Abstract

This paper first describes the complications that need to be overcome in order to describe the different senses of lexical items for translation purposes and presents an interpretation table of conventional senses of the lexical item business in specialized expressions and compound terms for English-French translation. Contrary to the recognised divide between words in general vocabulary and terms in specialized language, conventional senses of lexical items do contribute to the creation and interpretation of specialized expressions and compound terms they form. The fact that a limited number of conventional senses of lexical items are used in the creation and understanding of most expressions and compound terms makes it possible to describe recurrent translation procedures and competencies and to teach them. The second part of this paper provides a textual characterization of contextual features, semantic selection restrictions, grammatical properties and syntactical patterns of the senses of business that have been collected in a small corpus randomly selected from the Wikipedia Corpus of Mark Davies (2015) available on-line. This analysis makes it possible to establish correlations of textual features with specific senses in the corpus and to propose conditional interpretation rules and a method of interpreting business lexical items as constituents of expressions and compound terms found in the corpus.

Keywords: specialised translation, corpus-based translation studies, intertextuality, textual features, lexical patterns, business

1. Introduction

The focus of this study is the interpretation of specialized uses of the English term business in economic, business and financial texts. The purpose of this paper is to identify interpretation rules of the meanings (or more technically, the senses) of the lexical item business that is a constituent of expressions and compound terms in specialized texts for the purpose of translation from English to French. As regards translation (in any language), the identification of interpretation rules represents half the work for translators that also have to produce an acceptable correspondent or equivalent once the meaning is identified.. As opposed to the interpretation rules which are analytical and have a unique unequivocal solution, translation decisions initiated with an interpretation of source text are creative and imply the production of multiple solutions in target language, as have shown Pym (2003). Because of that, translation decisions and target language productions can be studied separately from a given interpretation of a source text. The study of creative translation decisions are not the focus of this paper, although it is important to recognize that translation decisions are initiated from a given interpretation. The systematic description of the interpretation rules followed by translators offer the advantage of describing the analytical skills translators must demonstrate and acquire first in the translation of expressions and compound terms formed with the lexical item business.

Subject-matter vocabulary, conceptual ontologies, processes and activities to be acquired by specialized translators are for the most part represented linguistically by terms, as opposed to words which define the basic lexical items used in general texts, as suggested by Chueca Moncayo (2005: 42). Technically, as defined in any handbook of terminology, terms are characterized as “a word (simple term), multiword expression (complex term, symbol or formula) that designates a particular concept within a subject field” (Pavel and Nolet 2001: 117). In fact, the class of terms overlaps considerably with the class of words. Some simple or unitary terms belong exclusively to a specialized field of knowledge, others share the same form with words but differ from them in their specialized uses and meanings, and some terms are even comprised of a combination of words (or of words and unitary terms) in the case of complex or compound terms. Only a very small number of terms (symbols and formulas) are not words or comprised of words. Except for those terms, the definition of words (and of their meaning properties as pointed out below) plays a crucial role in the definition of terms and of special language knowledge, ontologies, processes and activities. In order to describe which lexical, terminological and phraseological items are processed in the translation of specialized documents, and necessarily how they are processed, translators, and especially translator trainers need to have a clear understanding of what those elements are precisely.

2. Words, terms, and lexical items

It is well known that the word word is an ambiguous and vague term for such a central concept in translation where words are the building blocks of meaning and translation.[1] Following Mel’čuk (1993: 98-100), the word word has two major senses: a word₁ is a word-form and a word₂ is a lexeme (the eminent linguist suggested to use subscripts to differentiate the two senses that have the same word-form – see below). The first sense of the word word (word₁) is a concrete entity, a textual unit most often separated by blank spaces. The second sense of the word word (word₂) is an abstract entity, a lexicographic unit to which a core (and specific) meaning is associated. This distinction is in reality similar to the distinction made by Chueca Moncayo (2005) citing Jackson (1988: 8) regarding the following third and fourth meanings of the word word:

We have identified orthographic words, words distinguished from each other by their spelling. Secondly, we have identified phonological words, distinguished from each other by their pronunciation. Thirdly, we have identified word-forms, which are grammatical variants. And fourthly, we have identified words as ‘items of meaning’, the headwords of dictionary entries, which are called lexemes.

In the context of Mel’čuk’s (1993) systematic description of morphology, the first and the second kinds of words (orthographic words and phonological words) would probably be considered as linguistic properties of lexemes and not of word-forms which a) have one specific orthography that normally distinguishes separate word-forms and b) have no specific phonological component since they are textual units and are instances or occurrences of lexemes from which their phonological component could be retrieved when needed. These definitions have significant consequences for translation. First and foremost, word-forms and lexemes form an inseparable pair in which word-forms are concrete textual occurrences of lexemes. This relationship is clearly of the well known token-type kind, as suggested by Lyons (1977: 13). For translation and lexicographic purposes, this particular token and type relationship between word-forms and lexemes has a second important consequence which is that a lexeme is, in fact, an abstract entity of lexemes (plural intended) that each have their group of word-forms. This means that lexemes or lexical items are not equivalent to headwords in dictionaries but to the various subdivisions (syntactical or semantic) of the dictionary entries made in practice to distinguish different senses grouped under one particular “type” or “sub-type” of lexeme. This approach to lexemes and to word-forms relationship with specific lexeme results in the existence of different lexemes considered as homonyms, i.e. having the same form but differing in meaning, even if those different lexemes are described in a dictionary entry under one single headword (as different senses of a single polysemous lexeme). Whether they are treated in a dictionary under one or several headwords, homonymous lexemes are just different lexemes for translation purposes and are not occurrences of the same type of lexeme since, as Lyons (1977: 20) explained, the type-token distinction does not apply to lexemes per se.[2]

For the purpose of this article, the term “lexical item” will be used as a particular kind of lexeme which is a group of word-forms sharing the same meaning, combinatorial and syntactical properties and that could be associated with either a dictionary headword or, when the lexicographers and the usage distinguish more than one sense for a lexeme, a distinct lexeme such as illustrated by subdivisions of a (monolingual) dictionary entry. As stated above, lexical items will be either word-forms expressing general language meaning units of a text or terms expressing specialized language meaning units of a text.

