Issues and Perspectives in Bilingual Phraseography

In this article, we consider the problems normally associated with bilingual phraseography and the paradigm that should be used to guide its development. We first present the fundamental aspects that should be taken into account in the characterization of bilingual phraseography. Secondly, we give a general overview of both theoretical and practical bilingual phraseography, paying special attention to the main problems existing in this area. Thirdly, we consider the factors that can determine the development of bilingual phraseography, starting from the notion of the “user needs paradigm” and the contributions from the Function Theory of Lexicography. Finally, we make a brief summary of the main ideas dealt with in the article.


Introductory questions
From a traditional point of view, bilingual phraseography can be understood as a sub-area of phraseography1 which includes all the phraseographical work regarding general bilingual dictionaries and bilingual phraseological dictionaries.As a specific branch of phraseography, it should include, in the first place, a metaphraseographical perspective (of historiography, theory, criticism and research).Furthermore, it should contain a practical aspect focusing on the compilation of bilingual phraseological dictionaries (BPD) (including technical and methodological aspects), as well as everything pertaining to the treatment of phraseological elements in any type of bilingual lexical repertoire.It is, however, necessary to bring some clarifications to this characterization: firstly, it must be remembered that the different facets have not developed uniformly.Consequently, there are numerous descriptive works, whereas those of a metaphraseographical nature are few and far between, above all those referring to the use of the dictionary and those based on empirical studies.
Secondly, it is essential to highlight the vagueness of the term "bilingual", since this can lead to the exclusion of "multilingual phraseography".However, this latter field has aroused considerable interest and has been the subject of study in a series of recent articles (Berthemet 2009;2012;Dawes 2014), for which reason it must not be left out of the field of phraseography.
Thirdly, special attention must be paid to the aspect of interlinguistic equivalence, which is considered to be the cornerstone of bilingual lexicography, and, consequently, of bilingual phraseography.As we will see, this aspect is especially important due to the complexity of phraseological units.It is also a forceful reminder of the need to bear in mind the notion of "interlingual lexicography" developed by Hartmann (2007)2, when defining and characterizing bilingual phraseography.In addition, the possibility of developing an "interlingual phraseography" could be contemplated, as a logical extension of interlingual lexicography.Following from these notions, it is possible to deal with the questions that arise in any phraseographical work involving more than one language.These questions are of a mainly cultural nature.
The second aim of this paper is to reflect on future advances in bilingual phraseography.We support the idea that, in order to be consolidated as a field of linguistics and to guarantee its own development, phraseography (both monolingual and bilingual) should move closer to the so-called user needs paradigm, and, thus, to a functional approach to lexicography, based on the Function Theory of Lexicography (Bergenholtz and Tarp 2003, 2010: Fuertes-Olivera and Tarp, 2011;Nielson, 2013).In the following sections, we will examine these questions in detail.

Theoretical and practical bilingual phraseography: problematic questions
In this section, we present a general overview of the most important problems affecting bilingual phraseography.In order to provide this overview, we have borne in mind different sources related to three broad areas: lexicography, phraseography itself and phraseology.Within these areas, the subject is mainly dealt with in works centred on the teaching and translation of phraseological units; on the compilation and revision of BPDs or bilingual dictionaries (BD); and on pedagogical lexicography.To a lesser extent, there are also relevant references in publications about translation in general.Beyond the features that these works have in common, a notable difference of opinions can be observed, probably as a consequence of the different standpoints taken by each author.Thus, there is a huge range of opinions regarding the types of phraseological units that should appear in dictionaries, or the information that should be provided in addition to the equivalents.We will now examine the main criticisms directed at bilingual phraseography.
Theoretical and practical studies from different fields have highlighted the failings and deficiencies that affect the treatment given to phraseological units in BDs and BPDs.Granger & Lefer (2012) consider that this treatment has progressed in some monolingual dictionaries, above all in learners' dictionaries (MLDs); however, this is not true of bilingual dictionaries.As these authors explain: Bilingual dictionaries, however, are lagging behind in this field: "[t]he extraordinary range of lexical and grammatical information they [MLDs] include is rarely even approached by the best bilingual dictionaries available" (Rundell, 1999).Although the last decade has seen some progress, as testified by the work reported in several studies (cf. Ferraresi et al. 2010, Murano 2010, Todirascu et al. 2007), bilingual dictionaries still have a long way to go before they can match the phraseological coverage of MLDs.More generally, though, Sinclair's (2008: 408) observation that dictionaries are still "almost exclusively focused on the word, so much so that it is difficult to treat a phrase adequately in a dictionary and difficult to find it once it is there" remains true today, whatever the type of dictionary.(Granger and Lefer 2012: 682) As we can see, there is a prevailing sense that phraseology does not receive the attention it deserves in lexicography in general and in phraseography in particular.In fact, everyone seems to agree that phraseological units are not treated correctly and homogeneously in general bilingual dictionaries (Wotjak 1983;Roberts 1996Roberts , 1998;;Corpas Pastor 1996;Nunes 1998;Santamaría Pérez 1998;Sevilla Muñoz 2000;Sánchez Merino 2002-2003;Navarro 2005;Svensén 2009;Fragapane 2011;Valero Gisbert 2012;Mulhall 2014;Penadés Martínez 2015a) or in BPDs (Molina Plaza 2004;Alonso-Ramos 2015).As a result, in most of the works on this subject, negative opinions prevail and there are few positive opinions of any dictionary (and these are almost always accompanied by some other, less favourable comment, as in Kubarth (2014).To summarize, the current situation of bilingual phraseography is not encouraging.Although advances have been made in certain aspects, there continue to be numerous problems, some similar to those affecting monolingual phraseography.
