger and Paquot 2008). From a corpus-based perspective, it can be defined as:

the co-occurrence of a form or a lemma of a lexical item and one more or additional linguistic elements of various kinds which functions as one semantic unit in a clause or sentence and whose frequency of co-occurrence is larger than expected on the basis of chance.(Gries 2008, 6)

The advent of corpus linguistics was precisely the main factor making the traditional models “old” (see Section 4) and shifting the attention from idioms to less stable and more or less restricted word combinations that are usually found in LSP texts (Kjær 2007, 508).

These definitions stress the fact that phraseological units function autonomously in a clause or sentence and tend to co-occur with a certain frequency in a language or sublanguage (LSP). The key features have also been highlighted by Burger (2010, 14), who focused on two main characteristics shared by these word combinations: they consist of two or more words (polylexicality) and the combination is fixed (stability) (see Ruusila and Lindroos 2016, 122).

When it comes to legal phraseology, as emphasized by Ruusila and Lindroos, “the varying definitions and classifications of phraseological units used by researchers lead to difficulties in utilizing and comparing the research results across legal languages and legal systems” (2016, 128). To define a phraseological unit in legal language one could start by the definition of LSP phrasemes proposed by Gläser, such as “fixed, lexicalized, reproducible units that consist of two or more words” (2007, 487, translated by the author). This fixedness may or may not have a specific legal meaning, as previously mentioned. In the former case, as indicated by Kjær (1991, 115), such repetitively used formulaic expressions should be used as such and cannot be replaced by synonymous expressions. In opposition to other LSP (e.g., medicine or economics), legal phrasemes are not universal in their meaning, but bound to a particular legal system (Kjær 2007, 508), which leads to the necessity of studying legal phrasemes in connection with the legal system and legal culture in question (cf. Lindroos 2015, 166; Ruusila and Lindroos 2016, 130). An ad-hoc definition of legal phraseology could therefore be the following: crystallized lexical and/or morphosyntactic patterns, varying in complexity and internal cohesion, which are “handed down” from texts to texts and become recurrent in specific legal cultures and discourse communities. These patterns frequently co-occur in legal documents as a result of its use by legal professionals who, as part of a specific community of practice, employ them in their daily jargon so that these word combinations represent a discursive pattern typical of certain legal genres. These units may have a specific legal meaning or not, depending on the textual coordinates of the legal genre.

One of the key features of legal phraseological units is that they tend to cluster around legal terms, which is why it is necessary to make a distinction between a legal terminological unit (LTU) and a legal phraseological unit (LPU), even though the boundaries between the two concepts are fuzzy.11.An example could be “The A to Z guide to legal phrases” (http://​www​.plainenglish​.co​.uk​/files​/legalguide​.pdf): most of the terms labelled as “legal phrases” in the UK guide to plain English are legal terms and not phrasemes. The very fact that they can be defined and have a legal definition confirms that they refer to legal concepts and therefore are terminological units. LTUs tend to refer to concepts which can be legally defined whereas LPUs tend to be used for discursive/stylistic reasons, especially in the case of patterns belonging to the category phraseology in legal language. From a syntactic point of view, LTUs are often nominal categories whereas LPUs may include different categories (verbs, adjectives, prepositions, etc.). From a semantic point of view, LTUs are lexical units having a denominative and referential character, whereas LPUs are combinations of words having a relational character. As pointed out by Bevilacqua (2004, 28 in Pontrandolfo 2013, 93–94), LTUs denotes a node of the conceptual structure of a specialized domain whereas LPUs are generally made of a terminological nucleus (simple or multiple LTUs) combined with verbs, nouns, adjectives, etc., which represent the activities and processes typical of a specific domain. An example – partially based on Kjær (1990a, 18–19) – will clarify this distinction: bill of exchange is a legal terminological unit (a multi-word term) because it has a specific definition in law22.See its definition in the Black’s Law Dictionary: https://​thelawdictionary​.org​/bill​-of​-exchange​/#:~:text​=An%20open%20(that%20is%2C%20unsealed,or%20to%20the%20drawer%20himself. whereas to accept a bill is a legal phraseological unit (more precisely, a collocation made by a verb + a noun) which has a specific meaning in the legal domain even though the meaning of the verb accept can be paraphrased by other verbs such as take on or assume. Kjær stresses the fact that the word combination forms a lexically stable unit which can be broken only within lexicalized limits (1990b, 24) and these restrictions on combinability also depend on the user of legal language (1990b, 26). Consequently, in some cases the wording is prescribed by law whereas in other the author (legal expert) may alter the wording by using synonyms without affecting the meaning of the legal text.

