TM AND MT IN THE REAL WORLD: SHOULD TRANSLATORS BE ANGRY?

Anger, or the absence thereof, is in the news a lot these days.  Supporters of both major political parties are angry (for different reasons, of course), and the general public is angry.  Oddly enough, much of this anger is directed at the man in the White House, who does not seem to be angry at all.  Speaking of President Obama in a recent New York Times article, Ishmael Reed said that it is “risky for a black man to express anger…he’d be dismissed as an angry black militant with a deep hatred of white people.”  Nonetheless one can only wonder what would have happened had he expressed even a teeny-tiny bit of wrath or outrage.

Shortly after I posted part 3 of Bernie Bierman’s article, I received emails from a couple of followers.  Both said that, while they agreed with Bernie, they felt that he was “too angry”.  While this seems to be a matter of degree (how angry is too angry?), it certainly raises some important questions.  For example, how do we as translators feel about ourselves and the work we do?  Are we happy about “selling words”, new words, repeated words, recycled words?  Should we not be angry that those who “purchase” these words are trying to tell us how much they will pay for them?  And, should we not be angry that many of our colleagues have been seduced by vendors of technology into acting counter to their own best interests? 

It seems to me that we should be very angry about what the rush to embrace translation memory tools and now machine translation has done to the translation profession.  We should be angry that this was done “with the blessing” of ATA and other translators’ associations.  We should also be fearful that the “real” translators (mentioned in a comment to part 2 of Bernie Bierman’s article) are becoming extinct. 

Not all that long ago, translation was regarded as a writing profession and translators were “ghost” writers of sorts.  They “transformed” a message in the source language into a message that could be readily understood by target-language readers.  They were communicators.  To do this well requires knowledge, skill, training and, yes, talent.  It goes beyond just matching words.  Translators needed a broad background of knowledge, an insatiable curiosity, and the general ability to learn new things.  A “real” translator crafted each document with care, constantly searched for le mot juste, and was unhappy if a heavy-handed editor made stylistic changes.  The author of the translation felt a sense of ownership. 

All of that changed drastically with the advent of CAT tools.  The new breed of “translators” (also known as CAT operators) no longer works with documents.  Instead they are working with segments that come from many different sources.  They are now doing what one of my colleagues calls “fill-in-the-blank” translation and are only being paid their full rates (if they are lucky and the client doesn’t set the rate) for the words in the blanks they fill in.  However, for CAT operators, this is not a real issue because the fact that they only have to fill in the blanks allows them to accept translation projects or assignments which, in the past would have been beyond their abilities. Because they are “recycling” so many words, they can sell more words for less.    

While this may not be a financial issue for CAT operators, it certainly gives rise to questions about translation quality. In Denver I overheard a couple of agency owners grousing about problems with translation memories because “translators” do not update them.  What in the world do they expect?  If CAT operators are being paid next to nothing for repetitions and so-called 100% matches, and only slightly more for so-called “fuzzy matches”, there is no incentive to check for accuracy and they move on.  In spite of the fact that some members of our community would like to equate the word “professional” with behavior, the truth of the matter is that the “professional”, unlike the amateur or dilettante expects (or should expect) to be paid for his/her expertise.       

Only a few short years ago, translators were making their own decisions about whether or not to use a CAT tool.  They made this choice based on the type of work they do.  All of that has changed because many agencies require that a job be done with Trados or some other CAT tool.  It goes without saying that the overall emphasis on technology has had a deleterious effect on the translation profession, and this goes beyond declining rates.  A steady diet of “fill in the blank” translation can affect the translator mentally as well as financially.   As Gabriel Fairman points out in the June 2010 issue of MultiLingual:

The most tacit [sic] and perhaps most important issue takes place at an emotional level when the translator’s feeling of authorship over the document gets impaired.  As a translator’s work is reduced to translating segments rather than documents, power and responsibility get dimmed as well.  The translator who inputs segments ascribes [sic] more clearly to a machine metaphor, while a translator who crafts a text corresponds more to an artistic metaphor.  The more clearly a translator, or any professional for that matter, has sight of the overall purpose and art of the work, of his or her ownership of everything that lies within a certain realm, the greater the chances of self-improvement and self-satisfaction.

In other words, if you don’t use it, you lose it.  As linguists are pushed (and they are indeed pushed) to embrace technology as a means of making a living, they focus more and more on the selling of words, new words, repeated words, and “fuzzily-matched” words .  The idea of the translator as someone who communicates a message is being replaced by a technician who “processes” words.  Many of these translators cum CAT operators now call themselves language engineers. 

Ah, but all those CAT operators should beware because things change quickly and CAT tools are quickly becoming retro.  The prediction is that in the not too far distant future, they will be replaced by SaaS (software as a service), translation “in the cloud” and, yes, machine translation.  Well, if we thought that CAT tools had a deleterious effect on human translation, the effect of machine translation will be infinitely worse because the decline in compensation will be even more pronounced as translators become post-editors of machine translation.  Language Service Providers (also known as translation agencies) that are pushing post-edited machine translation (PEMT) as a viable alternative to human translation are telling their prospective clients that they can save more than 50% of the cost of human translation.  Post-editors will be trained to work in two modes:  light (cleaning up grammar and spelling in the target language) or heavy (adding some elements of style where some knowledge of the source language will be required).  Translators will be asked to work on jobs with machine translation, multiple translation memories, and even “new” words.  With so many translators’ associations discouraging or forbidding any discussion of rates and economic matters, translators have nowhere to turn except perhaps to the internet.

Since CAT has become the signifier for computer-assisted translation, perhaps HAT (human-assisted translation) would be a good descriptor for pre- and post-editing.  With all due apologies to Dr. Seuss, I wonder what the CAT in the HAT would think about that.   And, even more importantly, what do human translators think about these trends?  After all, “real” translators know that human language is idiomatic, idiosyncratic, and constantly evolving.  They know that it is wishful thinking on the part of MT enthusiasts who say that “we are not there yet, but we are getting closer”.  Yet, many are taken in by this talk.  After all, language is the thing that separates us from the beasts of the field.  While those beasts may communicate on a primitive level, there is no evidence that they attempt to persuade, dissuade or otherwise manipulate their fellow creatures in any way.  That is a particularly human characteristic and there are lots of human beings who are making lots of money “persuading”. 

We should all be very angry about the present state of affairs.  We should be angry that some of our colleagues think so little of their skills that they are willing to work for peanuts (cf. www.NoPeanuts.com).  We should be angry when clients attempt to set the rates THEY will pay.  We should be questioning the practices involving translation memories and who owns them, and a whole host of other things.  Ideally, this discussion should be opened up and led by our translators’ organizations but, if what I observed at the ATA gathering in Denver is any indication, it is not going to happen.  So, it is up to the “real” translators, the ones who are in danger of becoming extinct, to make their voices heard, to use their language skills to persuade others to take a long, hard look at what is happening in our profession and then make their voices heard.