Therefore, there is still a demand for the inclusion of more information about the treatment given to phraseological units in the introductions of dictionaries.The selection criteria used to choose phraseological units as opposed to other types of lexical units (in the case of BDs) are brought into question, as is the choice of one specific type of unit over another (when referring to BPDs).It is claimed that the samples of phraseological units are not sufficiently varied, and, as a result, some types of units, such as collocations or formulae, are under-represented.Moreover, the choice of the phraseological repertoire concerns not only the type of units, but also other criteria such as: their frequency of use, their current relevance and their linguistic variety.Once again, there appears to be no consensus in this case regarding the application of these criteria, whose reliability largely depends on the progress made in phraseological studies such as the development of data bases enabling researchers to draw conclusions regarding the sphere and frequency of use of a unit.The end result is that dictionaries are not always seen to follow the promised selection criteria.This means that, in a work in which one expects to find frequent, up-to-date units belonging to the standard norm, what one actually finds is vulgar or outdated units, to the detriment of well-known, more usual ones.
With regard to selection criteria, the phraseological status of certain units included in dictionaries is also discussed.This is another critical aspect relevant not only to phraseography but also to phraseology.At present, the limits separating a narrow concept of phraseology from a broad one are not stable and there is a noticeable tendency to broaden the scope of the notion of phraseological unit3.
Another object of severe criticism is the lemmatization of phraseological units.The tendency is to express reservations about the criteria used, since it has proved impossible to give an exact definition of what can be considered the most appropriate criterion4.Furthermore, the lack of systemization in the 3 This question of the widening of the limits of phraseology is linked to the use of linguistic corpora in phraseological and phraseographical research.For this reason, we re-examine it briefly in section 3, in which we consider the relevance of working with corpora for bilingual phraseography. 4 There are several proposals regarding lemmatization according to the criteria considered.In general, in phraseographical praxis the following criteria are applied: the structural criterion (based on the organizative structure of phraseological units, that is, the way in which their constituting elements are presented); the semantic criterion (based on the semantic peculiarities of the phraseological units, that is, their idiomaticity) and the categorial criterion (based on the category or type of words that make up the unit).Some of these criteria may prevail over others, but in some cases they may be used together (Olimpio de Oliveira 2007).application of these criteria is criticized, since the principles of lemmatization are not homogeneous in most of the works mentioned, bar a few exceptions.Likewise, there is criticism of the diversity of procedures adopted in registering the stem of the phraseological unit and in indicating alternative or optional elements that appear in the stem of certain units.
There are also negative opinions of the fact that phraseological units are occasionally included in BDs as examples of the use of the lemma but are not identified.In the same way, the absence of a distinction between different types of units, both in BDs and in BPDs, is also criticized.And, in works in which different types of phraseological units are separated, the way in which this is done is questioned: where devices such as orthographic signs are employed, their functions are not always made clear to users.Another negative aspect is the lack of standardized procedures in the positioning of the phraseological unit in the article, this can lead to a completely random distribution which is confusing to users.
Likewise, exemplification has been the target of several analyses, from which conflicting positions have emerged.Although there is consensus on the need to use examples illustrating the authentic uses of a phraseological unit, the lack of which information has been criticized, there is no consensus regarding whether these examples should be created ad hoc or whether they should be authentic, taken from corpora.And if they were taken from corpora, there is no consensus as to what extent they could be adapted without losing their originality.
The virtual non-existence of pragmatic, sociolinguistic or dialectal data on phraseological units, both in BDs and BPDs, has also been pointed out by several authors5.This question is closely related to the treatment given to equivalents.The absence of this type of data, together with the chaotic and decontextualized manner in which these elements are usually presented (Schmitz 1998: 163), is one of the reasons for which the contextual conditions governing the use of phraseological units are not made explicit.As a consequence of this, BDs and BPDs do not fulfil the one of the main functions for which they are conceived, the communicative function, since they are of no use in translating, understanding or writing texts.