The permeability of the distinction between legal terminology versus legal phraseology is also demonstrated by the role played by phraseology as “terminology in the making” (see Pavel 1993, 21–34): some phraseological units become terminological units as in the case of the Italian verbs collocating with prova (evidence), i.e., acquisizione/assunzione (gathering of evidence) versus ammissione (admission of evidence) (see Scarpa, Peruzzo, and Pontrandolfo 2017, 90).

Defining legal phraseology also means classifying it; as a matter of fact, most definitions found in the literature are based on the identification of the types of legal phrasemes which constitute the essence of their conceptualization.

3.Classifying legal phraseology

Attempts to provide all-inclusive classifications of legal phraseology have been made in the last 30 years (for an overview, see Pontrandolfo 2013, 100–104; Biel 2014b, 34–36; Goźdź-Roszkowski and Pontrandolfo 2015; Ruusila and Lindroos 2016, 128–132; Nikitina 2017, 46–51). Even though general typologies may be useful while describing word combinations in any legal text, many classifications in the literature are based on ad-hoc categorizations which reflect specific research carried out by single scholars. One of the first and most remarkable general classifications of legal phraseology was proposed by Kjær (1990b, 26–27), who studied context-conditioned word combinations in legal language. Her classification envisages four types of combinations which are generally used in legal texts:

1. prefabricated word combination directly prescribed by law (failure to employ those word combinations in accordance with the legally prescribed formulation will result in the invalidation of the whole text of which they form part);

2. word combinations only indirectly prescribed by law (variation of these word combinations will not render the whole text of which they form a part invalid, but its legal force will be affected);

3. word combination, the use of which is recommended for reasons of unambiguity (if they are varied, this may affect the security of the law; these are word combinations based on implicit quotation);

4. routine phrases whose use is merely habitual (if they are varied, the writer of legal texts will not spend more time, but otherwise, a variation will have no effect whatsoever).

In her 2007 study, Kjær updated her statements and proposed a more general term-based typology (2007, 509–510):

1. multi-word terms, with the most productive pattern [Adjective + Noun];

   1. Latin multi-word terms, e.g., ex officio;

2. collocations with a term;

   1. LSP phrases (Fachwendungen): [Noun + Verb];

   2. Support Verb Construction (Funktionsverbgefge): [(Preposition) + Noun +Verb].

3. formulaic expressions and standard phrases, including

   1. binomials (“word phrase patterns consisting of two words belonging to the same word class, connected by a conjunction”);

   2. phrasemes with archaic words or word forms.

The second proposal emphasizes the importance of terms in legal phraseology. Word combinations in legal language tend to cluster around the key concepts (terms) of that domain; therefore, collocations tend to have a term as the node of the expression, as in the case of the above-mentioned example to accept a bill of exchange. Phraseology acts as a strong link between the term and the text.

As far as the ad-hoc classifications are concerned, scholars, mainly with linguistic and translation backgrounds, have relied on specific typologies of phraseological units to carry out their empirical studies. One of the first classifications is proposed by Goźdź-Roszkowski (2011, 109–142) based on Biber et al.’s (1999, 32–34) notion of lexical bundle, i.e., multi-word sequences that frequently occur statistically in a given language variety. Based on his American Law Corpus (ALC) (2011, 27–34), which includes academic journals, briefs, contracts, legislation, opinions, professional articles and textbooks, Goźdź-Roszkowski distinguishes among:

1. legal reference: temporal (e.g., at the time of), location (e.g., in the district court), attributive (e.g., the amount of the), participative (e.g., on behalf of), institutional (e.g., the Due Process Clause), terminological (e.g., a breach of contract), procedure-related bundles (e.g., shall be entitled to);

2. text-oriented: causative/resultative (e.g., by reason of the), condition (e.g., in the event of), clarification/topic elaboration (e.g., on the other hand), focus (e.g., at issue in this), framing (e.g., on the ground that), structuring (e.g., as provided in Section), transition bundles (e.g., in addition to the);

3. stance: epistemic and attitudinal bundles (e.g., it is necessary to consider). (Goźdź-Roszkowski 2011, 109–142)

Another classification, in line with Goźdź-Roszkowski’s one, which is based on the notion of lexical bundles is that of Breeze (2013, 234), who identifies three main types of phraseological units in four legal genres (academic law, case law, legislation, and legal documents):

1. stance expressions;

2. text-organizing expressions;

3. referential expressions.