Since terms are generally defined as monosemic entities, there is no real practical reason to distinguish the concept of “term-form” as a term occurrence and the concept of term-type as a “terminological” lexeme, even if there is a difference between a term-form and a term-type. One example would be the relationship between the acronym (term-form) and its definition (term-type) such as the plural form of a term-type that is described in its singular form or ASAP standing for “as soon as possible”. As regards the relation of constituent word-forms to the compound term they are a constituent of, this relationship is predictable and transparent or unpredictable and opaque. In the first case, the relationship is better described in the grammar or syntactical patterns of one or the other constituent (head noun or noun modifier). In the latter case, the compound term needs the same treatment as the one reserved for phraseological units or idioms. However, the predictability or transparency and unpredictability and opacity of compounds are computed differently by native and second-language speakers. In the context of translation, source language (such as English in this paper) is most of the time the second language of translators. That means that English idiom criteria of definition for translators from English will be somehow target-language dependent and probably less restrictive than for native speakers of English.[3] The same could also be said about translators and subject matter specialists (which are normally less proficient in their second language than translators). In the interpretation (and translation) of expressions and compounds in specialized texts, the appreciation of their transparency or opacity is a very important source language expertise of translators and shall be taken into account in the definition of interpretation and translation rules based on lexical items constituent of terms.[4]

3. Meaning of compound terms

As shown in the previous section, terms are most comprised of words. For practical reasons and according to one of the generally agreed lexicographical practices (see below), unitary words are considered to belong to general vocabulary whereas compound phrases and terms are considered to belong to specialized vocabulary. Although not systematic, this principle allows us to draw a useful line between lexical items that are part of the general language and those that are part of specialized terminologies. For example, most specialized monolingual dictionaries do not have an entry for the term business as a specialized unitary lexical item and just describe compound entries formed with the lexical item business (almost exclusively used as an attributive noun modifier). This is the case with The New Penguin Dictionary of Business (2002) and the Oxford Dictionary of Business (2003). Both have no entry for business as an unitary term and between 14 to 28 entries for compound terms starting with the word-form business (such as business analysis, business angel, business cycle, business entity concept, etc.). However practical, this lexicographic cleavage of words and compound terms is not consistent with the essential contribution of general senses of words in the using and interpretation of specialized terms.

The typical uses in specialized texts of what Gezymisch-Arbogast (1989) called passepartout words provide circumstantial evidence for the contribution of unitary lexical items in the uses of compound terms. According to the author, passepartout words are vague unitary words commonly used as substitutes for compound terms. When used in specialized texts, passepartout words create interpretation ambiguities due to the impossibility to associate the unitary word occurrence as an occurrence of either one of its several compound term of which it is a constituent. She cites the example of assets that may stand as assets₁ for “tangible fixed assets” or assets₂ for “movable fixed assets”. The switch from compound to elliptical term[5] she explains is caused by a textual shift from the system level of language to the pragmatic level of parole (specific textual uses relevant for a short excerpt or a paragraph). These textual switches illustrate the contribution of unitary terms in the use and interpretation of compound terms. Although passepartout words brings about a vagueness in specialized texts (and increases the difficulty of interpretation for non-specialists), their high frequency suggests that specialists make those switches to focus their communicative efforts on the most economical terms without sacrificing too much readability for efficiency. The economy principle in language, rooted in the general principle of least effort, is well documented in linguistic (see Vicentini 2003) and language use. It could be even more significant in specialized texts written for specialists.

An important consequence of the contribution of lexical items in terms is that the main senses of words are an integral part of specialized vocabulary and terminologies. In other terms, some general senses of lexical items are crucial to the interpretation and translation of specialized compound terms. In fact, this inherent incorporation of lexical items in terms probably explains why most translators who have a good active knowledge of semi-specialized lexical items or even general vocabulary are often ahead of content-matter specialists in the translation of specialized texts.[6] They have better lexical and syntactical resources than most specialists who, by their strong bias towards specialization, are often less proficient in general culture and vocabulary[7]. For this reason, the study of specialized uses of business shall take into account some of the general senses of the word business described in monolingual general dictionaries that applies to the interpretation of compound terms in the field. In fact, this step is necessary for the description of the efforts and processes involved in the translation of compound terms formed with the term business. The interpretation rules are based on syntactic and textual “reading” of specialized sense distribution within compound terms and on the knowledge of syntactical patterns of construction of terms that can be recognized and interpreted similarly.