However, one of the aspects that has perhaps been the object of the most scathing comments and which has caused most arguments is phraseological equivalence.Reflections on this subject are closely linked to two fundamental perspectives: studies regarding the translation of phraseological units (Roberts 1998;Zuluaga 2001;Corpas Pastor 2000 and2003;Torrent-Lenzen 2012) and, related to these studies, research on the contrastive analysis of phraseological elements (Dobrovol'skij 1998, 1999, 2000aand 2000b, Corpas Pastor 2003).As a result of the adoption of these standpoints, certain notions are applied in the area of phraseography, such as full equivalence, partial equivalence, and zero equivalence, in order to judge the level of adequacy of the equivalents offered.
An analysis of the aforementioned studies reveals that many of the problems attributed to BDs and BPDs spring from the fact that they occasionally offer incorrect equivalents: they do not clarify the correspondence between the equivalents and the acceptions in the source language, and, as a result, they do not offer an adequate separation of the proposed equivalents, either graphically or in the lexico-semantic aspect.That is to say, the equivalents are listed one after the other as if they were interchangeable.Thus, the practice adopted in BDs of offering an indiscriminate list of equivalents is criticised because it is considered not only to be insufficient, but also incorrect and confusing: the user is unable to choose a correct equivalent if there is no explanation of the semantic changes that can take place when a unit is combined with one element as opposed to another.6 As occurs in the case of other phraseographical aspects mentioned, equivalence is dealt with from a largely descriptive point of view.Thus, in many works, authors have focused on the difficulties inherent to establishing equivalents and have set out to highlight the limitations of the suggested equivalents, showing that those lack validity both at lexical and textual levels.In other words, they have highlighted the fact that the equivalents given in BDs and BPDs do not function as such when a user tries to put them into practice.Furthermore, the lexicographical treatment of the notion of equivalence based on corpora does not solve this problem since "Not infrequently, the generally accepted equivalent of an idiom cannot always be used to translate authentic texts" (Dobrovol'skij 2014a: 340).
As well as the questions mentioned above, we believe it is necessary to consider certain contributions from authors of phraseographical works, such as Torrent-Lenzen (2008); Iglesias Iglesias (2010) or Mellado Blanco (2015) who, from their own practical experience, try to throw some light on the problems of equivalence.Among the most outstanding contributions from this group of authors is the notion that the equivalent of a phraseological unit may not always be another phraseological unit.(Torrent-Lenzen 2008 -this author calls them "equivalentes desfraseologizados [dephraseologized equivalents]"; Mellado Blanco 2015).This means that for a given phraseological unit, a simple lexical unit or a free combination of words (dephraseologized equivalent)7 can function as equally valid and adequate equivalents.
Furthermore, another important contribution is the idea that explanatory paraphrases (Torrent-Lenzen 2008; Iglesias Iglesias 2010) can play an important part in the process of establishing equivalents.However, it is important to make clear that these paraphrases serve more as a type of definition than as a translation strategy.Besides, they have certain limitations and present problems of composition, since they require a high level of semantico-pragmatic analysis and the use of precise terminology.
Also related to the notion of explanatory paraphrasing, there is another significant contribution to this field: the acknowledgement of the fact that equivalents can have a defining value.This is the standpoint of Torrent-Lenzen (2008: 1433), who introduces the concept of "equivalencia definidora" ["defining equivalence"] through which she shows the complexity of the notion of equivalence.This author considers that: (…) "equivalente y definición no son sistemas completamente distintos y que las fronteras son difusas, aunque entre los extremos que suponen el equivalente fraseológico por un lado y la paráfrasis por otro haya una diferencia considerable" ["equivalent and definition are not completely different systems and their borders are blurred, although there is a considerable difference between the two extremes, that is to say, between the phraseological equivalent at one end and the paraphrase at the other"] Torrent- Lenzen (2008Lenzen ( : 1436)).
Likewise, consideration has been given to the way in which phraseological units signify and their relationship with the establishment of equivalents.Different authors (Torrent-Lenzen 2008; Parina 2014) point out that phraseological units have a meaning which is complex and potential: that is, they do not denote directly and precisely, and consequently, even a native speaker may have difficulties understanding their communicative value.Hence, according to Ayupova and Zamaletdinov (2014: 323): "On the whole, a PU [phraseological unit] is a product of an inductive activity of the mind.As such, revealing its meaning in the dictionary is a deductive process and the most important part is the semantic analysis".
In addition, with regard to the relationship between exemplification and equivalence (Iglesias Iglesias 2010), it has been mentioned that a different phraseological equivalent can be given for each example of the use of a phraseological unit (Torrent-Lenzen), which can mean that "an idiom in an authentic context cannot be translated with its traditionally accepted equivalent" (Parina 2014: 362).It has also been shown that it is necessary to indicate the restrictions of usage that may exist for the suggested equivalents (Sánchez Merino 2002-2003;Iglesias Iglesias 2010;Parina 2014).In this regard, it has been said that: "Some of these usage restrictions and peculiarities are perhaps familiar to native speakers but [may] not be obvious for dictionary users, for whom L1 or L2 is a foreign language" (Parina 2014: 373).
