Ambiguity matters in linguistics and translation

Ambiguity implies that there are at least two distinct senses ascribed to one sign. It is inherent to language and speech. In this article, I reflect on the types of ambiguity, its typology, production and effect and propose an algorithm for tackling ambiguity in translation. I posit that the choice of a translation strategy and the need for disambiguation in general depend on the type of ambiguity, its sources and character, i. e. whether ambiguity is intended or not. Intended ambiguity occurs when the speaker intentionally does not follow the logic of conceptual clues (primes) and opts for a set of communicative strategies and linguistic means, which allow him/her to offer several possible interpretations of one event or even refer to several different events. I explore a rarely analyzed event-referential ambiguity, which requires additional conceptual information for disambiguation and, consequently, may pose a problem for translation. I argue that problems in disambiguation may occur for a variety of reasons: the translator and\or the recipient may have a wrong reference, have insufficient background knowledge to resolve the ambiguity or make wrong inferences since each recipient bears a different combination of cognitive, axiological, social, professional and gender attributes.


Introduction
To start a discussion on why ambiguity matters (the phrase, which is ambiguous in itself), I would like give a quote from Umberto Eco: 'It sometimes happens that one of my translators will ask me the following question: "I am at a loss as to how to render this passage, because it is ambiguous. It can be read two different ways' (Confessions of a Young Novelist, 2011:33).
Translators face the challenge of ambiguity and its resolution in their daily work since ambiguity is inherent to human communication though often described as a problem, a 'disease of language' . Unlike most language users, who are usually unaware of the presence of ambiguity, translators have to decide how to tackle the problem of simultaneous existence of two (or more) discrete senses or two (or more) possible interpretations when rendering the meaning in the process of intralingual and interlingual translation.
In this article, I will look into the phenomenon of ambiguity, its sources and types, resolution and translation decision-making patterns. More specif-ically, I will analyse ambiguity from the point of view of its a) production and typology; b) perception and effect; c) disambiguation and translation decision-making. I will also offer a Q&A algorithm for tackling ambiguity in translation.

Production and typology
The study of ambiguity and its sources began in Antiquity. It was Aristotle, who wrote: 'There are three varieties of ambiguities and amphibolies: (1) When either the expression or the name has strictly more than one meaning… (2) when by custom we use them so; (3) when words that have a simple sense taken alone have more than one meaning in combination; e. g. 'knowing letters'. For each word, both 'knowing' and 'letters', possibly has a single meaning: but both together have more than one -either that the letters themselves have knowledge or that someone else has it of them (Sophistical Refutations, Stanford Encyclopedia of Philosophy, 2016).
Since then, the study of the 'obscurity of meaning' has never stopped. The problem of ambiguity has been addressed in philosophy (Athrton 1993;Camp 2006;etc.), logic , lexical and cognitive semantics (Apresjian, Bach 1982;Chomsky 2000;Cruse 1986;Dunbar 2001;Fodor 1998;Greenough 2003;Lakoff 1970;Nunberg 1995;Saka 2007;Tuggy 1993, Zwicky&Sadock), pragmatics (Grice 1975;Recanati 2004;Zabotkina 2018), psychology (Barsalou 1982), automated text analysis and natural language processing (NLP) for artificial intelligence (AI) (Degani et al. 2016; Giammarresi, Salvatore & Guy Lapalme 2016), etc. It shows a ubiquitous interest in ambiguity despite its seemingly confusing character. Unfortunately, there have been very few research works in Translation Studies devoted exclusively to ambiguity proper. In the Benjamins Translation Studies Bibliography, out of 214 ambiguity search hits only separate works explore the problem Hendwerker 2004;Pym&Bei 2018;Fougner 2003;etc.). The majority of the articles explore the problem of ambiguity in the translation of literary texts and machine translation.
Yet, there are numerous articles, published recently, focusing on translation ambiguity (which, certainly, does not mean 'ambiguity in translation' but ambiguity as a process of sense identification in NLP and AI) and its consequences for lexical processing and language learning (see Natasha Tokowicz and Tamar Degani 2016).