4. The inherent vagueness of bilingual dictionaries in translation

Since lexical items clearly contribute to the meaning and interpretation of compound terms, specialized translators or translation learners may be inclined to consult a general bilingual dictionary for interpreting predictable and transparent lexical items in compound terms. However, bilingual dictionaries in this regards are often misleading and disappointing. Bilingual dictionary entries have inherent vagueness in the discrimination of senses that create major problems in specialized translation. That is because bilingual dictionaries do not usually identify specialized or even precise (whether specialized or not) senses of the source language lexical items. For practical reasons like if target language solutions are identical for two or more source language senses, source language senses are aggregated under one target language equivalent. At best, if by chance two source senses each match with one different target language equivalent the dictionary subdivisions may help to disambiguate the source senses, but when this is not the case, no distinctive identification of source language specific or specialized sense is provided to the users. For example, in the business entry of Collins-Robert Unabridged French Dictionary (2006), the first main meaning of business # 1 a: “[uncountable] “commerce” -> affaires^fpl” contains several unspecified senses under the same translation affaires, such as “to do business with [somebody] : faire des affaires avec [quelqu’un] ” (which is defined in Wordnet 3.1 (see below) as “the activity of providing goods and services involving financial and commercial and industrial aspects”), “his business is cattle rearing : il a une affaire d’élevage de bestiaux” (which is defined in Wordnet 3.1 (see below) as “the principal activity in your life that you do to earn money”), “the music business : le secteur musical” (which is defined in Wordnet 3.1 (see below) as “business sector”). Those three specific senses (specialized or semi-specialized in economics, business, and finance) are described in the same sense: [uncountable] commerce. The presentation of the entry imply that all senses are generally translated by affaires in French, and the examples provided above are either using affaire or another translation (such as secteur musical for music business). The interpretation given to the subentries is that its meanings are translated as a general rule with commerce and that the rest of the entry is dedicated to similar examples of that rule as well as counter-examples. This loose aggregation of specific source language senses also forces the users to take into account (ie read and analyze) all of the not necessarily relevant examples provided in the subentries instead of checking only for the specific sense in the source text. Another problem is that affaire is a word-form that has several lexical items and senses just as is the case for business. Since the dictionary entries do not discriminate the specific senses in source language and target language, the user is confused with the specific sense that is to be translated and that refers to the target language equivalent provided. For translators and students in specialized translation training courses, the specific senses of lexical items are inadequately specified in the bilingual dictionary which generates a high risk of inadequate choice of equivalent, one that does not convey the specific sense intended in the source text. Furthermore, this treatment is an important source of interferences in the correct uses of target language specialized terms and lexical items. In this case, the problem is not related to the disambiguation of meanings in source language text but has more to do with the incorrect attribution of specialized meanings to target language lexical items from source language senses (particularly in the case of false cognates and calques).

One could argue that bilingual dictionaries are not particularly intended for helping translators (which most usually translate in their strongest language in which they have native-like proficiency), but for second-language encoding communication situations where the user is presumed to have no active knowledge of target language terms and expressions (as opposed to their first language). Another argument that could be provided against the description of specific meaning of lexical items in bilingual dictionaries is the fact that bilingual dictionaries are obviously not designed for specialized translation. This line of reasoning seems flawed taking into account the role that simple lexical items play in the interpretation of compound terms. Whatever the issue at stake, the fact remains that there is an acute need for specialized translators and learners of translation to have access to better tools for interpreting simple lexical items used in compound terms.

On a contrastive and theoretical level, each word in both languages has several senses that are not necessarily compatible since any lexical unit has its structure or pattern of lexical senses. This is particularly true for false cognates that have been described in the famous work of Van Roey, Granger and Swallow (1988) where specific meanings of numerous pairs of cognates (circuler/circular, cité/city, clair/clear, etc.) are grouped into three sections: an equivalent section where meanings are equivalent for the two cognates, and two contrastive and differential sections where either the meaning of the French word cannot be translated by its cognate word in English or the meaning of the English word cannot be translated by its cognate word in French. By using a lexical item of the target language to represent the meaning of a lexical item in the source language, bilingual dictionaries systematically implies that all the senses of the lexical item in the target language belong to the senses of the lexical item in the source language, and, inversely, that all senses of the target lexical item belong to the senses of the source lexical item. In most cases, bidirectional inference is rarely true (as shown especially for false cognates) since lexicalized senses in words of two languages are rarely similar in nature and quantity.

The intricately confusing mixture of word-forms and senses between two equivalent lexical items from two different languages could explain the trend in English-French translation in economics, business and finance to align the structure of senses and meanings of lexical items in compound terms with their correspondent English senses, or even to borrow English compound terms, as Camilla Ferard (2009) study have shown in the case of “source language transference” for price earnings ratio, value at risk, correspondent banking and sub-prime.

5. General meanings of business in specialized terms

A first step in the description of interpretation rules of business in compound terms is the identification of specific senses contributing to compound terms. A good start is a monolingual dictionary that offers precise identification of general and specific senses of lexical items in a specially structured entry. In a general dictionary such as the Canadian Oxford Dictionary (2001) (COD), the entry business contains eleven main senses occasionally subdivided in “lettered senses”, as explained in the page xv of the Guide to the Use of This Dictionary, as well as eleven idioms listed after the main senses. The next figure presents the dictionary entry exactly as it appears.

Figure 1. Entry for business in COD

As explained above, only some senses of the lexical item business contribute to specialized compound terms. In the main senses of the lexical item, meanings “2 a thing that is one’s concern (none of your business) […] 3 b a reason for coming (what is your business?) 4 serious work or activity (get down to business), 5 derogatory an affair, a matter (sick of the whole business) and 10 N Amer. euphemism (esp. of pets) an occurrence of defecation or urination” can be considered as unspecialized uses of the lexical item business. Meaning 11 “Theatre action on stage, as opposed to dialogue” can be considered a specialized term that does not belong by definition to economic, business and finance translation. For the purpose of this paper, those senses were considered general senses and not considered as building blocks, expression or terms to learn to practice specialized translation. Although not definitive, these exclusions are made in practice for the purpose of analyzing the interpretation, and eventually, the translation of business uses in economic, business and financial documents. The exclusions mean that even if these meanings may appear in economic, business and financial texts, they are not typical of the topics, vocabulary, terminology and conceptual ontologies, processes and activities of the field.