Similarly, the idea that BDs can offer full equivalents has been refuted.Studies based on corpora make it clear that many of the equivalents offered for a certain phraseological unit by a bilingual dictionary, although presented as full equivalents, are no such thing.This means that, from a semantic and discursive standpoint, the differences between a phraseological unit in its source language and its equivalent are not random.Instead, they reveal inherent characteristics of this unit.Nonetheless, it is admitted that there is one issue still to be resolved: whether these divergences should be made explicit in the dictionary and whether users are able to understand and interpret them correctly.
Mellado Blanco (2015) has made a fundamental contribution to the debate on the concept of equivalence, in a work in which she discusses which parameters should be considered when determining the interlinguistic equivalents of phraseological units.This author has underlined the need to distinguish between the types of equivalence according to which level is being considered.Thus, it is possible to take into account a systemic level (referring to the language it can also be studied from a contrastive perspective), a textual level (related to the text, which links it to a traductological standpoint), and a lexicographical level (related to dictionaries).With regard to this latter level, Mellado Blanco states that it is important to pay attention to the notion of functional equivalence, and following from it, to the textual behaviour of phraseological units.However, although functional equivalence affects both textual and lexicographical levels, the author warns that it must be examined differently depending on the level being dealt with: (…) la equivalencia lexicográfica se diferencia de la equivalencia textual en que no pretende hacerse eco de toda la casuística posible de traducciones de un fraseologismo de la L1 en la L2.Además de ello, en los diccionarios sería aconsejable que junto a las equivalencias en la L2 se explicaran las posibles restricciones de uso, cuestiones combinatorias o divergencias en la estructura del significado de los respectivos fraseologismos en la L2 [(…) lexicographical equivalence is different from textual equivalence because it does not aim to highlight all the possible casuistics of translations of a phraseologism from L1 into L2.In addition, in dictionaries it would be advisable to explain, along with the equivalences in the L2, the possible restrictions of use, combinatory issues or divergences in the structure of the meaning of the respective phraseologisms in the L2].(Mellado Blanco 2015: 155) Another reason supporting the differentiation between lexicographical and textual equivalence is that the solutions offered by a translator on translating a phraseological unit are not comparable to functional lexicographical equivalents, "ya que las traducciones presentan potencialmente una tipología muy variada que incluye desde la paráfrasis hasta la modificación creativa e incluso la eliminación del fraseologismo en la L2" ["since translations may present a very varied typology, ranging from paraphrase to creative modification and even the elimination of the phraseologism in L2"] (Mellado Blanco 2015: 155-156).
Thus, this author's contributions establish a necessary and unavoidable delimitation between the fields of traductology and lexicography (or phraseography).In fact, the absence of borders between the aforementioned levels (systemic, textual and lexicographical) has given rise to a mistaken concept of equivalence at lexicographical level which is reflected in the adoption of erroneous criteria to analyse the validity and appropriateness of the phraseological equivalents included in dictionaries or in the staunch defence of the notion of full equivalence.
Therefore, for Mellado Blanco (2015), of all the parameters of interlinguistic equivalence8, the most relevant for lexicographical (and phraseographical) tasks are the following: that of the scope and semantic structure of phraseological units (which includes, amongst other factors, the so-called "polyequivalence"); that of the connotative-pragmatic component (which encompasses numerous elements, including the different marks of use that are attributed to phraseological units or the frequency of use); and that of syntactic and semantic valency (the syntagmatic combinatorics inherent to each unit).The author argues that on a lexicographical level the parameters of the identity of lexical elements, of the coincidence of the morphosyntactical structure or of the image would be less important, since, in this respect, it is the functional-communicative aspect that takes precedence.She considers that excessive importance has been attributed to the parameter of the image, when, in fact, this is not of great significance.She reflects that: (…) el hablante −salvo en el caso de las modificaciones formales o semánticas, o en los juegos de palabras intencionados− no elige un fraseologismo determinado por la imagen concreta o su sentido literal, sino por su significado fraseológico, así como por el valor comunicativo y expresivo que posee.De esta manera, la coincidencia interlingüística de la imagen no parece que sea determinante a la hora de hallar un equivalente lexicográfico en la L2 (…).Lo importante es que se mantenga la expresividad, y esta se garantiza siempre que exista idiomaticidad, aunque se consiga mediante imágenes diferentes.[(…) the speaker -except in the case of formal or semantic modifications, or in intentional wordplays -does not choose a phraseologism determined by the specific image or its literal sense, but because of its phraseological meaning, as well as for its communicative and expressive value.In this way, the interlinguistic coincidence of image does not seem to be decisive in finding a lexicographical equivalent in the L2 (…).The important thing is to maintain expressivity, and this is guaranteed as long as there is idiomaticity, even if it is reached by means of different images].(Mellado Blanco 2015: 163) It is necessary to underline that this statement is based on extensive empirical evidence, since the author uses her own experience as a phraseographer, derived from the compilation of the BPD Idiomatik Deutsch-Spanisch (Schemann et al. 2013)9.