It is interesting to observe that Breeze, based on Pecorari (2009), makes a distinction between content and non-content phrases, due to the fact that, as already shown, lexical bundles may refer to specific (legal) aspects of the content of texts (e.g., request for confidential treatment) or not (e.g., in terms of, the fact that). This distinction between content-related and non-content bundles proves to be essential in that it recognizes that legal texts are also characterized by non-legal word combinations (e.g., on the basis of vs. on the ground that / as described in paragraph vs. shall be entitled to). The main advantage of these two proposals is that they are based on a wide range of legal genres.

An interesting classification of phrasemes in the language of law, which is not based on a single type (lexical bundles as in the previous ones) but specifically designed for EU legislative texts, and used effectively also for other legal genres, is provided by Biel (2014a, 178–182). The author clearly indicates that this classification should be viewed within a phraseological continuum with fuzzy boundaries between each category, ranging from the global textual level to the local microlevel:

1. text-organizing patterns (e.g., the opening and closing sections of EU instruments);

2. grammatical patterns (e.g., shall, must, otherwise, provided that, in order to);

3. term-forming patterns (multi-word terms) (e.g., person acting in concert, cross-border merger of limited liability company);

4. term-embedding collocations (e.g., to hold shares, pro-rata issue of securities);

5. lexical collocations (e.g., subject to this Regulation, the costs […] shall be borne by).

Pontrandolfo’s (2016 based on Pontrandolfo 2013) classification is specifically designed for judicial texts and tested on the COSPE corpus. Based on Corpas Pastor’s (1996) LGP classification of phraseology, he focuses on four main patterns:

1. complex prepositions (e.g., without prejudice to, in pursuance of);

2. lexical doublets and triplets (e.g., adequate and proper, noble and learned);

3. lexical collocations (e.g., to quash an appeal, credibility of the appellant);

4. routine formulae (e.g., for these reasons, I would allow the appeal and quash the appellant’s conviction, I agree that the appeal should be allowed).

Finally, Nikitina (2017, 99–100), based on a corpus of pleadings before the European Court of Human Rights (ECtHR), proposes a classification of multi-word units in written pleadings which takes into account the other classifications proposed by the above-mentioned scholars:

1. formulaic units

   1. binomials/multinomials (e.g., rights and freedoms, legality and reasonableness)

   2. archaic words or word forms (e.g., herein, thereof, aforesaid, henceforth)

   3. routine formulae (e.g., on the grounds of aforesaid, I request ECtHR to)

2. term-related units;

   1. multi-word terms (e.g., Convention rights, Human rights)

   2. collocation with a term (e.g., to submit observations, the Court upholds)

3. grammatical units

   1. modal auxiliaries (e.g., shall, must, should, may, if-then, with a view to -ing)

   2. complex prepositions (e.g., by virtue of, owing to)

The list of classifications presented in this section is obviously not exhaustive (see also Table 3) although it is representative of the main categories identified in the literature. As specified by many authors, these categories should be placed along a phraseological continuum and read with a certain degree of flexibility and fuzziness (Granger and Paquot 2008, 29).

4.Old perspectives

The dawn of legal phraseology studies is characterized by a stable or traditional view of phraseological units. The very first studies specifically devoted to legal language dealt with the so-called binomial or multinomial expressions (also referred to as doublets/triplets) (Bhatia 1984, 1993; Child 1992; Crystal and Davy 1969; Gustafsson 1984; Maley 1987, 1994; Malkiel 1959; Mellinkoff 1963, 1982; Thorntorn 1987). One of the first sub-classifications of binomials is that of Malkiel (1959), who distinguishes four types of doublets:

1. near synonyms (e.g., null and void);

2. complementary (e.g., assault and battery);

3. opposite (e.g., assets and liabilities);

4. subdivision (e.g., months and years);

5. consequence (e.g., shot and killed).

