CC BY
9Столыпинский вестник №4/2023
Научная статья Original article УДК 81-25
ПРОБЛЕМЫ, СВЯЗАННЫЕ С ПЕРЕВОДОМ УСТНОЙ РЕЧИ В ПИСЬМЕННУЮ ЧЕРЕЗ СУБТИТРИРОВАНИЕ
PROBLEMS ASSOCIATED WITH THE TRANSMISSION OF ORAL SPEECH INTO WRITTEN THROUGH SUBTITLING
Ямалтдинова Диана Рустемовна, магистрант Института мировой экономики и бизнеса Российского университета дружбы народов, dianayama20@gmail. com
Сибул Виктория Владимировна, кандидат филологических наук, доцент, профессор кафедры иностранных языков экономического факультета Российского университета дружбы народов, [email protected]
Diana R. Yamaltdinova, master's student of the Institute of World Economy and Business of the Peoples' Friendship University of Russia, [email protected] Viktoria V. Sibul, candidate of of Philological Sciences, associate professor, professor of the Department of Foreign Languages, Faculty of Economics, Peoples' Friendship University of Russia, [email protected]
Аннотация. В статье рассматриваются изменения, связанные с переводом устной речи в письменную через субтитрирование. С развитием онлайн-технологий проблема субтитрирования аудио-визуальных выступлений
Столыпинский вестник
1777
становится особенно актуальной. Авторы дают определение тексту, а также его письменной и устной форме. Приводятся основные черты устной и письменной речи. Так, письменная речь отличается грамматичностью, автономностью, связностью. Устная речь характеризуется, в первую очередь, спонтанностью и аграмматичностью. В ней частотны повторения, паузы, оговорки, незаконченные предложения и мысли, слова-филлеры. Более того, устная речь характеризуется относительно короткими предложениями и грамматическими конструкциями, а также употреблением простых союзов. Авторами анализируются несколько выступлений на экономическую тематику, в речи спикеров выделяются черты устной речи. Далее авторы определяют различные подходы к переводу устной речи в письменную посредством субтитров. Авторами делается вывод о том, что наиболее частотными подходами при работе с субтитрами являются опущения, исправление грамматических форм (нейтрализация разговорных форм), а также нулевой подход, при котором особенности устной речи остаются в субтитрах.
Abstract. The article considers the changes associated with the transmission of oral speech into written through subtitling. With the development of online technologies, the problem of subtitling audio-visual content becomes especially relevant. The authors define the text, as well as its written and oral form. The main features of oral and written speech are given. Thus, written speech is distinguished by grammaticality, autonomy, and coherence. Oral speech is characterized primarily by spontaneity and agrammaticality. It also may contain repetitions, pauses, false starts, unfinished sentences and thoughts, filler words. Moreover, oral speech is characterized by having relatively short sentences and grammatical constructions, as well as the use of simple conjunctions. The authors analyze several speeches on economic topics, highlighting features characteristic for oral form. Furthermore, the authors define various approaches to transmitting oral speech into written by means of subtitles. The authors conclude that the most frequent methods used when creating
1778
subtitles are omissions, correction of grammatical forms (neutralization of colloquial forms), as well as the "do nothing" approach, i.e. when features of an oral speech remain in subtitles.
Ключевые слова: субтитрирование, устная речь, письменная речь, дискурс Keywords: subtitling, oral speech, written speech, discourse
INTRODUCTION
Without the internet, modern life is unimaginable. Our entire existence—our relationships with friends, family, and employers—take place online. It is understandable why a number of forums and conferences began to operate exclusively online. Additionally, everyone in the world had to get online in order to take health precautions when the COVID-19 epidemic struck. As a result, instead of having everyone in one place and limiting attendance, most forums are now hosted over Zoom or another broadcast application. The ability to hear from subject matter experts without traveling to a specific location has made conferences more accessible.
Speaking of accessibility, subtitles are a fantastic method to spread your message so that more people can understand it. They are a lifesaver for those who might not be native speakers of your language, are not specialists, or struggle with understanding some complex vocabulary or syntax. They also assist should there be any technological difficulties. So, the subject of the oral speech that is being subtitled is pertinent.