Upon examination of the previous contributions, it can be deduced that bilingual phraseography has some unresolved issues, since there are still many questions to be answered.Some of them arise from the inherent complexity of the subject of study, phraseological units, meaning that they are similar to those which have caused confrontations in monolingual phraseography (for example, the delimitation of the lemma, the determination of the valency, etc.)10; others are specific to bilingual phraseography, since they put into play elements that are inherent to the confrontation of two or more linguistic systems (remember the notions of anisomorphism (Zgusta 1971) and lacunarity11 (Szerszunowicz 2015).In this case, the problems and restrictions apply not only to equivalence but also to the structure of the dictionary as a whole, since the object is to deal with a set of cross-linguistic and cross-cultural gaps (which influences, for example, the decisions taken regarding the selection of units, the structuring of examples, etc.).
As we can see, these questions have as much to do with phraseology as they do with lexicography/ phraseography.Although in this paper we are not dealing with the former, it is worth remembering for a moment the old desire for a closer relationship between this discipline and phraseography (which would result in the inclusion in dictionaries of the achievements and advances made in phraseological research).With regard to the field of lexicography/phraseography, we suggest that the most important undertaking of bilingual phraseography should be to align itself with certain trends in lexicography, moving closer to 9 However, it is important to bear in mind that, in monolingual phraseography, Penadés Martínez (2013) has suggested the possibility that the underlying image of certain phraseological units -this author studies idioms-could be used as a criterion for diaphasic marking.She takes as a starting point Dobrovol'skij's contributions, from a paper of 2007, in which he states that the image conjured up by a certain phraseological unit not only connects with its meaning but can also impose some restrictions on its use.Penadés Martínez argues that some of these restrictions could have diaphasic value.The author explains: "Hay locuciones verbales cuya imagen se corresponde con acciones cotidianas que llevan a cabo los seres humanos o con situaciones que en algún momento de la vida se pueden experimentar o sufrir.Es el caso de caer de la burra, caerse de espaldas, caerse la baba, cagarse los pantalones, cortar el bacalao, (…) y un largo etcétera.La naturaleza cotidiana, habitual o usual de los sucesos designados literalmente por las locuciones puede servir de justificación para asignar a las que provocan imágenes relativas a tales sucesos la marca informal" ["There are verbal idioms whose image corresponds to daily actions carried out by human beings or to situations that can be experienced or suffered at some time in one's life.This is the case of caer de la burra, caerse de espaldas, caerse la baba, cagarse los pantalones, cortar el bacalao, (…) and many more.The everyday, habitual or usual nature of the events designated literally by the idioms may justify the practice of marking as informal those which suggest images related to these events"] (Penadés Martínez 2013: 29-30).According to this author, the aforementioned criterion seems to be more easily applied in the case of the informal mark (and, from what can be deduced, to a lesser extent in formal and vulgar marks).10 Many of these problems pertaining to monolingual phraseology have been pointed out by authors such as Olímpio de Oliveira (2007) and Penadés Martínez (2015).11 Lacunarity is a phenomenon which appears at different linguistic levels.It consists of the absence, in the target language, of certain elements from the source language.Given its complexity, it is the subject of study in the following disciplines: ethnopsychology, advertising research/market research, translation studies, film studies, foreign language teaching, lexicography and linguistics (Szerszunowicz 2015).Unauthenticated Download Date | 6/18/18 11:17 PM the user needs paradigm and, as a result, to a functional approach to lexicography.Thus, our position is that phraseography should incorporate these trends appropriately.Indeed, this would constitute the great challenge to be overcome by phraseographers and other specialists in this area.These questions, which constitute the second aim of this paper, are discussed below.

Perspectives for the development of bilingual phraseography: conditioning factors
In theoretical and practical lexicography, the field of "user research"12 is among the most complex and can be approached from various angles.Moreover, it is most probably the area that has received the least attention, both in monolingual and bilingual phraseography.However, it is a crucial issue for lexicography and should also be for phraseography.In fact, in lexicography, several voices have asserted its importance, meaning that in the past as in the present, this has been an important topic of discussion and occupies a primordial position in lexicographical advances.It is worth remembering Householder's (1967: 279) oftrepeated famous words: "dictionaries should be designed with a special set of users in mind and for their specific needs" (Householder 1967: 279).
It can be stated that there is a general consensus regarding the idea that dictionaries are products with a practical use which aim to satisfy certain needs of their possible users.Some theorists, such as Svenson ( 2009), go as far as defending the idea that user research should precede the compilation of dictionaries, with the intention of adapting them to the needs of the users for whom they are intended.
For all these reasons, it is by no means absurd to consider that user research should have received more attention in general, in both theoretical and practical phraseography.Furthermore, it can be considered that this gap in theory and research poses an obstacle that will prevent phraseography from existing in harmony with lexicography, bearing in mind the importance of the notion of user needs in the future (Nielsen 2013).In order to understand the importance of this notion for the development of phraseography, we outline below the way in which it has been dealt with within the General Theory of Lexicography and within the Function Theory of Lexicography.Finally, we suggest an application for this latter theory in bilingual phraseography.
