De labore interpretandi

Quality is when customers come back, and the products don’t.
Siemens motto (quoted 
by Philip Kotler)

Found in TranslationFound in Translation, Jost Zetzsche’s and Nataly Kelly’s book on translation, is an enjoyable read and a good Christmas gift for customers who speak English and are curious about translation.

It suffers, though, from two original flaws, that would not be serious if the book were published by a lesser publisher with limited resources, and had it not been authored by two of the most acclaimed authorities in the industry.

The first flaw is that it is clearly designed for a specific local audience.

The second flaw is that it is built on curious, fascinating and engaging anecdotes offering a very traditional and very static outlook of the job, with a strong inclination towards interpretation and language issues.

Digressions and personal notes

Two lines of business are touched in the book that I myself tried in the beginning: romance and erotic novels, and porn.

I’ve never been burned by the sacred fire of languages and I approached translation in a roundabout way.

My first experience was accidental. I was in the army and my battalion was asked to identify a possible interpreter as a liaison with the allied forces in an imminent NATO exercise.

At the commander’s request, a soldier came forward. He claimed to be studying as an interpreter and was, therefore, sent on a mission.

I took part in the exercise as the commander of my unit and had many contacts with fellow officers of the allied forces.

Back to base, I discovered that the Allied command had expressed a strong dissatisfaction with the performance of the “official” interpreter and complained about the lack of a liaison officer, but was satisfied with the “remedy”.

Without my knowledge, and that of my command, I was that “remedy”.

When I left the army, after my office, I went back to my engineering studies, and took another accidental opportunity to earn some money with translation.

I started translating for a few agencies, but spent more time working than studying. After a short time, I made my mind up to start a career in translation, and enrolled at the school for translators and interpreters.

One of my teachers told us that translating is the second oldest job in the world, and the two share the same reasons: one starts translating for love, continues for a few close friends, and ends up doing it just for money.

That is what happened to me as well.

Maybe this is why I easily accepted assignments that others disdained and had no illusions, let alone dreams, about earning a living by translating works of great authors.

Acetabulum cavityMy first paid job as a pro was an article for a medical journal on “hip joint replacements”.

I was so young, ambitious, arrogant and reckless to accept an assignment that was definitely beyond my capabilities.

I spent a whole day at the library to get on top of special terminology. My false sense of security made me make a mistake, which now I think should deserve more indulgence, but then unleashed the customer’s wrath: I translated “fossa dell’acetabolo” (acetabulum cavity) with “hip bone socket”.

The customer pointed out to me that a world-famous orthopedic surgeon would never have used a colloquial term, like say “shrink” or “bonesetter.”

The lesson was worth more than the entire translation course and I have never forgotten it, especially for the many implicit connections.

A few weeks later, I answered a classified ad for translators from the most important Italian publisher of romance novels.

I passed the test and, surprisingly, the editor in chief told me that, in her experience, men made better translators in that field, being capable of the necessary “detachment”. But the offer was poor, mostly made up of low-skilled women, although it was a well-paid job.

It was, indeed, a well-paid job, just like the translation of a screenplay for an adult movie. As for romance novels, the demand for translators in the adult movie industry was high while the offer was scarce and poor. And as for editorial translation, screenplays were poorly paid in respect of an over-abundant offer of translators.

From 2003 to 2011, I had the chance to teach terminology and localization at the faculty of translation and interpreting of a private-owned university in Rome.

I used to tell my students about the above experiences, and I could always see surprise, or even disapproval, in their eyes.

Because of the many social taboos, it will come as no surprise that there is hardly one translator boasting engagement in the industry of adult entertainment.

There was no disapproval for my involvement in the translation of romantic novels; on the contrary, it was viewed with some hypocritical indulgence, maybe because of the prejudices that feed the false idea of social prestige resulting from the translation of literature and that share the same origin with the taboos surrounding the industry of adult entertainment.

In 2010 Google estimated the total number of books in the world at 130 million. Therefore, being published is a sign of being worth something, at least of being read by someone. That misleading prestige above comes from here.