What is interesting is the fact that most idioms defined after the main senses could all be considered part of the specialized expressions and terms in use in the field of economics, business, and financial translation. This indicates that there are two types of specialized complex terms: those that are idiomatic, lexicalized and non-compositional, and those that are non-idiomatic, “grammaticalized” and interpretable compositionally. The latter type seems also to be very productive and is forming the bulk of terms in terminological databanks. Specialized dictionaries as we have seen, tend to not make this distinction between idiomatic and non-idiomatic complex terms and rely on complex versus unitary uses of terms as a general criteria for inclusion in their entry list.

The following figure shows the seven specialized senses or word forms of business that can be retrieved from the COD dictionary entry in figure 1. The numbers in parenthesis refer to the sense number used in the dictionary entry. In the discussion that follows, sense numbers refer to those indicated in the figure below.

1. (1) one’s regular occupation, profession or trade
2. (3a) a task or duty
3. (6) a thing or series of things needing to be dealt with (the business of the day)
4. (7) volume of trade (did a lot of business)
5. (8a) a company or a corporation
6. (8b) commercial enterprises collectively (the government needs the support of business)
7. (9) patronage; custom (take my business elsewhere)

Figure 2. Specialized meanings of business in COD

The remaining senses in the previous figure 2 show a high similarity with the results obtained in Wordnet 3.1 which offer the definition and classification of concepts associated with the word-form business. This resource is not a traditional monolingual dictionary but “a large lexical database of English” where “Nouns, verbs, adjectives and adverbs are grouped into sets of cognitive synonyms (synsets), each expressing a distinct concept”. The concepts of the term business are almost similar to the lexicographic description of the same lexical item in the general monolingual dictionary. In the figure 3 below, like for the general meanings in the dictionary entry, word senses not typical of economics, business or finance have been excluded and presented as strikeout text. Each of the remaining specialized senses is numbered. In the discussion that follows, sense numbers refer to those indicated in the figure 3 below.

• 1 S: (n) business, concern, business concern, business organization, business organisation (a commercial or industrial enterprise and the people who constitute it) “he bought his brother’s business”; “a small mom-and-pop business”; “a racially integrated business concern”
• 2 S: (n) commercial enterprise, business enterprise, business (the activity of providing goods and services involving financial and commercial and industrial aspects) “computers are now widely used in business”
• 3 S: (n) occupation, business, job, line of work, line (the principal activity in your life that you do to earn money) “he’s not in my line of business”
• S: (n) business (a rightful concern or responsibility) “it’s none of your business”; “mind your own business”
• S: (n) business (an immediate objective) “gossip was the main business of the evening”
• 4 S: (n) business (the volume of commercial activity) “business is good today”; “show me where the business was today”
• 5 S: (n) business, business sector (business concerns collectively) “Government and business could not agree”
• 6 S: (n) clientele, patronage, business (customers collectively) “they have an upper class clientele”
• S: (n) business, stage business, byplay (incidental activity performed by an actor for dramatic effect) “his business with the cane was hilarious”

Figure 3. Lexical senses of business (Wordnet 3.1)

General senses excluded, five of the specialized senses of business in Wordnet are very similar with five senses of the COD as described in figure 2: sense 1 of Wordnet with sense 5 of the COD, sense 3 of the COD with sense 1 of Wordnet, sense 4 with sense 4, sense 5 with sense 6 and sense 6 with sense 7. Sense 2 in Wordnet is not present in COD (other than in the compound “in business 1” in the idiom section of the dictionary entry; and senses 2 and 3 of the COD are not present in Wordnet. This simple comparison of senses organization illustrates the editorial subjectivity (and freedom) of dictionaries and lexical databases in the selection, grouping and definition of senses and raises the important issue of incompleteness of lexicographic treatment of senses for translation purpose. Unlike the vagueness of bilingual dictionaries in the treatment of senses, general monolingual dictionaries offer a limited and selective description of senses.

Translators who do not master their source language as well as their native target language are directly impacted by the editorial practices and decisions of the monolingual resource they consult and are never totally sure if a specific textual use of a lexical item they need to translate is conventional but out of the scope of the dictionary’s lexical item coverage. In case of discrepancies between different resources, like shown with the COD and Wordnet, translators are left with the decision to compensate for the lexicographic gray areas of sense descriptions. This happens to be a very common challenge facing translation learners who are not prepared and trained to compensate for these comparative deficiencies of otherwise renowned monolingual dictionaries and resources. For example, the understanding of business in the sentence The corporation went out of business in 2001 may be difficult if not impossible with the sole description of the COD considering the absence of a meaning referring directly to commercial activities and the absence of the expression “out of business” that could have been associated with “in business 1” in the idiom section of the dictionary entry. On the other hand, a sentence such as “He is not there [in a city] on business, since he appears to have private means”, might give the same difficulty to the translators using the conceptual description of Wordnet where this meaning is totally absent.

