Can computers outperform human interpreters?

Unlike many people in the translation industry, I like to imagine that one day computers will be able to interpret simultaneously between two languages just as well as or better than human interpreters do, what with artificial neuronal neurons and neural networks‘ pattern-based learning. After all, once hardware capacity allows for it, an artificial neural network will be able to hear and process many more instances of spoken languages and the underlying content than my tiny brain will in all its lifetime. So it may recognise and understand the weirdest accents and the most complicated matter just because of the sheer amount of information it has processed before and the vast ontologies it can rely on (And by that time, we will most probably not only be able to use digital interpreters, but also digital speakers).

The more relevant question by then might rather be if or when people will want to have digital interpretation (or digital speakers in the first place). How would I feel about being replaced by a machine interpreter, people often ask me over a cup of coffee during the break. Actually, the more I think about it, the more I realise that in some cases I would be happy to be replaced by a machine. And it is good old Friedemann Schulz von Thun I find particularly helpful when it comes to explaining when exactly I find that machine interpreters might outperform (out-communicate, so to say) us humans (or machine speakers outperform humans).

As Friedemann Schulz von Thun already put it back in 1981 in his four sides model (, communication happens on four levels:

The matter layer contains statements which are matter of fact like data and facts, which are part of the news.

In the self-revealing or self-disclosure layer the speaker – conscious or not intended – tells something about himself, his motives, values, emotions etc.

In the relationship layer is expressed resp. received, how the sender gets along with the receiver and what he thinks of him.

The appeal layer contains the desire, advice, instruction and effects that the speaker is seeking for.

We both listen and speak on those four layers, be it on purpose or inadvertently. But what does that mean for interpretation?

In terms of technical subject matter, machine interpretation may well be superior to humans, whose knowledge base despite the best effort will always be based on a relatively small part of the world knowledge. Some highly technical conferences consist of long series of mon-directional speeches given just for the sake of it, at a neck-breaking pace and with no personal interaction whatsoever. When the original offers little „personal“ elements of communication (i.e. layer 2 to 4) in the first place, rendering a vivid and communicative interpretation into the target language can be beyond what human interpretation is able to provide. In these cases, publishing the manuscript or a video might serve the purpose just as well, even more so in the future with increasing acceptance of remote communication. And if a purely „mechanical“ translation is what is actually needed and no human element is required, machine interpreting might do the job just as well or even better. The same goes e.g. for discussions of logistics (“At what time are you arriving at the airport?”) or other practical arrangements.

But what about the three other, more personal and emotional layers? When speakers reveal something about themselves and listeners want to find out about the other person’s motives, emotions and values or about what one thinks of the other, and it is crucial to read the message between the lines, gestures and facial expressions? When the point of a meeting is to build trust and understanding and, consequently, create a relationship? Face to face meetings are held instead of phone calls or video conferences in order to facilitate personal connections and a collective experience to build upon in future cooperation (which then may work perfectly well via remote communication on more practical or factual subjects). There are also meetings where the most important function is the appeal. The intention of sales or incentive events generally is to have a positive effect on the audience, to motivate or inspire them.

Would these three layers of communication, which very much involve the human nature of both speakers and listeners, work better with a human or a machine interpreter in between? Is a human interpreter better suited to read and convey personality and feelings, and will human interaction between persons work better with a human intermediary, i.e. a person? Customers might find an un-human interpreter more convenient, as the interpreter’s personality does not interfere with the personal relation of speaker and listener (but obviously not provide any facilitation either). This “neutral” interpreting solution could be all the more charming if it didn’t happen orally, but translation was provided in writing, just like subtitles. This would allow the voice of the original speaker to set the tone. However, when it comes to the „unspoken“ messages, the added value of interpreters is in their name: They interpret what is being said and constantly ask the question „What does the speaker mean or want?“ Mood, mocking, irony, reference to the current situation or persons present etc. will most probably not be understood (or translated) by machines for a long time, or never at all. But I would rather leave it to philosophers or sci-fi people to answer this question.

About the author:
Anja Rütten is a freelance conference interpreter for German (A), Spanish (B), English (C) and French (C) based in Düsseldorf, Germany. She has specialised in knowledge management since the mid-1990s.

Hello from the other side – Chinese and Terminology Tools. A guest article by Felix Brender 王哲謙

As Mandarin Chinese interpreters, we understand that we are somewhat rare beings. After all, we work with a language which, despite being a UN language, is not one you’d encounter regularly. We wouldn’t expect colleagues working with other, more frequently used languages to know about the peculiarities of Mandarin.