A series of specious illustrations of a distorted reality of the translation industry combine with those taboos, especially about the specificity, the importance, and the difficulties inherent in a translator’s job. There is a reason why translators refuse to use the word “job” and prefer by far the more ambitious and misleading “profession”.

It is basically the same attitude that leads someone to stupidly and bluntly insult those who show an opinion other than the majority’s, even because it is easy to attain a benevolent condescension of such a majority by pouring one’s own venom on common bogeys.

Like Hotel California, checking in is not a problem, but checking out can be an issue: we are all prisoners here, of our own device, you can check-out any time you like, but you can never leave! You can get caught in there, as in a terrifying nightmare, which at first seemed a sweet dream.

What Don Felder, Don Henley and Glenn Frey could see as the “dark vulnerability of the American Dream,” a metaphor of the slavery to drugs, translators can find it in the traps scattered from peers and teachers. First of all the quality myth.

The Quality Myth

Translation scholars and translators have made quality controversial and irrelevant.

It is no justification that the topic has been the subject of analysis and disputes for centuries. The same Jerome wrote his letter to Pammachius in defense of his work. He did so by quoting those who had to face the same problem before he did, from Ennius to Horace, from Terence to Quintilian, and especially Cicero, by paraphrasing the title of one of his whose most famous works.

As Diego Bartolome noted recently, “giving gold to the client requesting iron is a waste”: the gold will still be paid and treated as iron. Your efforts will be wasted and you’ll be disappointed. And, in any case, you are the sole responsible.

In this regard, I always presented the same example to the students who attended my localization class, asking them to translate a single string, according to the most typical increasingly spreading practices in software localization:

This key is listed in the Stolen Codes Database for the program.

Given the usually short lead times and the increasingly low fees, to be profitable, the translation has to be done in a maximum of 2’15”.

All versions were never better that the translation produced by GT:

Questa chiave è presente nel database dei codici rubati per il programma.

The versions of several highly experienced professionals did not differ much from the “baseline”.Piero Manzoni: Merda d’artista

Interestingly, to support their poor performance, the “samples” cited the most classic and imaginative excuses, from “contextualization” to “respect for the text.”

And even when provided with the string label (64423_arm_R_ARM_ERROR_ACCOUNT_LOCKED_OUT), the result generally did not improve.

A version that would meet the criteria of software localization, starting with communication effectiveness, could be:

Questa chiave risulta rubata.

I then used a parallel with one of the most famous works of modern art, Piero Manzoni’s “Artist’s shit”, and his ironic and subtle reference to the artist’s work.

I pointed out that, although theoretically perfect, the “ideal” version would be rejected by virtually any customer — typically LSPs — as “too short”.

Innovation and the Race to the Bottom

According to some observers and industry players, anecdotal evidence indicates that the translation industry is engaged in a race to the bottom.

If that is the case, there is something wrong in the statements of those who keep arguing that 2012 has given the language services industry plenty of reasons to celebrate – strong growth rates, a positive job outlook, spikes in volume, and good sales figures.

There is a small problem, though. The main translation market is overseas, where courses in translation and interpreting are definitely scarce (at least in comparison with Europe) despite the need for language specialists in government and commerce, and where the brand-new graduate degree program at the University of Maryland is considered as a piece of important news.

In fact, educational programs are critical, not only to develop the labor pool, but because they shape the future of the industry.

Not surprisingly, then, many are scared of machine translation, as it could produce a tsunami.

Crying wolf, tearing one’s own clothes and hair, however, is useless. It just shows ignorance and feeds the worst moods of those who think that the industry is basically populated by hopeless unqualified workers.

In reality, the translation industry remains a highly fragmented industry with relatively inefficient production and business models. Machine translation is a viable solution of real value in limited domains and not for everyone.

The translation industry needs innovation, especially when it comes to business models.

In a recent report, Common Sense Advisory seems to re-invent the wheel when saying that “most LSP production models are freelancer-centric.”

And yet, this is the core of the problem. According to CSA, though, it would be possible for the industry to exist without freelancers “only if we assume that all translation work could be performed by full-time employees instead of freelancers. Such models are not flexible, scalable, or cost-effective enough to respond to market demands. In other words, they are not practical”.