Those limitations of monolingual resources emphasize the need for a better and more systematic description of the lexical item senses contributing to compound terms. No monolingual dictionary or language database (like in Wordnet 3.1) is designed to fulfill the translators’ need for a description of most if not all of the specialized senses of lexical items involved in the creation and interpretation of specialized compound terms. This is a very important conclusion for the description of techniques and meaning-based processes of interpretation and translation of specialized compound terms that are based on the textual correlations of specialized lexical item senses.

6. The interpretation table for specialized business senses

The incompleteness of monolingual databases and dictionaries makes it necessary to build a complete representation of the lexical item senses. As for the selection of senses belonging to the field of economics, business and financial texts, specialized sense representation and organization shall be flexible and tailored to translator’s needs which are partly determined by conventional translations of senses in their target language, as is shown below. For the design of the interpretation table, one good strategy consists of putting together the similar definitions found in the resources and decide on the addition of other specific senses or meanings described in another resource. Another strategy is to proceed like lexicographers and examine textual translations of different senses to get a broad view of the most common senses and equivalents for them. Both strategies may be used in parallel, and that is what has been done for this paper.

The interpretation table in the next figure provides a description of conceptual and referential senses that are conventionally associated with specialized occurrences of the lexical item as used in compound terms from the economic, business and finance vocabulary. Except for neologisms and perhaps regional variations (which may be easily integrated), the existence of counter-examples and departures from senses and meanings in the interpretation table is to be understood as what Patrick Hanks (2013) describe as “exploitations of the norm”. The first column in figure 4 presents three main referential meanings of the lexical item business in compound terms. The second column in the figure describes the specific senses that are contributing to the meaning of compound terms formed with the lexical item business. The three main meanings result from the identification of referential concepts in source language that are recurrent in all occurrences of the lexical item. In the case of business, those referential concepts are an occupation or duties, a commercial activity, and a profit-making entity or group of such entities or people seeking profits. The existence of the three main meanings facilitates the search or the analysis of lexical item occurrences by reducing the alternatives in the interpretation process to three instead of eight for the specific senses. For the first and third main meanings, it could have been justified to separate the referential concepts duties and people seeking profits as two other main meanings. The sense organization has been determined while the corpus was analyzed so too many changes in the interpretation table would have impacted the codification of the data. It was decided not to implement the changes required for the addition of two main meanings to the three already distinguished. Also, it was estimated that the impact on the following analysis would be marginal. As a matter of fact, the number of occurrences of senses 1b and 3c found in the corpus (see below) is limited to 2 and 9 respectively. Still, the referential uniqueness of the duties and people seeking profit components of main meanings 1 and 3 nevertheless appear to justify a distinct treatment in the interpretation table. What should be noted is that the identification of the main meanings is based on referential meanings recurrent in source language. The specific sense definitions are followed by typical phrase examples and expressions.

Figure 4. Interpretation table for specialized business senses

The identification of specific senses and their connection to a main meaning are more dependent on recurrent translations in target language than main meanings. Their definition and structural role in the table is partly based on target language equivalents recurrently found for those occurrences. For that reason, the sense definitions and organization might be specific to English-French cross-linguistic conceptual semantics. For example, the connection of sense 3c to main meaning 3 is due to its recurrent translation by secteur which refers more closely in French to an industry or to a set of “economic agents” than to a group of similar activities. This is different for example as the treatment of the same sense in the Longman Dictionary of Contemporary English, where this sense is instead connected to the main meaning 2 referring to the activity of “buying or selling goods or services”. This situation clearly shows how translation can induce a different perspective on source language interpretation of lexical items. It does not necessarily mean that senses perceived differently in two languages are entirely different. On the contrary, they have a lot in common. The most obvious impact of translation of the senses are their organization and not on their definition.

The interpretation table design is based on the referential and semantic properties of the lexical item business in compound terms. In order to analyze the contribution of other criteria in the interpretation process, other crucial information are processed by the translators such as the grammatical number (plural is relevant in the case of business, as will be shown below) as well as the syntactical status and, in some other cases, semantic classes of lexical items in certain syntactic roles and collocational elements in syntactic constructions. Even specific contextual information may contribute to the interpretation process. This information needs to be processed and correlated with the sense of business in compound terms and are discussed in details in the next section.

7. Distribution of senses in the corpus

For the purpose of this paper, a corpus of occurrences of business was collected in the Wikipedia Corpus from Mark Davies (which is available on-line). The Wikipedia Corpus was selected from other available English Corpora to get a better coverage of specialized texts. To validate the interpretation table, and to study textual correlations with the lexical item senses of business contributing to the meaning of compound terms, a small corpus of 301 occurrences randomly selected have been codified semantically as defined in the interpretation table with each sense roughly translated in French.

The sample taken from the Wikipedia corpus was comprised of the first page of 101 occurrences of businesses in plural and two different pages (1 and 5) of 100 occurrences (number 1-101[8] and 400-500 respectively) of the word-form business in singular. Each occurrence has been classified in one of the senses provided in the interpretation table. For some occurrences, it was not possible to decide on one specific sense and so the correct classification of the occurrence required the consultation of the referred Web page in order to obtain more contextual information. The following figure shows the classification of token senses among the three samples of the corpus.