Overall, within the General Theory of Lexicography (which encompasses differing paradigms, such as those represented by Scerba and Wiegand), the works on user research adopt the following strategy, regardless of whether they follow a positivist or naturalist model [Lew 2014]: they start from the typology of existing dictionaries and investigate, among other aspects, the users' needs as regards these works; who these users are and why they use dictionaries.Thus, this field can include, for example, research on all kinds of linguistic activities (reception, production and translation); the different types of dictionaries used for these respective activities (bilingual, monolingual and learners' dictionaries, etc.); they also take into consideration the users' skills and the strategies used to find the desired information in the dictionary (and to use it correctly) as well as the users' deficiencies (Welker 2008;Svenson 2009;Nesi 2013).Therefore it is clear that, in this context, the starting point for the compilation of a lexicographical work would be the typology of the dictionary, which must take into account the specific needs of real users.
However, this is no easy task.Nesi (2013) draws attention to the fact that the users, the types of use and the contexts can vary greatly.As a result: "In some other field of research large-scale controlled trials can test how effectively a given treatment works, but the effectiveness of a dictionary cannot usually be investigated by this means because it is difficult to enlist the aid of a representative sample of all potential users" (Nesi 2013: 65).For this reason, the author explains, there is a tendency to use empirical studies with small, specific groups, representing the users of a particular type of dictionary in a certain context.Nonetheless, researchers always take into account the limitations of this type of research and, in order not to risk extrapolating and generalising the results obtained and to increase their reliability, they carry out complementary studies.
As a general rule, in this type of research, we only find incidental references to phraseology and phraseological units: most works focus on other types of lexical units.In addition, it should be mentioned that there seems to be a tendency, as can be seen in Lew (2012b), to focus on a very specific aspect: the procedures used when searching for phraseological units.Despite all this, in works on the compilation of BPDs, for example in García Benito (2000), Torrent-Lenzen (20089 or Dobrovol'skij (2015), or in texts that review the treatment given to phraseological units in BDs, such as in Nuccorini (1988), repeated reference is made to the users of the dictionary, their needs and expectations.However, in these studies, there is normally no indication of the availability of data from the research with which to corroborate the statements and opinions expressed.
On the other hand, the Function Theory of Lexicography (Bergenholtz and Tarp 2003;Tarp 2008)13 adopts a radically different perspective.In the first place, it questions the exact nature of the oft-mentioned "specific needs", since this notion is understood to be too general and abstract.For this reason, it seeks to define exactly which type of need is lexicographically relevant (and concludes that it is the need for information) and it asserts that this need is always linked to a particular type of social situation, meaning that the needs of a single user may vary from one situation to another.In effect "para determinar las necesidades de los usuarios, éstas deben relacionarse no sólo a un determinado tipo de usuario (ya que los diferentes tipos de usuarios tienen diferentes necesidades), sino también a un determinado tipo de situación social que incluso influye con mayor peso que aquel en la caracterización del tipo de necesidades" [in order to determine the users' needs, they must be related not only to a certain type of user (since different types of users have different needs), but also to a certain type of social situation which is even more influential than the user in the characterization of the type of needs"] (Fuertes Olivera y Tarp 2008: 69).
In the second place, and following on from our previous points, the centre of attention now becomes the potential users of the dictionary and their situation, as opposed to the actual users.Another important aspect of this theoretical school of thought is the possibility for the user to have quick, easy access to the lexicographical information sought (an idea that is in accordance with the new trends that link lexicography to information sciences and information and communication technology [Nielsen 2013]).
In this way, instead of starting with their needs, the users' specific social situations are identified and analyzed first in order "to determine which type of needs a specific user may have in each type of situation" (Bergenholtz and Tarp 2003: 173); only then will a profile be drawn in order to reveal these users' characteristics.Therefore, the characterization of these needs is subordinated to the social situation.All this information is collected by means of deduction (working, as has been mentioned, with potential users).Thus, the definition of the type of users and their specific characteristics, together with the determination of the types of situations and the specific needs connected with said situations, frame the function of the dictionary.Only by starting from this base is it possible to determine the types of data to be included in the work and how these data will be presented.To sum up, the starting point of the compilation of a lexicographical work will be to establish the (extralexicographical) social situation of the potential user.The following social situations are considered to be lexicographically relevant: communicative, cognitive, interpretative and operational: 1) las comunicativas, donde se trata de ayudar a usuarios que encuentran problemas en diversas situaciones comunicativas (como la producción, recepción, traducción, revisión y corrección de textos); 2) las cognitivas, donde se trata de ayudar a usuarios que necesitan o desean aumentar su conocimiento de algo; 3) las operativas, donde se trata de ayudar a usuarios que necesitan consejos o instrucciones para realizar alguna acción física o mental no relacionada directamente con la comunicación; y 4) las interpretativas, donde se trata de ayudar a usuarios que necesitan interpretar algún signo, símbolo, 13 According to Fuentes Olivera (2012: 27), the "Teoría Funcional de la Lexicografía o teoría de las funciones es una construcción teórica iniciada en la década de 1990 en el Centre for Lexicography de la Universidad de Aarhus.Presenta la lexicografía como un área de práctica social y ciencia independiente que se ocupa del análisis y la construcción de los diccionarios que pueden satisfacer las necesidades de un tipo específico de usuario con problemas específicos que se presentan en una situación de uso específica" ["Function Theory of Lexicography or the theory of functions is a theoretical construct dating from the 1990s in the Centre for Lexicography of the University of Aarhus.It presents lexicography as an area of social practice and an independent science devoted to the analysis and compilation of dictionaries able to satisfy the needs of a specific type of user with specific problems that arise in a specific context"].etc. no lingüístico [1)Communicative functions, which serve to help users who find problems in different communicative situations (such as production, reception, translation, revision and correction of texts); 2) cognitive functions, which serve to help users who need or wish to increase their knowledge of something; 3) operative functions, which aim to help users who need advice or instructions in order to carry out some physical or mental action not directly related to communication; and 4) interpretative functions, which help users who need to interpret a non-linguistic sign, symbol, etc.].(Tarp 2014: 70) These four social situations in turn correspond to four basic categories of functions14.