A subtitle, which takes the shape of two lines of text at the bottom of the screen, is a shortened version of the original oral speech. When the speaker makes the corresponding utterance, the subtitles appear and vanish at the same time. The audio of the original video and the visual content of the subtitles are both concurrently perceived by the audience when something is subtitled.
1779
APPROACHES TO THE TEXT NOTION
A text is a result of a speech creation process that is characterized by being complete, assigned to a specific genre, having a title and a number of units, interconnected by the lexical, grammatical, logical and syntactical means; and a pragmatic purpose. (Galperin, 2005) Galperin suggests that a text should always be presented in the written form to be considered a text, while an oral speech is complete opposite to its written counterpart. Thornbury, on the other hand, characterized a text as a continuous piece of spoken or written language, especially one with a recognizable beginning and ending.
According to Horowitz and Samuels (1987), oral language is characterized by the face-to-face exchange and is adapted to a specific audience and extralinguistic settings. Meanwhile, written language is planned and future-oriented, instead of focusing on the "in the moment" interaction.
THEORETICAL BACKGROUND
Even though both oral and written speech are similar in a way that they are cohesive and coherent, have a communicative purpose and are (generally speaking) complete, they have a number of significant differences.
First of all, written speech can be characterized as being more grammatical. Written text does not have the luxury of being spontaneous, it is written down, reread and edited. Thus, it is "perfect" in nature. (Dilmurodova, 2022) It has no reason to have any lexical, grammatical or syntactical irregularities, as well as fragmented sentences. When writing down a speech, the author tries to choose the best words possible as to make the intended thought as clear as possible.
Most written sentences have the subject-predicate structure. The analysis of oral discourse shows that the same sentence in oral speech could be stated in the same order but have a different meaning or have the same meaning but be stated in a different order, all with a use of emphasis, which is not as explicitly present in the written form (Fanelli, 2018).
1780
Moreover, written texts have autonomy. While oral speeches generally require viewers/listeners and some sort of reaction, written texts are self-sufficient in a sense that they are created for a broad audience that cannot and are not expected to give an immediate feedback. Situational context is absent and the phrases can take meaning by themselves.
Furthermore, written texts can be characterized as being more coherent. It is true that both oral and written texts are coherent (otherwise they hardly would be able to be considered a text), yet oral speech suffers from the overuse of filled pauses, vague language and simple connectors, all of which make it much less coherent than a written version that is usually stripped of those aspects. While a "stripped down" text is easier to read, if that stripped down version was said out loud, it may not necessarily be easy to listen to. Without the pauses, repetitions and false starts, the information load becomes much more compressed which makes it harder to comprehend just by listening.
If we talk in detail about the features of the oral speech, we can name spontaneity, expressiveness and context-dependency as the main ones. It is important to note that oral language has many more features that will not be addressed to in this study as they are irrelevant for the purpose of this research.
Let us take a look at spontaneity and agrammaticality. While delivering an oral speech, a speaker does not necessarily have a strict script to follow. And even if they do, the idea may get developed in the process of speaking. The idea that is being expressed, thus, gets to control how the sentences are modeled. If a speaker does not read their entire speech of the script, it is likely that the sentence structure and the word combinations will not be grammatically correct 100% of the time. Mistakes mostly go unnoticed in a natural conversation and we can only realize how common they are while reading a direct transcript (i.e. "There's another secretary too who I don't know what she's responsible for").
1781
Most common "mistakes" of spoken speech are verbless clauses, ellipses (including omitted pronoun subjects), lack of concord and omitted relative particles ("there's a few problems are likely to crop up"), false stats, slips of the tongue and changes of direction midstream in a grammatical structure ("if you like we could there's food in the fridge why don't we could have something if you're hungry"), etc. (McCarthy, 1991) And while those imperfections can be fixed before they get to the reader in the written text, oral speech cannot be edited.