Figure 5. Distribution of senses in the 301 occurrences of the Wikipedia Corpus

From all the occurrences in the corpus, just one referential meaning and sense could not have been interpreted and identified and is shown in the last row of the table. The occurrence appears on the Wikipedia page of Sergey Mironov and describes a type of medal given to him.[9] Before that last row, five general language meanings and senses have been found in the corpus and marked GL. In those case, the interpretation of business in compound terms relate to senses such as a difficult job, a preoccupation or an important issue. Figure 5 shows no example of sense 2b and 2c and only two occurrences of sense 1b. Both occurrences of 1b seem to be related to uncountable nouns status such as in “to attend to other business” and “the last piece of business”. More robust interpretation rules based on textual correlations of senses 2b, 2c and 1b could certainly be obtained with a larger corpus and with more occurrences of these senses. In the last section of the paper which suggest a decision-making process, these senses will not be taken into account for the lack of data on their textual features.

The figure 5 allows to draw two important conclusions for the interpretation of business in expressions and compound terms in the corpus of 301 tokens that have been collected for this paper. First, the grammatical plural form of business (businesses) is consistently correlated (99 occurrences out of 101) with sense 3a that makes the interpretation of the token businesses very straight forward. The only exceptions of this are the two occurrences of 2a that refer to a business entity that has several businesses and confirm the strong bias towards sense 3a. In one of the context, the word-form businesses is clearly used as a synonym of division which is a sub-entity of a business organization. The two examples show a specific ambiguity of business when the lexical item refers to activities conducted by an entity (a division or a corporation) that is the property of a parent company or body or a holding corporation. In that case, the translators have the choice to interpret business as an activity or as an entity. One of the contextual information that have to be taken into account is the fact that this sub-entity has a specific proper name or not. In that case, an adequate equivalent would be a target language term that refers to an entity and on the contrary, if there is no obvious presence of a specific entity with its own corporate name, then the translators may use a term that refers to activities.

Second, the most statistically common senses for business in specialized expressions and compound terms seem to be limited to senses 2a, 3a, and 3b of business in singular. Senses 1a and 3c are used marginally in the corpus used (this conclusion would need to be verified in a larger corpus). That means that in order to interpret any occurrence of business lexical items in the occurrences of the corpus, most of the efforts shall be directed firstly towards senses 3a, 3b and 2a, and secondly towards 1a and 3c of business in singular. But for translation practitioners and learners, this statistical approach is unsatisfactory and counter-intuitive because it does not account for textual features that are co-occurring with tokens and that are used by translators and learners to understand specific senses of business lexical item in specialized compound terms. The next and last section of this paper will explore a more symbolic approach to the interpretation of business senses 1a, 2a, 3a, 3b and 3c in complex terms. Textual features such as contextual features and grammatical patterns will be analyzed in order to correlate them with specific senses of business in singular.

8. Contextual features relevant to the interpretation of business

The corpus studied has shown a marginal but significant feature in the interpretation of a specific business sense. For the sense 3b, a few occurrences are associated with the name of the academic topic of business in expressions like business administration, business expertise, business skills, business (three unitary occurrences), business and management, language of business. The occurrences in the corpus were not significant to require a specific sense, but a larger corpus may provide significant evidence of it. In some cases, the interpretation of this specific sense of business is correlated to collocations such as administration, expertise, skills. In other cases, business lexical item appears as a unitary expression and no obvious criteria seems to characterize this sense. Still, it’s possible to interpret this sense of business with the awareness of contextual information that are present with the sense and which all describe academic disciplines or fields of study such as management, administration, accounting, marketing and even medicine, law, engineering, humanities, science, and so on. The interpretation of business in this expression seems to be related to contextual information related to academic disciplines. The interpretation of the other senses of business does not rely on such contextual information, as will be shown in the following section.

9. Grammatical patterns of business senses in compound terms

After the manual classification of sense occurrences according to the interpretation table, the 301 occurrences of the corpus were analyzed as regards syntactical patterns. Each occurrence was characterized either as a minimal phrase (unitary use), as a head noun with modifiers or as a modifier in a compound term or as part of an idiomatic compound or expression. A further and more detailed classification of patterns was realized. For example, when business is used as a lexical item in a minimal phrase, the different constructions that were identified were occurrence of business alone (Unitary), or with article indefinite (Ua) or definite (Uthe) or quantifier (Quant). Sense 1a was correlated with two prepositional phrases such as on business and for business that are related to instances of traveling or getting into places for one’s occupation or work. Also, a specific grammatical property of sense 3c is the presence of the indefinite article the at the beginning of the compound terms formed with business as a head noun. As regards the uses of the article the, the difference between 3a and 3c relies on the systematic absence of a noun modifier for the senses 3a, as opposed to 3c where the article is consistently preceded by noun modifiers that characterize the activity sector denoted with the compound term. Other more specific grammatical status were codified, but no specific correlation were identified with business senses. For instance, the presence of a possessive modifier of the term was not correlated with a specific sense of business. This textual feature was present in 1a, 2a and 3a but not in 3b and 3c occurrences. That means that this feature can be used in combination with other features for the interpretation of business sense 3b or 3c when business is used as a unitary expression or as the head noun of a complex term.

The subcorpus of business in plural excluded, the significant senses of business lexical item in expressions and compound terms (senses 1a, 2a, 3a, 3b and 3c) have the syntactical patterns described in the figure 7. Combined with semantical properties, these syntactical patterns allow to make some interesting distinctions in the patterns of use of business senses.

Figure 6. Syntactical patterns of senses 1a, 2a, 3a, 3b and 3c

To these 192 occurrences of business, eight other occurrences were processed. Two of them were classified as sense 1c (not analyzed for insufficient data to make textual correlations), five were considered as belonging to general language and one was considered uninterpretable as already shown in figure 5. The category of idioms here refers to constructions such as do business or business-to-business in which business cannot be classified as a unitary, head noun or modifier.