Following Yu. Apresjan (Apresjan 1974), in Russian linguistics, ambiguity is referred to as neodnoznachnost' (not-single-meaningness); it is differentiated from mnogoznachnost' (polysemy) and neopredelennost' (vagueness) (Zaliznyak 2007). There have been numerous attempts to distinguish polysemy, ambiguity and vagueness. However, there is no evidence that the proposed tests give reliable results, since participants of the tests largely rely on their individual perception of the difference between the three phenomena (Kilgariff 2011).
In common parlance, ambiguity implies more than two interpretations, more than two senses of a word. However, some of the latest lexicographic definitions (for instance, Prinston University's WordNet 3.1) provide some food for thought on the difference between ambiguity and vagueness: ambiguity is defined as 1) an expression whose meaning cannot be determined from its context 2) unclearness by virtue of having more than one meaning, whereas vagueness implies 1) unclearness by virtue of being poorly expressed or not coherent in meaning) "the Conservative manifesto is a model of vagueness"; "these terms were used with a vagueness that suggested little or no thought about what each might convey" 2) indistinctness of shape or character) "the scene had the swirling vagueness of a painting by Turner" From the cognitive and pragmatic points of view, the definitions above present a rather interesting approach to the analysis of ambiguity and vagueness: vagueness is clearly perceived as a negative phenomenon (by virtue of being poorly expressed and not coherent) whereas ambiguity proper is regarded with somewhat greater tolerance (unclearness by virtue of having more than one meaning).
Researchers traditionally distinguish language ambiguity and speech ambiguity (Tuggy 1993). Language ambiguity is the capacity of a word or phrase to have distinct senses, i. e. the property of linguistic units, whereas speech ambiguity is the realisation of this property in an utterance . Some languages are notoriously famous for being polysemous and, consequently, have a much higher potential for ambiguity. According to some authors, over 40 % of English words are polysemous (Traxler, 2012:117). Given the fact that most words in natural languages are likely to be polysemous, i. e. having more than one meaning, ambiguity is not a rarity in language and speech. This brings us to the need to reflect on the types of ambiguity, some of which were also described in (Zabotkina et al. , 2018. From the point of view of language mechanisms, there are lexical, morphological and syntactical ambiguities. The most common, but nevertheless, challenging type of ambiguity is lexical ambiguity. It occurs when ambiguous senses are either a result of polysemy (for instance, words having a metonymic or a metaphoric sense extensions), or a result of homonymy.
The translator has to be aware of the words, which are not only polysemous but intrinsically ambiguous. In linguistic literature, they are called different names -auto-antonyms, or Janus words. They are polysemous words having contrasting and seemingly unrelated meanings, for instance: (Eng.) apparent 1. An apparent situation, quality, or feeling seems to exist, although you cannot be certain that it does exist. There is at last an apparent end to the destructive price war.
2. If something is apparent to you, it is clear and obvious to you. It has been apparent that in other areas standards have held up well.
The presence of a star is already apparent in the early film.
Auto-antonyms pose a problem for both translation and interpreting, particularly when they occur at the beginning of a sentence and the interpreter, unlike the translator, has to identify and process contextual clues for disambiguation during a fairly short period of time and without having the context necessary for disambiguation. Consider, for instance, a piece of transcript of a business meeting and reflect on the scanty conceptual information available for the interpreter\translator preceding and following the autoantonym apparently as well as the general ambiguity of the excerpt: 'I think one of the things we discussed in the branch action group meeting is actually, weren't quite sure what the (pause) cos (laughing) we'd been sending out (unclear) for quite less than ten of the branches because (pause) to (pause) do something to them which was (pause) mm (pause) apparently lots of them in but erm (pause) well they keep disappearing so that the (pause) what we've been talking about is to try and get some points (pause) erm (pause) health and safety station (pause) within the branch, which actually just has all this stuff for it (Available at https://www.english-corpora.org/bnc/). Idioms provide numerous examples of lexical ambiguity, especially those having a lower use frequency. They pose a problem since they can be easily mistaken for their literal readings, for instance, she is in the driver's seat may easily by interpreted as somebody has taken the front seat opposite the steering wheel or somebody who is in control. One may jump to conclusions without realizing the sentence is ambiguous as both interpretations are plausible ones.