In 1963, Mellinkoff tried to systematize the use and function of these expressions in legislative texts, defined as a sequence of two or more words or phrases belonging to the same grammatical category having some semantic relationship and joined by some syntactic device such as and or or (Bhatia 1984, 90 in 1993, 108) (e.g., signed and delivered, in whole or in part, to affirm or set aside, act or omission, advice and consent, by or on behalf of, under or in accordance with, unless and until, consists or includes, wholly and exclusively, the freehold conveyed or long lease granted). As far as their function is concerned, Mellinkoff (1963, 349) distinguishes between: worthless doubling (e.g., force and effect) and useful binominals (e.g., full faith and credit). Gustafsson (1984, 134) adds a further categorization: synonymous (e.g., last will and testament); antonymous (e.g., be present in person or by proxy); complementary (e.g., shoot and kill).

Defined as “mannerism” and therefore style markers in legal language, their purpose was initially a laudable one: to increase clarity, accuracy and unambiguity. However, the (ab)use that has been made of these patterns has resulted in a style which is, as Mellinkoff puts it, “wordy, unclear, pompous and dull” (1963, 24); this is the reason why these doublets have been stigmatized as unnecessary by plain language experts, who believe that these expressions clash with the pragmatic and economy principle in language, thus affecting the linguistic quality of the final texts33.“The use of the term and/or is pervasive in legal language. Lawyers use it in all types of legal contexts – including statutes, contracts, and pleadings. Beginning in the 1930s, however, many judges decided that the term and/or should never be used in legal drafting. Ardent attacks on the term included charges that it was vague, if not meaningless, with some authorities declaring it to be a “Janus-faced verbal monstrosity,” an “inexcusable barbarism”, a “mongrel expression,” an “abominable invention,” a “crutch of sloppy thinkers,” and “senseless jargon.” Still today, critics maintain that the construct and/or is inherently ambiguous and should be avoided whenever possible–which, many detractors would argue, is always” (Robbins 2018, 311). (see, among others, the indications of the CMLJ in Spain, 2011, 155). On the other hand, there are authors (see, among others, Crystal and Davy 1969, 202; Mattila 2006, 232–234; Tiersma 1999, 59–61) who are in favour of the use of these formulaic expressions, since they are the result of the historical tradition of legal language, characterized by a ritual, almost magic, language (see also Borja Albi 2000, 54). In a similar vein, Bhatia attributes a specific function to these multinomial units, namely, precision and inclusion (1993, 108) (see also Robbins 2018); this all-inclusiveness has been also interpreted as a need to preserve vagueness in legislative (and in general legal) texts (see, among others, Mattila 2006, 234)44.As indicated by Borja Albi (2000: 56), the use of these binomials/multinominals could also have etymological purposes. Garner (1987) mentioned that the use of these synonymic expressions had etymological reasons since a term deriving from Latin or French could be accompanied by an equivalent Anglo-Saxon term (e.g., acknowledge and confess, old English and French; act and deed, Latin/French and old English; etc.). (see also Section 7 below and Biel in this volume).

Another category of legal phraseological unit, closely related to multinominal units, which gained attention in the early studies on legal phraseology were the so-called fixed, routine or stereotyped formulae. This type of phraseological unit differs from both collocations and idioms (“composites”) for the kinds of meaning they convey and the structural levels at which they operate (Cowie 1994, 3170). Formulae evolve meanings which largely reflect the way they are used in discourse. As a matter of fact, legal phraseology was initially studied in terms of formulaicity, traditionally regarded as one of the most typical and conspicuous features of legal style (Crystal and Davy 1969, 194). Formulaic expressions have been found to lie at the core of much of the formal and ritualistic language so ubiquitous in legal proceedings and documents (Tiersma 1999, 100–104). Exactly as in the case of multinominal units, on the one hand, these expressions have been perceived as an obstacle to the understanding of professional-lay communication but, on the other hand, they have been recognized as vital signposts helping interactants to orient themselves at different stages in the course of legal proceedings.