As far as we have been able to establish, function theory has not been applied to bilingual phraseography15.However, based on the proposals that are being adopted by this theoretical framework, which are related, above all, to specialized and pedagogical lexicography, we can contemplate different possibilities for its application.The studies carried out by Fuertes Olivera and Tarp (2011); García Llamas (2013); van der Merwe and Fuertes Olivera (2014) and Tarp (2014) bear out this idea.
In accordance with what these authors propose, we believe that it is possible to direct the design of a BPD (or the inclusion of phraseological information in a BD) towards certain types of users, instead of opting for works described as "multiuse" since they are intended indiscriminately for professionals (translators and teachers), students (of languages or translation) or researchers (phraseologists, philologists, etc.).On this matter, and following in the steps of other works carried out according to the aforementioned theoretical paradigm, we believe that it is feasible to envisage the compilation of a BPD based on collaboration between a phraseographer and a specialist (depending on the context, the specialist could be a foreign language teacher, a translation teacher or a specialized translator, a phraseologist, etc.), with the aim of determining, by means of deduction, the potential situation of use, the user's characteristics and needs.These experts, being familiar with translation processes, learners' difficulties and needs, and so on, could help to deduce the "phraseographically pertinent" problems and needs (i.e. the problems and needs that can be solved or helped by means of a dictionary); which phraseographical data are required; or which functions are relevant (from our standpoint, it is very probable that, in the fields mentioned -research, translation, and teaching/learning-the communicative and cognitive functions will be most relevant).
In addition, Function Theory offers a series of parameters which enable us to establish the profile and characteristics of a group of potential users, which is fundamental since, according to Tarp (2014): "One type of user with specific characteristics will not necessarily have the same needs as another type of user with different characteristics, even when they are carrying out the same type of task".This type of procedure might help to attenuate one of the problems attributed to BDs.As Svenson (2009: 254) explains: "In a bilingual dictionary, the entries are often replete with information that is not needed by the individual user in a specific look-up situation.This is a disturbing factor which causes many users to give up before finding the information sought".
In fact, the possible scope of Function Theory is considerable and we have no intention of exhausting its possible applications to phraseography in this paper.As has already been pointed out, the idea underlying this study is that contributions from Function Theory can give rise to advances in both theoretical and practical bilingual phraseography.In the case of the former, it can take the form of the development of theoretical bases for the compilation of "specialized" BPDs, according to their users and projected uses.In practice, this would mean the production of works with a more uniform format which would be more appropriate for their target users.It would also entail the improvement of the means of inclusion and presentation of lexicographical data in order to make it effectively useful and relevant16.It could even bring about an increase in the formative and informative nature of dictionaries (with the development of "systematic introductions" or the use of data bases [Fuertes Olivera & Tarp 2011;Nielson 2013]) Similarly, we consider that the adoption of this theoretical perspective fits in perfectly with other lexicographical questions that are currently in vogue and which are essential to bilingual phraseography.That is, the adoption of function theory, as well as placing an emphasis on the user that was previously non-existent in theoretical and practical phraseography, would also motivate its connection with electronic lexicography and the appropriate use of parallel and comparative corpora.Both these points are essential in function theory, since they respectively help to improve the speed and ease of access to relevant lexicographical data and help to access pertinent lexicographical information according to the users' needs.