Other features that may not necessarily be considered a mistake but that would look out of place in a written text are:
• repetitions;
• disfluencies (stops and starts, pauses);
• clause-by-clause and phrase-by-phrase construction; formulaic language ("chunks") (You should've heard him; can't abide the thought of; and he said; and I said)
• frequent use of expressions such as "you know", "well", "oh", "umm" (filled pauses) as well as instances of vague language or filler words such as "sort of", "or something" and "and stuff"
• discourse markers (well, yeah - as a response; I mean - there will be a clarificaition; but - the clarification contrasts with what was said; you know - shared knowledge) which equals to or can be referred to a cohesive device
Futhermore, many sentence and clauses are connected by relatively simple connectors "and", "but" and "because". Native speakers are also prone to use such constructions as: "The problem is is that..."
Two more features of oral speech cannot be separated from one another: they are expressiveness and context-dependency. The perception of the oral speech largely depends on the emphasis that the speaker puts on certain words (word combinations), the emotion with which they speak, their tone, facial expressions, gestures. This communicative behavior goes beyond just the speech spoken. Most
1782
of these features cannot be recreated in the written mode without adding descriptions or emotionally colored vocabulary and removing deictic references. The transcript of an oral speech may contain references impossible to decode without visual and audial information. Also, on the grammatical level spoken narratives can also be characterized as having more vocatives, expletives, exclamations and abbreviations. Moreover, we can't forget that any oral presentation comes with anxiety and/or excitement. No speaker is ever completely neutral while presenting, which can influence the delivery of the speech.
In addition, most oral speeches follow the rule of "one-clause-or-phrase-at-a-time". When a reader is going through a written text, they have the ability to go back and re-read some parts that they may not have understood right the first time. But one cannot do that with an oral speech. This is why while talking presenters usually try not to use complicated and long sentences: to ease the comprehension process of what is being said. Speech is produced in smaller units that do not form any larger units and are often linked by simple conjunctions.
Oral speech can also be characterized by having sentence slots. Oral utterances usually have some sentence "slots" that would be considered a mistake in a written text. Some common slots would be: a tail ("...so."), question tags (isn't he?), adverbials that convey speaker's attitude (actually, really), vagueness expressions (and that sort of thing) and topic clarifiers (It needs a bit of prod, that fire). A similar type of slot can also be inserted at the middle of a sentence to announce the general topic. (Analogue phones. Don't talk to me about analogue phones). (Thornbury, 2005)
MATERIALS AND METHODS
Material of the research is represented by the subtitled videos of speeches on economic topics. The research was done mainly by the means of deductive analysis of gathered examples. The main aim of the research is to study how the same exact speech changes while being transformed in the form of subtitles. For the purposes
1783
of the research 15 videos presenting speeches on economic issues with a total runtime of about 650 minutes were selected and analyzed.
STUDY AND RESULTS
Based on the material of the research we can determine some key aspects of the oral speech. Those examples include:
Repetitions, filled pauses and discourse markers
• so at least when we we talk about bitcoin there are some trust issues, there have been some hacks and uh... so there is obviously a need to to work on on trust and feeling that its uh... its the safe technology
• the current state is in this. the research state and it's being uh developed to to be applied in many many areas... it might be used almost everywhere to. where, like, we won't even imagine how we lived without it but I'm I'm not sure where it's going to go, I mean, no one really knows, right.
• I mean I would assume this technology is taking away business or activity from these middlemen
• Well, we have a lot of uh public blockchains, like, we have bitcoin, we have the ethereum network, nxt, but a lot of companies and consortiums are getting together to build private blockchains, so, ones that are more closed off at first [...] But in the real world projects in energy and in pharmaceuticals and in retail and lots of different fields are starting experiments
Disfluencies, changes of direction, ellipses
• Um well I mean I've heard of it and I trust it a lot because there's. they have like all their guarantees so I. I know that I'm gonna get money
• So you could kinda like. I don't know, say, trade . you trade apples but you would just see, like, random letters for it so you wouldn't be able to, like, track it, i guess?