Syntactic (structural) ambiguity develops from an ambiguous structure of an utterance. All types of texts offer numerous examples of structural ambiguity including excerpts from the Bible: "And all the people saw the thundering, and the lightning, and the noise of the trumpet, and the mountain smoking: and when the people saw it, they removed, and stood afar off" (The Bible). The pronoun it may refer to either of the three events thundering/lightning/noise or the whole situation. This type of ambiguity is often referred to scope ambiguity. This type is is closer on the gradient scale to vagueness, since the scope of conceptual information activated is ample, non-homogenous and changeable depending on the presented stimuli. However, this type is a frequent occurrence and poses relatively little difficulty for translation.
Another type of ambiguity is a morphological one when one part of speech can be used instead of another (see the title of this article). In the phrase, no road works 'works' can morphologically be interpreted as a plural noun or a verb. Consequently, the sentence can be interpreted as there is no repair works or none of the roads is in good condition to be used.
Transformational ambiguity, introduced by J. Lyons (Lyons 1975) and reflected on by N. Chomsky (Chomsky 1957), is a type of ambiguity that exists, in most cases, only out of context, for instance, 'the shooting of the hunters' (ibid).
The translator is to be aware of another type of ambiguity, which is seldom analysed in linguistic literature -event-referential ambiguity. I hold there are at least two types of it. The first one is a possible reference to two or more events in one context resulting from insufficient or poor conceptual stimuli: Event-referential ambiguity may also result in another scenario -several possible interpretations of one single event depending on the point of view: Re-election of the chairman -1) election of the same person for another term or 2) election of another person; In the case described above, the event is the same -that of the election (or re-election) of the chairman. However, there could be two different interpretations of one event. This type of ambiguity occurs due to a variety of reasons, ranging from conceptual, semantic (polysemy proper) to purely pragmatic ones -prevalence of individual pragmatic factors in the process of disambiguation.
Event-referential ambiguity may be generated by separate words (polysemy and homonymy), ambiguous morphological and syntactic structures. It may occur due to the poverty of conceptual stimuli, or appear as ambiguity of the whole sentence or text as is the case of scope and transformational ambiguity.
From the pragmatics point of view, the translator has to differentiate between intended and unintended ambiguity. Intended ambiguity is aimed at creating a specific pragmatic effect. It has its bearing on the way the source text is understood, interpreted and perceived by the recipient. The translator has to differentiate between the intended ambiguity of the speaker (author) and that of the translated text as such.
There are certain types of discourse that are particularly demonstrative of intended ambiguity: advertising, politics, minority issues and diplomacy. The latter is notoriously famous for ambiguities as one can see from a much quoted aphorism: when a diplomat says yes, he means perhaps; when he says perhaps, he means no; when he says no, he is not a diplomat.
Depending on the intention of employing ambiguity in the abovementioned types of discourse, one can identify two subtypes of ambiguitystrategic or 'constructive' and manipulative. However, broadly speaking, any kind of intended ambiguity can be considered manipulative since the speaker when generating an utterance aims at achieving a particular goal or effect. Yet, strategic ambiguity can be defined as a deliberate choice of words or language structures allowing the freedom to alter interpretations, which, however, may produce the most positive effect in pursuing a positive goal: 'The U. S. policy towards the unification of China and Taiwan has been described as a policy of 'strategic ambiguity', one that allows the U. S. to be nonspecific in its assertions about the status of Taiwan (British National Corpus).
Lately, strategic ambiguity has been more and more often considered as a communicative strategy whether it is for better or for worse: 'As a strategy, constructive ambiguity sounds like an ornate euphemism for "sitting on the fence", but this is the label some have applied to the Brexit position of the UK's opposition Labour party…… Of course, the party itself is not openly calling it constructive ambiguity, as it tiptoes around the increasingly desperate negotiations between Theresa May's government and the EU, and time rapidly runs out before the UK's deadline to leave the union in six months' time (https://www. thenational. ae/opinion/comment/how-the-uk-labourparty-has-turned-brexit-ambiguity-into-a-clear-advantage-1.776666).