The most interesting numbers concern the sense 3b where it is mostly used as a modifier. The unitary senses of 3b can be separated into two groups: three occurrences of the expression in business and five occurrences of business as a discipline as discussed in the previous section. Among three other unitary uses, two of them are also uses of business as a discipline. These numbers are evidence of a distinct sense of business to refer to an academic discipline. Other than the discipline, the typical interpretation of this sense is mostly used as a modifier referring to the world of business, a concept that is dominant in the Western European and North American capitalist economies and democracies. As shown in figure 6, sense 2a is very often used as a modifier.

An interesting semantic selection restriction distinguishes sense 3b and 2a. Head nouns for the senses 3b mostly refer to people in general and not to a specific business entity, as opposed to head nouns for the senses 2a which refer specifically to people involved in the operating of a specific business entity. This is what can distinguish for instance the compound terms such as business leaders, business community for the sense 3b and the compound terms such as business owners, business executives, business partners for the sense 2a. The distinction between singular and plural form does not seem to be relevant for the distinction between the two senses since business in plural is never used as a modifier. As regards non-human nouns, one can find the same difference in semantic restrictions between the two senses: 3b is used with more abstract and general concepts such as business ethics, business world, business news while 2a is commonly used with nouns referring to properties or attributes of one’s specific business entity such as in business deal, business machine, business process or business strategy or business decision.

For sense 3a as a modifier, most of the examples in the corpus, 10 out of 13, can be correlated to determinative modifiers uses where the lexical item business can be interpreted as a syntactical complement of the head noun. For example, in this sense, business market can be interpreted as the market for/of the business, or the market targeting businesses. This interpretation rule highlights the role of business sense 3a in unitary expression or as a head noun in compound terms. Unlike senses 2a and 3b as a unitary expression, sense 3a often appears with the definite or indefinite article.

10. Discussion and textual correlations with sense interpreting

The corpus analysis of 301 occurrences of business has shown that out of 8 specialized senses of business defined in the interpretation table, only three of them are very common (2a, 3a and 3b), two of them were absent from the corpus (2b and 2c) and three others (1a, 1b and 3c) were marginally found in the interpretation of business in expressions and compound terms. Regarding the interpretation process for translation purpose, many contextual, semantic and syntactical features of the occurrences have been analyzed and checked in the previous sections for their correlation with specific senses. This section aims at proposing conditional features identifying specific senses and suggests rules of interpretation for the occurrences of business lexical items in expressions and compound terms. It is our view that contextual, semantic and syntactical features all contribute at some point to the interpretation of the lexical item business in specialized expressions and compound terms. In this last section of the paper, we present the sense correlations that have been found with some of the textual features analyzed. It should be said that those correlations express translation process decisions made by professional translators and specialized translation learners. The correlations do not rule out all the ambiguities that may be present in specific occurrences of business lexical item in specialized expressions and compound terms. The correlations are not intended either to exhaust all of the occurrences of business that can be found in expressions and compound terms such as the one found in the corpus studied, especially the senses 2b, 2c and 1b that have been excluded for practical reasons. Their purpose is to cover most of the textual features that are recognizable in the expressions and compound terms to be interpreted in English and eventually translated into French.

A general outline of the correlations is presented in the following figure 7. Conditional decisions number 1 to 7 are mutually exclusive. Their order of application is important since correlations are ordered by certainty. The non-application of a lower number correlation should lead the interpretation to the next level of correlation. The first condition represent the strongest correlation of textual feature to specific sense of business lexical item while the last condition represent the weakest correlation of textual feature to specific senses and the most ambiguous conditions of analysis of textual features in their correlations with specialized business senses. The examples provided are taken from the occurrences found in the corpus created for the purpose of this paper.

1. When business lexical item appears in a specific complex lexical item (such as do business, mean business, business-to-business), it is part of a conventional expression or compound that need to be listed and translated as idiom (with its global meaning).
2. If business lexical item is used in plural, the sense is 3a and is generally associated with a proper name. For some rare occurrences, when the entity has no proper name the sense could also be 2a.
3. If business lexical item is a head noun, the sense may be 2a, 3a or 3c. To distinguish the three potential senses, the following conditions need to be checked.
   1. If the modifier refers to activities known as standard industry activities (insurance for example), the sense of business lexical item is 3c “group of entities sharing similar activities or operations”.
   2. If a possessive is part of the complex term that business lexical item is a constituent of, business sense is either 2a or 3a.
   3. If the head noun has a proper noun otherwise specified in context, if the modifier is a proper name, is an adjective or a generic noun not part of standard industry activities (such as an export business, a family business), the sense of business lexical item is 3a “specific business concern”.
   4. Otherwise, if the modifier refers to such activities that are illegal, parallel or of non-economical nature, the sense is 2a “series of activities for a commercial or economic, or other, purpose”.
4. If business lexical item is a modifier, the sense may be 1a, 2a, 3a or 3b. The following conditions need to be checked.
   1. If the modifier is a determinative modifier and not an attributive modifier, the sense of business lexical item is 3a.
   2. If the head noun refers to objects or circumstances happening in one’s duties or functions (such as traveling or for working purposes), the sense is 1a (a professional occupation).
   3. If the head noun refers to people, functions or concrete projects associated with a specific business entity or concern, the sense is 2a.
   4. If the head noun refers to people, functions or abstract objects or projects not associated with a specific business entity but to an undefined group that implicitly include most if not all business entities of a community, the sense is 3b.
   5. If the head noun refers to academic topics, studies or competencies, the sense is 3b (special use of this sense as an academic discipline or field of study) [this sense also appears in unitary occurrences, see below]
5. If business lexical item is a unitary expression preceded by the determiner the, business sense is 3c.
6. If business lexical item is a unitary expression not preceded by a determiner but preceded by the prepositions on or for, business sense is 1a. When preceded by the preposition in, business sense is 3b.
7. If business lexical item is a unitary expression not preceded by a determiner or a preposition, the sense may be 1a, 2a, 3a, 3b, 3c. This situation describes the most ambiguous occurrences of business lexical item in complex terms.
   1. In the case of 1a, no specific pattern correlate with this sense. More corpus data need to be analyzed in order to make interpretation rules in correlation with textual features.
   2. In the case of 2a, 3a, and 3b, semantic features of referential meaning of the sense seem to be correlated to each of the unitary use of the senses. If preceded by a possessive, business sense is either 2a or 3a (3b and 3c are excluded). In the case of 3b, two interpretations are possible, one based on referential feature for the sense 3b as a group of profit-seeking people, and another one based on contextual features for the sense 3b that was identified as a discipline or academic field of study.
   3. For 3c, the two unitary expressions found in the corpus suggest a contextual reduction or concentration closely related to the passepartout words identified by Gerzymisch-Arbogast (1989). The presence of a correct contextual expansion is a textual feature of this sense used as a unitary expression.