1784
• but we're actually seeing a lot more use cases for blockchain that aren't around the currency side they're more around how do you take any asset and be able to trade that using the same technology
• whenever we have a transaction and let's say I buy something from you this information gets logged and it gets verified by a third person or third party
• the way that the technology is changing nothing's gonna be like it used to be. there's no, like, firm and then the buyer and the seller and it's basically. we'll have to rewrite a lot of rules and economics as well
• I mean, just do. like, as an example . one thing that I think about a lot in terms of possible blockchain applications is electricity
• if I could trade with any kid, I would trade. well, I would trade something I don't like so much
Vague language
• we have a lot of companies that sort of help us make sure that our trades happen
• so it's kinda this like really big ledger or accounting system for all sorta things that get traded
• a lot of people think about blockchain as bitcoin because it's sorta in the news a lot and it's this new cryptocurrency and it's kinda exciting
Simple connectors
• so you can't really see exactly what happened but you know it happened because it's like a marker
• Well, we have a lot of uh public blockchains [.] but a lot of companies and consortiums are getting together to build private blockchains, so, ones that are more closed off at first and then may evolve into a public network when people feel comfortable using it . And some are also proof of concepts . But in the real world projects in energy and in pharmaceuticals and in retail and lots of different fields are starting experiments and we'll see in the next few years how all of those interact
1785
and what we learn about. how the best use cases for blockchains and what it means for trade.
Tail
• so, education will be a big part before we actually can use the technology very in in the wide sense, right.
• I'm not sure where it's going to go, I mean, no one really knows, right.
• We've had a lot of PR , a lot of proof of concept, but truthfully this is more like a science, right.
Most of these examples come from the speech given by Bettina Warburg, co-founder of Animal Ventures, political scientist and blockchain researcher. The video was posted on the WIRED YouTube channel and is called "Blockchain Expert Explains One Concept in 5 Levels of Difficulty".
When compared to the subtitled versions of the same phrases we discovered several approaches to different aspects of the oral speech.
Omission: filled pauses, repetitions and discourse markers were often omitted. It is no wonder why: it is extremely difficult to comprehend a written utterance when it is filled with "um"s and unnecessary repetitions. Discourse markers, on the other hand, sometimes managed to stay in the text, since in some cases they play a cohesive role, but in no more that 50% of times.
Simple connectors and tails are not really considered a mistake and, thus, do not require any special treatment. Nevertheless, tails are often omitted in order to reach an optimal reading speed.
For example, compare:
1786
Table 1: Omissions in the subtitles
Utterance Subtitle
So at least when we we talk about bitcoin there are some trust issues, there have been some hacks and uh... so there is obviously a need to to work on on trust and feeling that it's uh... it's the safe technology So at least, when we talk about bitcoin, there are some trust issues. There have been some hacks, so there is obviously a need to work on trust and feeling that it's the safe technology.
the current state is in this. the research state and it's being uh developed to to be applied in many many areas. it might be used almost everywhere to. where, like, we won't even imagine how we lived without it but Im I'm not sure where it's going to go, I mean, no one really knows, right. We are currently in the research state and it's being developed to be applied in many areas. It might be used almost everywhere and we won't even be able to imagine how we lived without it. I'm not sure where it's going to go, no one really knows.
Stay as-it is: disfluencies, changes of direction and ellipses in most cases are left just as they are in speech. This is also easily explained: subtitler's job is to mimic the spoken utterance as adequately as possible and they cannot add content to make a sentence sound "better" or more complete, as by doing so they would change the author's intent and flow of thoughts. Most times a simple ellipsis sign suffices in place of an abrupt stop or the start of a new thought.
For example compare:
Table 2: Disfluencies in subtitles
Utterance Subtitle
the way that the technology is changing nothing's gonna be like it used to be. there's no, like, firm and then the buyer and the seller and it's basically. we'll have to rewrite a lot of rules and economics as well The way that the technology is changing, nothing's going to be like it used to be. There's no firm or the buyer or the seller. It's basically. We'll have to rewrite a lot of rules and economics, as well.
I mean, just do. like, as an example. one thing that I think about a lot in terms of possible blockchain applications is electricity I mean, just. As an example, one thing that I think about a lot in terms of possible blockchain applications is electricity.
1787
"Grammaticize": vague language (such as kind of and sort of) is usually pronounced in it's informal form: sorta and kinda. Such forms are not welcome in subtitles, and that is why they are usually made more grammatical (sort of and kind of).