Many studies, especially of German tradition, started to investigate recurrent lexical sequences, of different length, which appear to be prefabricated. Stolze defines these units Standardformeln/standard formulae (1992, 190), habitualisierte juristische Formeln/habitual legal formulae (1999a, 176), standardisierte Formeln/standardized formulae (1999b, 56), a term also used by Sandrini (1996, 256) and Koutsivitis (formules standardisées/standardized formulae 1991). The use of these formulae is highly standardized, which is why, for specific types of legal genres, such as contracts, these expressions are usually collected in templates. An example would be the boilerplate clauses (see also Tiersma 1999, 59) which demonstrate that legal language clusters may be very long, getting to entire parts of the documents (e.g., This contract acknowledges that […] and […] agree to the following conditions, to be considered in effect after […]. […] agrees to the following: […]. In return, […] agrees to […]. The provisions of this agreement are as follows: […]. This agreement is amenable to both parties and can only be altered with the consent and signature of both parties. […] Signature. Date).55. https://​www​.contract​-template​.org​/boilerplate​-contract​.html Another example is the ritual procedure of giving evidence in court (see also Giurizzato 2008). When called to give evidence, people in the UK (but also in many other countries) are asked if they wish to take an oath or make an affirmation that the evidence is true. These types of oaths or affirmations used in court can also be considered as standardized (and highly recognizable) formulae (e.g., the oath: I swear by […] (according to religious belief) that the evidence I shall give shall be the truth the whole truth and nothing but the truth or the affirmation: I do solemnly, sincerely and truly declare and affirm that the evidence I shall give shall be the truth the whole truth and nothing but the truth.).66. https://​www​.nidirect​.gov​.uk​/articles​/giving​-evidence​-court There are many other examples of routine formulae adopted in various fields of law, such as those used in EU legal documents, which make the texts highly predictable and repeatable (see, among others, Nystedt 2000).

Another strand of research which can be included in the old perspectives under analysis is the area of lexical collocations in legal texts.77.The use of the term “lexical collocation” here slightly varies from Biel’s (2014a), who refers to these types of units as “term-embedding collocations” (collocates of terms which embed terms in cognitive scripts and the text, evidencing combinatory properties of terms). Instead, the author defines “lexical collocations” as “routine formulae at the microstructural level which are not built around terms” (2014a, 180–181). Defined as “associations of two or more lexemes (or roots) recognized in and defined by their occurrence in a specific range of grammatical constructions” (Cowie 1994, 3169), they reflect the strong tendency in language use to reuse certain patterns, which make them particularly useful in legal language. Following the types of lexical collocations categorized in the BBI Combinatory Dictionary of English (Benson, Benson, and Ilson 1986), many scholars decided to investigate the recurrent use of lexical collocations in the language of law. Among these authors a mention could be made to the following studies (based on Pontrandolfo 2013, 151–164): Berdychowska (1999), who studied lexical collocations in the German Civil Code; Lombardi (2004), who carried out a corpus-based study on lexical collocations in normative and interpretative legal texts in German and Italian; Cruz Martínez (2002), who studied the lexical and grammatical collocations used in criminal proceedings (e.g., to appear before a court, to hear a case, to charge somebody with an offence, to indict somebody for, to be tried for, etc.); Bhatia et al. (2004), who analysed the lexical collocation noun + verb in a corpus of case law starting from key verbs, such as to submit, to grant, to reject, to dismiss; Giráldez Ceballos-Escalera (2009), who studied lexical collocations in French and Spanish civil texts; Fernández Bello (2008), who extracted a wide range of collocations typical of the judicial style from a corpus of criminal and civil judgments delivered by the Spanish Supreme Court.

These three areas of phraseological interest, which characterized the early studies in the field of legal phraseology, confirmed (also empirically) the highly formulaic nature of legal language. However, the scope of these studies was heavily limited: phraseology was mainly studied monolingually as a lexical environment of terms (it had an ancillary function compared to terms) identified mostly manually within a very narrow range of legal genres (legislative texts being the preferred genre). These first studies also demonstrated the potential for relying on repetition, fixedness and frequency to uncover patterns of lexical combinations which may be otherwise difficult to intuit. The advent of Corpus Linguistics has radically changed the landscape of contemporary legal phraseology, resulting in an array of different approaches and perspectives and giving fresh impetus for the study of this area of language.