It must be remembered that these two areas, electronic lexicography and corpora, are not foreign to phraseography.Therefore, they have been extensively dealt with in several phraseographical works, a fact that confirms their relevance and pertinence to this field (Dronov 2011;Berti & Pinnavaia 2014;Bargalló Escrivá 2014;Granger & Lefer 2012;Paquot 2015, and others).Nonetheless, they are hardly ever directly linked to the users of dictionaries.Indeed, in general, their importance for the study of phraseology is highly considered and the way in which this can be reflected in the compilation of reference works is made clear.For example, in relation to BDs, Dawes (2014) considers that working with corpora can facilitate the inclusion of phraseological units that are loans, neologisms or of low frequency, and which, for those reasons, do not normally appear in lexicographical repertoires, although they do appear in the press and especially in demotivated uses (the author refers specifically to paremies such as: Publish or perish; The show must go on; Small is beautiful [Dawes 2014: 115]).
However, it is important to bear in mind that, if an adequate perspective is not adopted, there is a risk that the world of possibilities offered to phraseography by electronic lexicography and corpora could become yet another obstacle, since it would be possible to have much more exhaustive BPDs, but these could also be much more generic and inclusive.In fact, as we have mentioned, one of the consequences of research using corpora is the expansion of the frontiers of phraseology, which would give rise to the inclusion of new elements, such as lexical bundles17, resulting in the widening of the scope of phraseography.In our opinion, this widening can only be viewed positively if the new phraseological elements in dictionaries respond to the so-called user needs paradigm.
All these questions have obvious repercussions for the field of electronic lexicography, whose spectacular rise, mainly conditioned by the advantages of electronic dictionaries, is also reflected in bilingual phraseography.In this way, it is normally said that this type of work serves to facilitate access to phraseological units and to eliminate the problem of the limit on the amount of information offered.However, we believe that, here too, it must be remembered that this type of lexicography/phraseography also 16 For instance, this would enable us to judge the extent to which a BD of onomastic phraseological units is necessary (Szerszuniwicz 2008) or the level of coverage of idiom modifications in BDs (Dronov 2011).17 Work carried out with corpora has shown that certain lexical bundles perform crucial tasks in discourse.These are elements which, until the present day, did not appear to be important for lexicographical/phraseographical purposes, i.e. they were not potential candidates for inclusion in dictionaries.As has been shown by Biber, Conrad and Cortes (2004), lexical bundles do not have a clearly defined theoretical status.However, taking into account the fact that they function as blocks, they appear to be closely related to what is understood as a phraseological unit.It is clear that this is a question pertaining to theoretical phraseology, as can be seen in Steyer (2015), which also has general repercussions in phraseography.In this regard, some authors, such as Granger and Lefer have demanded the inclusion of these lexical items in BDs.They state that: "There is converging evidence that a large number of current web-based lexical searches concern multi-word strings (Lannoy 2010, Verlinde & Peeters forthcoming).It is therefore imperative to facilitate the access to longer strings and consider them not as second-rate citizens, but as fully-fledged entries on a par with single words.Our analysis shows that the addition of lexical bundles to bilingual dictionaries could contribute to improving their phraseological coverage.Coupled with efforts made by other scholars in the area of collocations (cf. Ferraresi et al. 2010), our results show that major advances are within reach as regards both the actual number of phrasal units included in bilingual dictionaries and their translation".(Granger and Lefer 2012: 691) requires that attention be paid to the user needs paradigm.As has been pointed out, electronic dictionaries require changes in the traditional concept of how dictionaries are used (Lew 2012a(Lew , 2012b(Lew , 2013)), as well as needing great care in the presentation of data.Otherwise, there exists the risk of overwhelming the user with such a huge amount of information, codes, and resources that it may prove counterproductive: if the user has no specialized knowledge he/she will be unable to turn all these data into information.

Conclusions
This article has covered some questions relative to bilingual phraseography.The existence of a large volume of publications related directly or indirectly to this topic shows the interest it arouses and, at the same time, reveals a very diverse and sometimes contradictory panorama.There seems to be consensus only in relation to the negative or inadequate aspects of the treatment of phraseological units in BDs and BPDs.On the other hand, the virtual absence of homogeneous parameters regarding what is considered positive is striking.Among the wide range of independent proposals, formulated according to the authors' own criteria, sometimes not transferrable to other proposals or simply not valued in other works, two aspects are outstanding: firstly, a great concern for the object of study, that is, phraseological units; secondly, a lack of reference to the user needs paradigm.As a result, this aspect has been rather neglected and, in our opinion, this has had negative repercussions for the development of phraseography.
For the reasons given above, we support the idea that the development of two main approaches must be accepted in bilingual phraseography: one aimed at the general framework of phraseography, since it is necessary to take a stand in the face of certain phraseographical problems, such as that of equivalence, and the other focusing on certain aspects that have been disregard, but which are also an integral part of bilingual phraseography, such as users' needs.In this regard, we suggest a possible means of solving these problems: the adoption of the user needs paradigm and, consequently, the contributions of the Function Theory of Lexicography (as well as this theory's relationship with other advances, such as electronic lexicography and the use of parallel and comparative corpora).We believe that phraseographers and lexicographers should not ignore this option, bearing in mind the scope of the problems and deficiencies that remain to be solved in the area of phraseography.