For example compare:
Table 3: Grammaticality in subtitles
Utterance Subtitle
I know that I'm gonna get money I know that I'm going to get money
the way that the technology is changing nothing's gonna be like it used to be The way that the technology is changing, nothing's going to be like it used to be
so it's kinda this like really big ledger or accounting system for all sorta things that get traded So it's kind of this really big ledger or accounting system for all sort of things that get traded
a lot of people think about blockchain as bitcoin because it's sorta in the news a lot and it's this new cryptocurrency and it's kinda exciting A lot of people think about blockchain as bitcoin because it's sort of in the news a lot, it's this new cryptocurrency and it's kind of exciting.
Moreover, when a speaker uses never-ending connectors it is important to stop a sentence at one of those points to make it readable and much easier to understand.
CONCLUSION
While both oral and written texts must be cohesive and have a communicative purpose, they have a number of significant differences. Oral language can be characterized as a "here and now" language, it is interpersonal, spontaneous, situational; it uses such features as ellipsis, repetitions, simple structures, paralinguistic clues, etc. On the other hand, written language is not space or time bound, it is objective and distanced, edited, more grammatical and explicit. Written texts are linear, while oral speeches are a dynamic, unstable process.
1788
When transforming an oral speech into the form of subtitles several approaches
are used, the most common ones being omission, leaving it as it were in oral speech
and formalizing it.
Литература
1. Гальперин И. Р. (2005) Текст как объект лингивстического исследования, Москва, Россия
2. Ямалтдинова Д.Р. (2020) Псевдоустность и особенности аудиовизуального перевода под субтитры, материалы 74-ой международной студенческой научной конференции СпбГУАП
3. Assis R.(2001) Alexandra, Features of Oral and Written Communication in Subtitling in (Multi)Media translation: Concepts, Practises and Research, Edited by Yves Gambier and Henrik Gottlieb, pp 213-223
4. Fanelli, F (2018) Written Discourse, September 2018
5. Georgakopoulou, P. (2009) Subtitling for the DVD Industry in Audio-visual Translation: Language Transfer on Screen, Edited by Jorge Diaz Cintas and
6. Horowitz, R & Samuels, S. (1987). Comprehending oral and written language. Bingley, England: Emerald Group.
7. McCarthy, M. (1991). Discourse analysis for language teachers. Cambridge, UK: Cambridge University press.
8. Thornbury, S. (2005). Beyond the sentence. Cambridge, UK: Cambridge University Press.
References
1. Assis R.(2001) Alexandra, Features of Oral and Written Communication in Subtitling in (Multi)Media translation: Concepts, Practises and Research, Edited by Yves Gambier and Henrik Gottlieb, pp 213-223
2. Fanelli, F (2018) Written Discourse, September 2018
3. Galperin, I. R. (2005). Tekst kak ob'ekt lingvisticheskogo issledovanija [Text as an object of linguist^ research]. Moscow, Russia: Editorial.
1789
4. Georgakopoulou, P. (2009) Subtitling for the DVD Industry in Audio-visual Translation: Language Transfer on Screen, Edited by Jorge Díaz Cintas and
5. Horowitz, R & Samuels, S. (1987). Comprehending oral and written language. Bingley, England: Emerald Group.
6. McCarthy, M. (1991). Discourse analysis for language teachers. Cambridge, UK: Cambridge University press.
7. Thornbury, S. (2005). Beyond the sentence. Cambridge, UK: Cambridge University Press.
8. Yamaltdinova, D. (2020) Pseudoorality and the features of subtitling, materials of the 74 international student scientific conference of the Saint Petersburg State University of Aerospace Instrumentation
© Ямалтдинова Д.Р., Сибул В.В., 2023 Научный сетевой журнал «Столыпинский вестник» №4/2023.
Для цитирования: Ямалтдинова Д.Р., Сибул В.В. PROBLEMS ASSOCIATED WITH THE TRANSMISSION OF ORAL SPEECH INTO WRITTEN THROUGH SUBTITLING // Научный сетевой журнал «Столыпинский вестник» №4/2023.
1790