Interestingly enough, there is a difference between being ambiguous and being strategic about ambiguity. Take, for instance, the following example of scope ambiguity from business discourse and think of its possible meaning and translation options: "Well, we don't want to limit ourselves, so we're going to try to be all things to all people" (https://www. marshallstrategy. com/think-big-understanding-thevalue-of-strategic-ambiguity/). Manipulative ambiguity proper is aimed at achieving a goal, which is not morally or otherwise right. One of the most common techniques of achieving manipulative ambiguity is a deliberate choice of an 'ambiguous' communicative strategy and the corresponding lexical units. The analysis of numerous examples shows that one of such strategies is averaging and generalisation, for example: 'The increase in secondary school enrolments in DET schools has far outstripped the rate of population increase. Despite some expansion of the system, classes are still very large, on average forty or more, and facilities are overstretched' (British National Corpus).
Deliberate use of lexical means as the ones in the sentence above -'On average forty or more', 'far outstripped' and 'some' -leaves the recipient certain space for a conceptual manoeuver, which was obviously the intention of the speaker.
On the one hand, the use of the generalisation and averaging strategies results in pragmatic hedging used to protect the speaker from possible criticism concerning the use of unverified or inaccurate data. Naturally, these strategies lead to multiple interpretations of the event depending on the interlocutor's pragmatic attitude.
A similar effect is achieved by the use of linguistic means of expressing modality since some types of modality (for instance, probability) provide an ample opportunity for conceptual and interpretive options, as shown in the examples below: 'So it's hardly likely they're going to write off their star when the ink is barely dry on his $50 million contract!' (https://corpus.byu.edu/iweb/). All strategies and techniques of achieving intended ambiguity create an opportunity for manipulating public opinion since ambiguity provides several reference points. The multiplicity of reference points for the formation of a judgment creates a favourable environment for the emergence of ambiguity.
Unintended ambiguity can be a result of the speaker's unawareness of the effect produced. However, from a textual point of view, this lack of clarity, uncertainty is 'rich in connotations and innuendos that are very fruitful for the overall textual strategy' (Eco 2011:34).
Unintended ambiguity may result from a low degree of language proficiency since non-native speakers often use words either in a wrong context, or in a wrong meaning. It definitely poses a problem for the translator, especially if the translator is asked to 'proofread the text' written in a foreign language, which is often the case.

Perception and effect
From the point of view of perception and effect, ambiguity can 1) complicate the understanding of a message or 2) can be enjoyed by the recipient.
A fairly recent research work by a team of German psychologists proves that people do not need to fully understand a painting or any artwork to like it. Their experiments showed that "the higher the subjectively perceived degree of ambiguity within an artwork, the more participants liked it, and the more interesting and affecting it was for them." The most striking result: "The higher participants assessed the ambiguity of a stimulus, the more they appreciated it" . This might be also true of different types of discourse other than that of visual art. Ambiguity may have a positive effect on the recipient as he/she may enjoy the elegance of expression or its wit.
However, many researchers promote the principle of avoidance of polysemy and, hence, ambiguity. One of Grice's maxims also advises to avoid ambiguity at all costs (Grice, 1975), meaning the translator has to disambiguate since ambiguity may result in a communication failure. I argue that this may happen for a variety of reasons: having come across an ambiguous word (phrase, sentence, etc.), the translator and\or the recipient may: 1) have a wrong reference; 2) have insufficient background knowledge to resolve it; 3) make wrong inferences since each recipient bears a different combination of cognitive, axiological, social, professional and gender attributes.
Ambiguity can be perceived and not perceived. In other words, the translator or the recipient may remain unaware of the presence of ambiguity. When generating an utterance, the speaker chooses words according to his/her communicative and pragmatic intention. Polysemy (or homonymy, for that matter), as it may seem, is a problem for the recipient only (as it was shown above) since the recipient has to adequately infer the intended sense of the ambiguous word in a given context.
In everyday discourse, both the speaker and the listener are unaware of the presence of polysemy. They do not realize the complexity of the structure of polysemous words and the entire scope of conceptual information encoded by them. The identification of the meaning of a polysemous word occurs so effortlessly that polysemy is perceived as monosemy . Consequently, ambiguity may go unnoticed.