5.New perspectives

The rise of legal phraseology studies is commonly associated with the advent of corpus linguistics. As indicated by Buerki (2018, 17), with the availability of large corpora and the large-scale corpus-linguistic exploration of phraseological phenomena, it has become clear that idiomatic formulaic sequences are vastly outnumbered by conventional, non-idiomatic sequences that should nevertheless be considered as such, and this also applies to legal phraseology. The frequency-based approach to phraseology (Nesselhauf 2004 in Granger and Paquot 2008, 29), made possible by corpus-driven insights in the identification of lexical co-occurrence, allows one to enlarge the boundaries of the traditional phraseological unit (generally identified on the basis of linguistic criteria by means of a top-down approach), thus including a wide range of word combinations that do not all fit predetermined linguistic categories (Granger and Paquot 2008, 29).

Corpora of legal texts have been increasingly compiled over the last decade (for an overview, see Biel 2018a; Marín Pérez and Rea Rizzo 2012; Pontrandolfo 2012; Pontrandolfo 2019; Vogel, Hamann, and Gauer 2017),88.As pointed out by Vigier and Sánchez (2017, 261), despite the widespread use in other fields within Translation Studies, the development of corpora has been rather slower in the field of Legal Translation, most probably due to the confidential and private nature of many legal documents. This is what Biel calls ‘legicentrism’, that is to say the tendency of existing corpora to be composed of legislation mainly (2018a, 29), which results in an underrepresentation of other genres. even though not all of them have been specifically built to study legal phraseology. Table 3 offers a non-exhaustive overview of a number of corpus-based studies specifically devoted to the exploration and extraction of legal phraseological units.

As shown in Table 3, most of the studies that fall within the new perspective are based on distributional categories (Granger and Paquot 2018, 38–41). The legal phraseological units under investigation in these studies are categorized on the basis of the different extraction procedures and software adopted and not, as was the case in the old perspective, on pre-determined (often fine-grained) linguistic categories. The result is a wide range of word combinations, e.g., lexical bundles, n-grams, clusters, chains, recurrent word combinations, that rely on quantitative and statistical thresholds more than on strict categories, which responds to the inevitable fuzziness of some combinations.

Large corpora of legal texts become the testbed for quantitative evidence, which is proving to be of immense value to the field of legal phraseology. The move from the traditional to the distributional/corpus-based approach to legal phraseology has enlarged and deepened the analysis of legal genres. Semantic sequences (Mazzi 2010; Goźdź-Roszkowski and Pontrandolfo 2013; Goźdź-Roszkowski 2018), for example, have enabled researchers to investigate the role of phraseology in legal discourse in terms of explicit or implicit positioning strategies of legal interactants in the text. These recurring patterns become powerful means to isolate argumentative and evaluative devices, thus demonstrating that phraseological research involves examining the organization of language also beyond the level of a sentence or a clause towards larger linguistic units. By way of example, by means of a corpus-based analysis of US Supreme Court judgments, Mazzi (2010) identifies a number of reiterated discourse strategies indicating the judge’s stance in text: verbs, adjectives, and most interestingly the whole pattern “this/these/that/those + labelling noun”, a pattern which would have not been considered as phraseological in the old approach to legal phraseology. These studies have confirmed the key (also semantic) role played by phraseology in legal discourse, well beyond the mere question of style or standardization. This is probably one of the most vibrant strands of legal phraseology research with a host of possible applications, some of which include the automatic extraction of n-grams, which become a powerful vehicle to explore the evolution of legal language.

6.Research perspectives and applications

Legal phraseology can be studied from different perspectives and adopting different methods. The studies mentioned in Sections 4 and 5 adopt a wide range of methods and tools, often combining them to enhance the empirical results.

One of the basic methods is that of comparing LGP versus LSP (legal language) in order to isolate those patterns which are overused in legal discourse and can therefore be considered as typical of legal texts. Corpora can help researchers carry out this type of investigation semi-automatically. Another interesting comparison can consist in the analysis of variation of legal phraseology across legal genres or within a specific genre. Some of the studies mentioned in Table 3 also adopt a contrastive perspective (e.g., Spanish vs. German or English vs. Italian) and some of them are based on comparable corpora (original texts) whereas others rely on parallel ones (original vs. translated texts). Legal phraseology is, as a matter of fact, one of the challenges of legal translation, which is why many studies adopt a contrastive and translation perspective.