This applies not least to terminology tools. Many of the tools available to interpreters do now support Chinese-script entries. And indeed: Interpreting from English into Chinese, terminology software works as well for Chinese as it would for any other language – next to good old Excel, I myself have used InterPlex, Interpret Bank and flashterm. It’s rather when working from Chinese into English that things get tricky – and that’s not necessarily a software issue.

Until recently, many interpreters were convinced and rather adamant that simultaneous interpreting with Chinese is downright impossible, and I am sometimes tempted to agree. Compared to English – left alone German – Chinese is incredibly dense. Many words consist of only one syllable, and only very few of more than two. Owing to the way Chinese works grammatically, the very same idea can be expressed a lot more concisely in Chinese than English. To make matters worse, we replace modern Mandarin expressions with written, Classical Chinese equivalents in formal Mandarin. As a rule of thumb, the more formal the Chinese used, the more succinct it will be as well – rather different from English or German. This also is the case with proper names and terminology, which will usually have abbreviated forms that are a lot shorter syllable-wise than their English equivalents.

Adding to that, Chinese natives are incredibly fond of their language and make ample use of its full range of options: Using rare and at times byzantine expressions and words is appreciated and applauded as a sign of good education; it is never perceived as pedantic or conceited. This includes idioms (chengyu 成語), which usually refer to a story from the Chinese Classics in a highly condensed fashion: They generally contain a mere four syllables and usually function as adjectives, in contrast to English or German. In English, we will need at least a full sentence to explain what is being said, even if the same or a similar idiom exists. Chinese also frequently uses xiehouyu 歇後語: proverbs consisting of two parts, the first presenting a scenario, the second outlining the rationale of the story. Usually, the second part will be left out because Chinese natives will be able to deduce it from the first – similar to speak of the devil (and he will appear) in English. Needless to say, there hardly ever is an English equivalent, and seeing that we are operating in entirely different cultural contexts, ironing out cultural differences when explaining xiehouyu will take additional time.

It will be no surprise to hear that Chinese discursive and grammatical peculiarities make it a difficult language to interpret from: relative clauses tend to be lengthy – and are always placed in front of the noun they describe; Chinese doesn’t mark tenses as such but rather uses particles to outline how different events and actions relate to each other, in contrast to linear notions of time and tenses in European languages, so we are often left guessing; he and she are homonyms in Mandarin; to name but three examples.

Considering all of this, we see that more often than not, simultaneous interpreting from Chinese is a race against the clock and an exercise in humility – and there isn’t much time to look up words in the first place.

In Modern Mandarin, there are only around 1,200 possible syllables, with each syllable being a morpheme, i.e. a component bearing meaning; in English, we have a far greater range of possible syllables, and they only make sense in context, as not every syllable carries meaning: /mea/ and /ning/ do not mean anything per se, but meaning does. For Chinese, this implies that homophones are a common occurrence. And while we aim for perfect clarity and lucidity in English, Chinese rather daoistically indulges in ambivalence. Clever plays on words, being illusive and vague and giving listeners space to interpret what you might actually mean: not bad style, but an art to be honed. Apart from having to spend more capacity and time on identifying terms and words used in the original, this adds another layer of difficulty with regards to looking up terminology in the booth: The fastest way to type Chinese characters is by using pinyin romanisation, but owing to the huge number of homonyms, any syllable in romanized transliteration will give you a huge range of options. This means that we would have to spend at least another second or so simply to select the correct character from a drop-down list – and we will not enjoy the pleasure of word prediction that works for other languages.

In practice, this means that besides very intensive preparation before the event, we rely on what might be the oldest terminology tool in the world: our booth buddy. They are particularly important because in Chinese, we obviously don’t have any cognates – something than might get us off the hook working with European languages. We also heavily rely on our colleagues sitting next to us for figures: Chinese has ten thousand (萬) and one hundred million (億) as units in their own right, so rather than talking about one million and one billion, the Chinese will talk about a hundred times ten thousand and ten times one hundred million, respectively. This means we will have to be calculating while interpreting: a feat hard to accomplish if you are out there on your own.