Another interesting kind of ambiguity is the one induced by cultural traditions, for instance, in some Asian countries, where it is easier to be ambiguous to 'save face' rather than to simply say no or give a straightforward negative reply.
The translator is to be aware of ambiguity, its effects and the possible difficulties of its resolution. The choice of strategy obviously depends on whether ambiguity is intended or not and on the availability of translation means in the target language.

The problem of disambiguation
In this section, I will look into the problem of the resolution of ambiguity. The prevailing point of view in traditional semantics and in translation is that ambiguity needs to be resolved one way or another. However, language being an effective tool for communicating ideas, emotions etc. often plays practical jokes and easily shifts meanings.
Having faced the puzzle of ambiguity, the translator has two options. He\she can either disambiguate polysemous words or exploit them for conversational (meaning pragmatic) profit, and this is in spite of the fact that keeping multiple meanings in mind has some cognitive cost (Nerlich 2003).
Disambiguation is triggered by context. Broadly speaking, by 'context' I understand any type of conceptual, linguistic and extralinguistic information available at the moment of translation decision-making. If ambiguity occurs as a result of polysemy, then the presence of clues of one meaning or the absence of clues (primes) of a different meaning (or meanings) is a tool of disambiguation. Conceptual primes facilitate the process of disambiguation. Each word is associated with a set of dynamic cognitive contexts that store a significant amount of conceptual information, referring to any number of conceptual domains that are relevant to the identification of a particular sense of the word. Consider, for instance, the following example: 1. "While a central bank-backed digital currency could pose a disruptive threat to current bank operations, Citi views this as an improbable "long-tail" risk" (https://corpus. byu. edu/bnc/).
2. The Premier's success in his role as President of the European Community came as America prepared to carry out threats of trade sanctions on EC imports (https://corpus.byu.edu/bnc/). Two different conceptual primes in (1) and (2) -pose a disruptive threat and prepared to carry out threats -facilitate the process of disambiguation between the two senses of the noun: something that is a source of danger (1) and a warning that something unpleasant is imminent (2).
However, there are contexts, which are intrinsically ambiguous since they are not informative enough about their meaning in general or the meaning of the word in question: (1) 'We attended the same school' may mean, for instance, 'we were classmates' or 'we studied in the same school but were not classmates'; (2) 'She is plain' may be interpreted as ' she is ugly' or 'she is simple'; (3) 'The empirical data sources are included alongside the other references in alphabetical order' is definitely confusing since 'alongside' means both 'next to' and 'together' (https://dictionary.cambridge.org/dictionary/english/alongside). Example -Lege Artis Journal. Guidelines for authors. Available at https:// lartis.sk/wp-content/uploads/2019/04/LArtGuidelines_for_authors.pdf).
The examples given above demonstrate that insufficient context or poor cognitive stimuli may result in translation errors. Hence, this topic requires further research employing methods of the cognitive paradigm.

Conclusions
The main point I attempted to argue is the need to be aware of ambiguity as a phenomenon of language and speech as well as the rich variety of its types leading to numerous options in the translation decision-making process. There are at least ten questions the translator has to answer before deciding on what to do about ambiguity: 1. What type of ambiguity is it? 2. How it is achieved? 3. Is the ambiguity intended or not? 4. If the ambiguity is intended, is it constructive or purely manipulative? 5. Is it intended ambiguity of the author or intended ambiguity of the text as a whole? 6. Is the whole text ambiguous or are there only separate elements, which are perceived as ambiguous? 7. If the ambiguity is unintended, why does it occur? 8. Does the unintended ambiguity need disambiguation? 9. Are there linguistic means in the target language to express the source language ambiguities? 10. What conceptual, linguistic and extralinguistic information is needed to approach the problem?
In the majority of cases, speakers are aware of ambiguity provided it is intended. Intended ambiguity occurs when the speaker intentionally does not follow the logic of priming since each meaning of a polysemous word is primed differently.
The multiplicity of interpretations created by event-referential ambiguity may present a problem for translation and requires a broader cognitive context for its successful disambiguation. Poor cognitive stimuli, insufficient cognitive context available at the moment of disambiguation may result in translation errors.