Research into legal phraseology can be fostered by different research interests. The applications and usefulness of this area of research can be summarized in (at least) six main groups:

1. Contrastive linguistics

2. Legal translation and training of legal translators

3. Lexicography and Terminography

4. Electronic management / (semi)-automatic search of legal phrasemes

5. Corpus-assisted drafting of legal documents

6. Plain legal language (simplification of legalese)

As far as contrastive linguistics is concerned, research into the regularities of legal discourse in different languages may allow for an investigation of the legal culture behind languages. Pontrandolfo’s (2013, 2016) investigation of legal phraseological units in three different legal cultures (England and Wales, Spain and Italy) shows the generic and discursive construction of criminal judgments in the three legal settings, connecting linguistic and stylistic observations to the conceptual underpinnings of the judicial reasoning. The study has specific applications for translation purposes. Another example of translation-oriented contrastive study which straddles the two areas of applications is López-Arroyo and Moreno Pérez’s (2019) corpus-based study of contractual documents. Using a comparable corpus of English and Spanish sales contracts, the authors focus on recurrent phraseological patterns, classifying them according to form and meaning. The results of the study are useful for legal translators and drafters, especially because phraseology is a quality-enhancing factor in legal translation. Familiarizing oneself with the routines of the genre, as well as mastering their use (both at receptive and productive level) are crucial factors in legal translators’ training (Garzone 2007). As a matter of fact, phraseology is one of the discursive elements which mostly contribute to the naturalness of the translated text.

Another application of legal phraseology research is the area of lexicography and terminography. As demonstrated in different studies (see, among others, Buendía Castro and Faber 2015), phraseology is generally absent in legal dictionaries, which focus almost exclusively on specialized terms, most of which are multi-word terms (term phrasemes). Corpus-based studies can help improve the quality of these resources by providing them with important units which play a pivotal role in legal genres.

The extraction of legal phraseological units and especially fixed, highly repetitive patterns may also be used to populate online specialized databases. This is the case of TermWise (Heylen et al. 2014), which aims to leverage online translation data for terminological support to legal translators at the Belgian Federal Ministry of Justice. By means of a complementary database, Term&Phrase Memory, it integrates with existing Computer-Assisted Translation tools. The repetitiveness and patterning of certain legal genres may allow for the design of platforms that help legal professionals to draft their texts. Corpus-assisted drafting of legal documents is one of the practical applications of legal phraseology extraction. A recent example is Da Cunha’s ArText99. http://​iriadacunha​.com​/FundacionBBVA2015​/es​/acceso​-al​-sistema/ application which exploits routine formulae and standardized sentences in the machine-assisted drafting of administrative documents.

Finally, legal phrasemes may also be the target of simplification; redundant expressions are often criticized for hindering the comprehension of legal texts. Corpus-based studies of legal phraseology can help scholars identify worthless versus useful patterns (see Mellinkoff 1963, 349).

7.Final remarks

The words of law naturally tend to pattern and such patterning is crucial in the construction of legal texts. Phraseology weaves the intricate webs of law; as indicated by Biel (2014b, 36–37), legal patterns may help the reader navigate through the text, thus easing the progress of discourse. Moreover, prescribed formulaicity institutionalizes the discourse by limiting drafters’ creativity and idiosyncrasy with the final result of having a text which is organized both at macro and microlevel. However, this patterning has also some disadvantages. Table 4 summarized contrastive views on phraseology in legal texts.

The arguments for and against legal phraseology in legal texts confirms one of the first assumptions made in this chapter: the need to adopt a flexible view on a slippery discursive phenomenon. When analysing the old and new perspectives or the methods and applications of legal phraseology research, it is fundamental to avoid clear-cut classifications in favour of eclectic perspectives which recognize the fuzzy nature of these units. A wide range of methodological perspectives on legal phraseology enable scholars to obtain different and complementary results. The traditional and the new approaches can be reconciled, as several current studies empirically demonstrate and will continue to demonstrate in the future.