While I started out thinking that not being able to use terminology software to the same extent I would use it for German-English would be quite a nuisance, I have found that this is rather an instance of the old man living at the border whose horse runs away1, as you’d say in Chinese. Interpreting is teamwork after all, and working with Chinese, we are acutely aware that we rely on our booth buddy as much as they rely on us and that we can only provide the excellent service we do with somebody else in the booth. With that in mind, professional Chinese interpreters always make for great partners in crime in the booth.

About the author:

Felix Brender 王哲謙 is a freelance conference interpreter for English, Chinese and German based in Düsseldorf/Germany. He also teaches DE>EN at the University of Heidelberg, and ZH>EN interpreting as a guest lecturer in Leeds, UK, and Taipei, Taiwan.

1 (which, as the story goes, then returns, bringing a fine stallion with it, which is then ridden by his son, who falls of the horse and breaks his leg, which is why he is not drafted and sent to war, ultimately saving his life; meaning that any setback may indeed be a blessing in disguise, similar but not entirely identical to every cloud has a silver lining. One of the most frequently used Chinese sayings, eight syllables of which the latter four are generally left out: 塞翁失馬,焉知非福, which literally translates as ‘When the old man from the frontier lost his horse, how could one have known that it would not be fortuitous?’. I rest my case.)

Mein Gehirn beim Simultandolmetschen| My brain interpreting simultaneously

Gehirn einer Simultandolmetscherin
conference interpreter’s brain

+++ for English, see below +++

Für gewöhnlich fragen wir uns ja eher, was gerade um Himmels Willen im Kopf des Redners vorgeht, den wir gerade dolmetschen. Unsere Kollegin Eliza Kalderon jedoch stellt in ihrer Doktorarbeit die umgekehrte Frage: Was geht eigentlich in den Dolmetscherköpfen beim Simultandolmetschen vor? Zu diesem Zweck steckt sie Probanden in die fMRT-Röhre (funktionale Magnetresonanztomographie), um zu sehen, was im Gehirn beim Dolmetschen, Zuhören und Shadowing passiert. Und so habe auch ich mich im November 2014 aufgemacht an die Uniklinik des Saarlandes in Homburg. Nachdem uns dort Herr Dr. Christoph Krick zunächst alles ausführlich erklärt und ein paar Zaubertricks mit dem Magnetfeld beigebracht hat (schwebende Metallplatten und dergleichen), ging es in die Röhre.


Dort war es ganz bequem, der Kopf musste still liegen, aber die Beine hatten zu meiner großen Erleichterung viel Platz. Dann habe ich zwei Videos im Wechsel gedolmetscht, geshadowt und gehört, ein spanisches ins Deutsche und ein deutsches ins Spanische. Neben dem Hämmern der Maschine, das natürlich ein bisschen störte, bestand für mich die größte Herausforderung eigentlich darin, beim Dolmetschen die Hände stillzuhalten. Mir wurde zum ersten Mal richtig klar, wie wichtig das Gestikulieren beim Formulieren des Zieltextes ist. Nach gut anderthalb Stunden (mit Unterbrechungen) war ich dann einigermaßen k.o., bekam aber zur Belohnung nicht nur sofort Schokolade, sondern auch direkten Blick auf mein Schädelinneres am Computer von Herrn Dr. Krick.

Natürlich lassen sich bei einer solchen Untersuchung viele interessante Dinge beobachten. Beispielhaft möchte ich zum Thema Sprachrichtungen Herrn Dr. Krick gerne wörtlich zitieren, da er mir das Gehirngeschehen einfach zu schön erläutert hat: „Da Sie muttersprachlich deutsch aufgewachsen sind, ergeben sich – trotz Ihrer hohen Sprachkompetenz – leichte Unterschiede dieser ähnlichen Aufgaben bezüglich sensorischer und motorischer Leistungen im Gehirn. Allerdings möchte ich nicht ausschließen, dass der Unterschied durchaus auch an der jeweiligen rhetorischen Kompetenz von Herrn Gauck und Herrn Rajoy gelegen haben mag … Wenn Sie den Herrn Gauck ins Spanische übersetzt hatten, fiel es Ihnen vergleichsweise leichter, die Sprache zu verstehen, wohingegen Ihr Kleinhirn im Hinterhaupt vergleichsweise mehr leisten musste, um die Feinmotorik der spanischen Sprechweise umzusetzen.“

simultaneous interpreting German Spanish
Simultandolmetschen Deutsch (A) – Spanisch (B)