Figure 7. Decision-making in the interpretation of business

The textual correlations to be used in the decision-making process of interpretation seem complex to be implemented in concrete translation applications, although it might be very helpful for learners in specialized translation training courses. However, they do describe all of the textual information to process in the interpretation of occurrences to be translated. What is interesting is that the textual correlations with specific senses are for the most part binary: idiom or non-idiom, plural or singular, as a head noun or as a modifier, as a determinative or attributive modifier, with or without a determiner, etc. This view of decision-making with binary textual information in interpretation is probably central to the decision-making process in translation. Further studies on the translation techniques and decisions that follow the interpretation of lexical items in expressions and compound terms could shed lights on the decision-making processed effectively realized when translating.

11. Conclusion

The first step in the translation process is the active understanding or analysis of source language vocabulary and specialized terms. This paper describes the interpretation process in the translation of the lexical item business that is a constituent of specialized compound terms. As a matter of fact, only eight senses of business at most contribute to the creation of thousands of compounds and compound terms with the lexical item business. The analysis of a small corpus of 301 occurrences of business has even shown that only three of them seem to be common and productive (2a, 3a, and 3b) and that three others (1a, 1b, and 3c) appear to be only marginally represented. Considering the huge amount of compounds and compound terms that can potentially be created (with basic syntactical combination rules), it makes sense to focus the learning of specialized translation on the textual features of the senses of lexical items in compound terms like business in economics, business and finance texts. The advantage of the conditional interpretation rules of conventional senses is that it makes it possible to describe decision-making in translation as an active processing activity involving textual features (contextual and grammatical) relevant in the rendering of lexical item sense. The interpretation rules of simple lexical items for translation purposes also provide crucial knowledge of the interpretation (and translation) techniques that are actively applied to specialized texts, other than the acquisition of the specialized passive knowledge of content matter specialists which translators might assimilate only after practising their trade for a long time or by following a dual career training in translation as well as in their field of specialization.

The small corpus analysis of 301 occurrences has shown correlations of textual features with specific specialized senses of business lexical item in expressions and compound terms. Contextual, syntactical and semantic features of specialized senses contribute to the interpretation of business lexical items and can be correlated with specific senses. The conditional correlations discussed in this paper can therefore be part of the learning activities designed in specialized translation training classes and will guide learners in the adequate English-French translation of business in compound terms in the field of economics, business and finance.

The study of textual uses of compound terms and their translation is strongly dependent on access to rich corpus data. For the interpretation of lexical items in compound English terms, the monolingual corpora created by Mark Davies have been crucial to the analysis of typical textual features of senses of a lexical item in compound terms. For the translation process, it would be of interest to focus on large parallel translation corpora to better describe translation rules and conditions of translation that apply to terms and expressions in specialized texts. To that end, the project of the corpus COMENEGO (Corpus Multilingüe de Economía y Negocios or Multilingual Corpus of Business and Economics) specialized in economics, business and finance directed by Daniel Gallego Hernández (2013a, 2013b) will provide useful evidence.

The approach suggested in this paper for the interpretation of business applies in theory to most of the other lexical items in English commonly used in specialized compound terms. The creation and validation of interpretation tables and textual correlations for this large amount of lexical items represent a huge amount of work. Since it is virtually impossible to accomplish this task for the foreseeable future, there is no reason not to start with most significant keywords in each specialized fields of translation. Of course, the interpretation rules described in this paper fill half the gap in the description of the translation process and competencies. However, it also opens up new horizons in the description of a strongly needed and most welcome representation of translation competencies and offers instructive learning material for the training curriculum in specialized translation.

About the author(s)

Éric A. Poirier, associate professor of translation and program director in the Department of Modern Languages and Translation at the Université du Québec à Trois-Rivières, has worked as a professional translator for 18 years and teaches mainly translation methodology and financial translation from English to French. He is the author of the textbook Initiation a la traduction professionnelle : Concepts clés. His research interests include the empirical evaluation of translated texts and the science of translational data and its applications to translation practices, teaching and learning.

Email: [please login or register to view author's email address]

•  Previous article