In reality, it is not practical to share the same freelancing resources all over the world, despite the notorious and annoying scarcity of good translators (or maybe because of it).

The practicality of the traditional freelancer-centric business model lies in the possibility to dump one’s operational inefficiencies on freelancers.

On the other hand, CSA provides an indirect counterargument when saying that “a large number of agencies in the industry act as single-language vendors (SLVs) or regional language vendors (RLVs). These companies largely derive their revenue from multi-language vendors (MLVs).”

This could be solved with consolidation, as the recent story of the industry tells us. In 2012 we still have over 25,000 LSPs of varying quality and professionalism.

It is also quite arguable that “buyers understand the importance of freelancers:” information asymmetry is still a major harm to the industry’s credibility.

Finally, where freelancers receive great importance (manufacturing, financial services, and life sciences), industry players operate mostly on the models above dismissed as “impractical” and the use of freelancers is exceptional and direct.

Recently, Renato Beninatto held a webinar for his company Moravia on “microtranslations as new economic reality”, arguing that gone are the days of large, standalone projects taking months to plan and complete. Instead, it’s really more about a continuous flow of small chunks or drips of content that need to be picked up, processed and published within a short period of time.

In reality, the shift “from drops to drips” is mostly due to the inability of MLVs and LSPs to devise new, more efficient, competitive and recognizable business models, making the translation industry a real industry.

A big company is usually expected to release a large number of important innovations, although maybe not “disruptive”. Scrolling through the list of the 100 language service providers, however, no one in the top ten, or the top twenty, or even in the top thirty, stood out for significant innovations.

Maybe that’s why the translation companies listed on stock exchanges still do not receive any special attention and their shares generally come with a suggestion to sell.

Most of the disruptive innovations come into the market because the majority of clients don’t need most of the features initially added into the product or service.

A vast number of useful innovations come not from some scientist and engineer tinkering in a lab, but from users solving their own problems.

Digitalization has moved toward costless sharing. The translation industry is desperately reacting to digitalization by replicating the offerings available in the physical world.

Can you envisage a model for your business based on sharing that provides adequate returns for the cost of production?

Clinging Like Ivy

The costs of getting a product to consumers limits competition. Digitalization reduces these costs. The massively digitized translation industry should therefore apply strategy 101: ask yourself what you provide that is unique and charge for it.

Ada Lovelace’s case is emblematic.

Her fame was mostly due to an article that she translated and that was written by the Italian mathematician Luigi Menabrea. The topic was Charles Babbage’s analytical engine. Other scientists did not grasp the concept. Ada Lovelace spent 9 months on it, appending a set of notes, which were longer than the article.

By the way, Luigi Menabrea’s article was written in French.

For the same reason, every LPS, including large MLVs, hold on to good translators because they are very difficult to find.

And it’s exactly this difficulty that leads LSPs to cooperate with less-than-good translators and pay them less and less.

Here comes Gresham’s Law.

Who is responsible, then, for the commodification of translation? Certainly not technology.

In a recent paper, Stanford University biochemist Gerald Crabtree suggests that evolution and the eventual simplicity of modern life are making us dumber. According to Dr. Crabtree, with the development of agriculture came urbanization, which may have weakened the power of selection to weed out mutations leading to intellectual disabilities. In practice, we could say that our intelligence is much weaker than that of our ancestors of a few thousand years ago.

Dr. Crabtree speculative and controversial thesis seems in contrast with the diagram used by Nobel laureate Robert William Fogel in The Escape from Hunger and Premature death, 1700-2100 : Europe, America and the Third World. The diagram in questions shows the relationship between the growth of world population and some major events in the history of technology.

In some cases, however, technological progress seems to have definitely played against us. On the one hand, technology has made our existence more comfortable and durable; on the other, it made us intellectually lazy.

For example, the first home computers, like Commodore 64, were hard to govern, programming skills were a must, any achievement was a feat, and the enfant terrible (like the one depicted in John Badham’s War Games by a fresh-faced Matthew Broderick) was very common.