„Wenn Sie aber den Herrn Rajoy ins Deutsche übersetzt hatten, verbrauchte Ihr Kleinhirn vergleichsweise weniger Energie, um Ihre Aussprache zu steuern. Allerdings musste Ihre sekundäre Hörrinde im Schläfenlappen mehr sensorische Leistung aufbringen, um den Ausführungen zu folgen. Dies sind allerdings nur ganz subtile Unterschiede, die in der geringen Magnitude den Hinweis ergeben, dass Sie nahezu gleich gut in beide Richtungen dolmetschen können.“

simultaneous interpreting Spanish German
Simultandolmetschen Spanisch (B) – Deutsch (A)

Dies ist nur einer von vielen interessanten Aspekten. So war beispielsweise auch mein Hippocampus relativ groß – ähnlich wie bei Labyrinth-Ratten oder den berühmten Londoner Taxifahrern … Welche wissenschaftlichen Erkenntnisse sich aus der Gesamtauswertung der Studienreihe ergeben, dürfen wir dann hoffentlich demnächst von Eliza Kalderon selbst erfahren!

PS: Und wer auch mal sein Gehirn näher kennenlernen möchte: Eliza sucht noch weitere professionelle Konferenzdolmetscher/innen mit Berufserfahrung, A-Sprache Deutsch, B-Sprache Spanisch (kein Doppel-A!), ca. 30-55 Jahre alt und möglichst rechtshändig. Einfach melden unter

PPS: Auch ein interessanter Artikel zum Thema:


Normally, we rather wonder what on earth is going on in the mind of the speaker we are interpreting. Our colleague Eliza Kalderon, however, puts it the other way around. In her phD, she looks into what exactly happens in the brains of simultaneous interpreters. To find out, she puts human test subjects into an fMRI machine (functional Magnetic Resonance Imaging) and monitors their brains while interpreting, listening and shadowing. I was one of those volunteers and made my way to the Saarland University Hospital in Homburg/Germany in November 2014. First of all, Dr. Christoph Krick gave us a detailed technical introduction including a demo of how to do magic with the helfp of the magnetic field (flying metal plates and the like). And then off I went into the tube.


To my delight, it was quite comfortable. My head wasn’t supposed to move, ok, but luckily my legs had plenty of room. Then Eliza made me listen to, interpret and shadow two videos: one from German into Spanish and one from Spanish into German. The machine hammering away around my head was a bit of a nuisance, obviously, but apart from that the biggest challenge for me was keeping my hands still while interpreting. I hadn’t realised until now how important it is to gesture when speaking. After a good one and a half hour’s work (with little breaks), I was rather knocked out, but I was rewarded promptly: Not only was I given chocolate right after the exercise, I was even allowed a glance at my brain on Dr. Krick’s computer.

There are of course a great many interesting phenomena to observe in such a study. To describe one of them, I would like to quote literally Dr. Krick’s nice explanation: „As you have grown up speaking German as a mother tongue, despite your high level of linguistic competence, we can see slight differences between the two similar tasks in terms of sensoric and motoric performance in your brain. However, it cannot be ruled out that these differences might be also be attributable to the respective rhetorical skills of Mr. Gauck and Mr. Rajoy. When translating Mr. Gauck into Spanish, understanding involved comparably less effort while the cerebellum in the back of your head had to work comparably harder in order to articulate the Spanish language.“

simultaneous interpreting German Spanish
Simultaneous interpreting German (A) – Spanish (B)

„When, on the other hand, translating Mr. Rajoy into German, your cerebellum needed comparably less energy to control your pronunciation. Your secondary auditory cortex, located in the temporal lobe, had to make a greater sensoric effort in order to understand what was being said. Those differences are, however, very subtle, their low magnitude actually leads to the assumption that you practically work equally well in both directions.“

simultaneous interpreting Spanish German
Simultaneous interpreting Spanish (B) – German (A)

This is only one of many interesting aspects. Another one worth mentioning might be the fact that my hippocampus was slightly on the big side – just like in maze rats or London cab drivers … I am really looking forward to getting the whole picture and reading about the scientific findings Eliza draws once she has finished her study!

PS: If you, too, would like to get a glimpse inside your head: Eliza is still looking for volunteers! If you are a professional, experienced conference Interpreter with German A and Spanish B as working languages (no double A!), about 30-55 years old and preferably right-handed, feel free to get in touch:

PPS: Some further reading:

PPPS: A portuguese version of this article can be found on Marina Borges‘ blog: