When is precision excessive?

When I was first living in Berlin and new to Germany I was surprised to regularly overhear telephone conversations on the subway that went something like this: “Hey Klaus, I’m on my way but I will be a bit late; you can expect me in 6 to 7 minutes.” Sometimes the delay cited would be 13 to 14 minutes, or 17 to 18 minutes, but it would invariably involve a narrow time range and at least one prime number. An American, of course, would never attempt to quantify the delay with such precision, as this would seem foolish (“will you really be there in exactly 6 to 7 minutes??”) and/or completely anal, as it violates the unwritten law that you must announce you are going to be late with numbers divisible by five. From the German perspective, however, providing a fastidiously accurate Zeitangabe (i.e. “time specification”) is completely reasonable, as everyone knows the average station-to-station travel time in Berlin is 2 minutes, so you merely have to multiply the number of subway stops by 2 and add the additional walking time required, considering that at a brisk pace you can cover 500 meters in 3.5 minutes. This simple calculation method yields an arrival time with a margin of error of +/-30 seconds. Of course, I’m exaggerating (slightly), but punctuality is a big deal in Germany, and accurately stating when you will arrive is simply considered polite.

The German penchant for accuracy in weights and measures is also particularly evident when figures are cited in magazine and newspaper articles. From a German perspective, extremely accurate figures add color to the story and also underscore the thoroughness of the reporting. A recent article in Der Spiegel about the Mexican drug war, for example, begins by recounting a drug bust in which an undercover police officer was carrying “155 grams of heroin”. As this figure could be interpreted by American and British readers as gratuitously precise, in the English version of the article, it was simply changed by the translators at Der Spiegel to “150 grams of heroin.”

This modification of the German original is an excellent example of effective adaptation, as it places consideration for the reader above a slavish adherence to the content of the source text, which may not be effective in a different cultural space. The fact that the numbers themselves need to be changed when translating into English speaks to the level of adaptation that is occasionally required.

On a side note, the modification made to the article’s title is also illustrative: Originally titled “Hombre, du bist am Arsch” (“Hombre, you’re fucked”), the English version reads “El Paso Vice.” This demonstrates an additional principle of good translation: if the direct equivalent in English is weird or otherwise inappropriate, you should simply change it.

Codified error and Trennbankensysteme

One of the most surprising phenomena in the world of translation that I encounter with great frequency is the tendency for clearly incorrect translations of certain terms to become accepted and widely used as standard terminology. This phenomenon is certainly attributable in part to native speakers of German translating into English. Yet indolence and timidity of many translators must also play a large role. To provide an example: The accepted translation for “Trennbankensystem” at LEO and in other specialized dictionaries is “separate banking system.” This translation really raises my hackles, as it is completely incorrect. The term “Trennbankensystem” refers to the separation that previously existed in the US between between commercial and investment banks, as per the requirements of Glass-Steagall. The German term thus refers expressly to circumstances in the United States, which should make the identification of a good translation an easy task. The accepted terms in English for this separation are “dual banking system” and “two-tiered banking system”, but not “separate banking system,” which almost seems to refer to some sort of “secondary” or “shadow” banking sector. The term is thus not only incorrect, but highly misleading. Nevertheless, it is found in all online dictionaries that contain a entry for the term, and has been used faithfully by translators in innumerable documents. It really is a sad state of affairs.

This is by no mean an isolated example. Consistent mistranslation – or “codified error,” as I like to call it – is extremely common: In the area of finance, for example, I often see the English term “participations” used as a translation for “Beteiligungen”, which is just about as correct as Arnold Schwarzenegger talking about the “advices” he gives to Franco Columbu. The correct way phrase this, of course, it to speak of an “ownership stake” or “holdings” in a company. “Participation” for describing investment interest can work, but should almost always be kept singular.

Dynamic adaptation

“Dynamic adaptation” is one way of referring to the various modifications and adjustments that are required when translating from one language to another in order to assure a clear and readable target text. It means taking a proactive approach to the translation process by actively searching for modes of expression in the target language that are effective for conveying the content of the original document. Often, to create a readable English translation massive modification is required, as equivalents for the German terms and phrases simply don’t exist. In many instances, the lack of equivalents is attributable to institutional differences, as in the following example:

“[Mit einem Minijob vermeiden sie] die hohe implizite Besteuerung durch das Ehegattensplitting und die beitragsfreie Mitversicherung in der gesetzlichen Krankenversicherung.”

This sentence describes labor market and tax regulations in Germany that don’t have direct correlates in the US or UK. The difficulties in translating this sentence are actually two-fold: While there are in fact weakly established translations for a few of the terms in the sentence, these “translations” are to some extent completely useless, for the reader, lacking knowledge of the circumstances that prevail in Germany, will simply not understand the “English” terms. The only solution, therefore, is to go about the translation in a way that explains the situation to the reader. This problem is posed in particular by “Ehegattensplitting”. While a direct translation might be “spouse splitting,” we should perhaps avoid implying that the Germans are running a government-subsidized “wife swapping” program. My translation thus includes a gloss: “… the high implicit taxes resulting from the ‘splitting rule’ (which allows spouses to pool and equally divide their income to determine tax liability) ….”

Yet the more crucial problem that emerges in this sentence – and the actual crux of my point – relates to the dynamic transformation that is needed to convert the phrasing “die beitragsfreie Mitversicherung in der gesetzlichen Krankenversicherung” into acceptable English. The outcome of a word-for-word translation is completely unintelligible: “the contribution-free co-insurance in the statutory health insurance.” The term “co-insurance” – apparently an accepted translation for “Mitversicherung” – is laughably bad, as it is totally misleading. Massive conversion is needed to arrive at a coherent English phrase. Here is one possible way to word the idea being presented: “tax-free extension of government insurance coverage held by a working partner to his or her spouse.”

If the goal is to produce an understandable and correct English text, this sort of adaptation is actually required in nearly every sentence that ones translates, demonstrating that translation is much less a form of transcription than actually a reinventing of the source text’s ideas to accommodate the possibilities and restrictions of a radically different system of words and sempahore. The individual words hardly correspond at all – because words are tangential to ideas, which tend to reside in phrases, and not just individual particles.