Today, technological evolution has mostly introduced a general waste of resources, that we see daily. The C64 required only 64 Kbytes of precious memory. The tons of RAM installed today on most PCs are usually spent to do the same things that the C64 could do, less with more. Not surprisingly, despite the endemic prevalence of computers, PDAs, and any other kind of electronic device, spotting a kid coding a rudimentary program or game is as hard as spotting a comet. Today, the “digital divide” is the gap between what users can do and what the machine can do. And this gap is widening every day.

In What Money Can’t Buy: the Moral Limits of Markets, Michael Sandel cited Michael Lewis‘s Moneyball: The Art of Winning an Unfair Game as illustrative of an “important intellectual revolution.”

In Lewis’s book, an underfinanced underdog team uses its wits and the tools of modern econometrics to compete with rich, powerhouse teams. The lesson is that exploiting market inefficiencies can pay off.

It’s hard to stand up and cheer for the triumph of quantitative methods and more efficient pricing mechanisms. The Oakland Athletics depicted in Lewis’s book, though, made the playoffs and haven’t had a winning season since. This is not because moneyball failed but because it spread. The Boston Red Sox won two championships later on.

In practice, moneyball turned out to be the wrong strategy for underdogs, at least in the long run.

The actual moneyball lesson is that when talent is priced efficiently, the teams with the most money to spend on player salaries can be expected to do better.

Moneyball made baseball more efficient, not better. Ten years later, the message in Trouble With the Curve is that statistics alone don’t make a team having the best players.

Technology is useful, but to exploit it efficiently in the long run, deep know-how and planning are needed.

Innovation doesn’t mean buying the latest piece of software or forcing your vendors to use it.

Yet, that’s what happens regularly, even with the most complex localization projects. And it is for this reason that in most localization projects, much of the file handling is still manual, although everyone in the industry knows this is a time-consuming and error-prone process.

Business Model Innovation

Before rising to current popularity, Zara’s business model has gone virtually unnoticed for over 30 years. It is designed to enable fast response to trends, with responsiveness and speed opposed to costs.

In fact, Zara produces in expensive locations and uses expensive shipping modes to react to trends faster.

However, for over 20 years, Zara’s model was rarely mentioned by the press and was misunderstood by the competition, with reactions ranging from dismissive mockery to indifference.

As in the case of Apple’s iPhone or Pfizer’s Viagra, the advantages coming from technology and product innovation are limited as both technology and product innovation can be copied.

Business model innovations, as opposed to product or technology innovation, have unique characteristics that make the benefits more sustainable.

First of all, the innovation is often present within the company’s processes that may not even be directly visible to competitors.

Secondly, copying a new business model from the competition is much harder than imitating the competitor’s product design.

Imitating someone else’s core operating (business) logic means changing and/or adapt one’s own existing one, which is definitely hard.

Translation companies produce only what customers wanted, but so far have been incapable of simplifying their supply chains. The only way to offer their products at a significantly lower price is to lower rates as the only way to cut their costs, thus fueling Gresham’s Law.

Business model innovation means changing the way one offers products or services that address existing customer needs using existing technologies. Very often this kind of innovation turns out to be more valuable and transformative than product- or technology-driven innovation.

What’s the problem, then, with business model innovation? Determining what changes to the model will be effective. How can companies systematically innovate their business models?

The value chain of translation companies is mostly the same and remains implicit for most of them, although designing a value chain is part of economics 101 and is based on three standard elements: revenue (price, market size, and ancillary sales), cost structure (direct and indirect costs, economies of scale and scope), and resource speed (the rate at which value is created from the applied resources, typically captured through lead times, throughput, inventory turns, and asset utilization).

These factors are well understood, and improving them is the main focus of management literature.

Business model innovation basically consists of creating value by adding risk through unsuspected opportunities.

Translation companies are typically very traditional companies with a very traditional approach to business. The approach in question consists of reducing risks by delaying production and financial commitments, transferring risk to other parties, or exploiting information asymmetry.


The moral? Innovation eats itself. Do you really want to stay hungry and foolish?