A Working Note on Words Slurred, Preferred and Absurd

The image “https://i2.wp.com/i113.photobucket.com/albums/n216/cbrayton/Stuff/workingterm.png” cannot be displayed, because it contains errors.
Dinking around with Glossword, a simple Web-based terminology database organized according to pretty good multilingual terminology and thesaurus-authoring practices, such as those employed by IATE.

As far as translators are concerned, terminology is primarily an ad hoc affair, more a matter of filling in the blanks in their knowledge than systematically studying a constellation of terms in a given universe of discourse. –Robert Bonnono, “Terminology for Translators—an Implementation of ISO 12620”

That, in a nutshell, is the problem that consumes my time these days.

A client has asked me to review a fairly huge collection of glossaries related to business reporting.

I have to find a an effective (and diplomatic) way of communicating to this client that the ad hoc method of constructing and compiling these glossaries means that using them in practice is likely to produce an unacceptable level of failures to communicate.

It is, to use one of my favorite New World Lusophone words, a gambiarra — a kludge.

Kludge: patched solution; a makeshift combination of hardware and software put together to solve a computing problem that is effective but not suitable for manufacture.

Making stuff up off the top of your head to meet the needs of the moment is often necessary, and a vital skill to have.

But in the long term, it means reinventing the wheel constantly.

Gambiarra is the name given informally to a procedure neeeded to configure an improvised device (gizmo), and is often used to refer to the device itself.

Bonnono (the article appeared in Meta, XLV, 4, 2000).

When translators think of terminology they imagine it as a kind of catchall roughly synonymous with a glossary or word list. Terminology is “all the words I don’t know and need to find out,” the words not found in a dictionary, or the latest jargon. While it is true that current terminological practice touches upon these meanings of the word, all of them misrepresent it in one way or another.


In the past, translators, perhaps because of the nature on their professional activity, have shown limited interest in terminology as a field of activity distinct from translation. There has even been considerable antagonism toward terminology on the part of translators, who often view it as a drain on their time and resources and an impediment to translation proper.

I find that to be true.

In the ideal world that terminologists have presented and translators dream of, terminological “issues” would be handled by terminologists and translation “issues” by translators. Queries would be submitted to terminologists for resolution and the results supplied to translators for integration in a translation product.

I believe that.

In large organizations like the United Nations, resources are available to implement such a model, but few companies and certainly no translator can afford the luxury of such a bifurcation of responsibilities. Although there is no easy way to ease the demands made on a translator’s time, the following illustrates one way of smoothly integrating terminology into the translation process.

The first step is to rediscover the principle of the division of labor.

Good paper. I have heard of this Bonnono fellow.

Robert Bononno is a full-time translator who lives in New York. His recent translations include The Singular Objects of Architecture by Jean Baudrillard and Jean Nouvel (Minnesota, 2002) and Cyberculture by Pierre Lévy (Minnesota, 2001).

Both texts likely to present intricate, nightmarish translation problems.

So that, in a nutshell, in my problem: Pointing out to the intelligent and capable but seriously disorganized bunch of professional translators who have asked for my advice that their current practices for managing terminology are described very succinctly and precisely by Bonnono as

1.2. How not to Do Terminology

Or rather

1.2 How Not To Do Terminology

In headline style, you cap the “to” in infinitives on the theory that the infinitive construction (“to headline”) is a single, er, lexical unit.

That is:

As I noted above, translators, at least those who bother to think about terminology, even in a vague way, generally associate it with collections of unknown words or terms, the critical entries that dictionary makers always seem to miss. So “terminology” tends to be thought of as parallel columns of words: source term on one side, target term on the other. Tables are wonderful for creating this sort of structure. It’s easy, it’s fast, you can do it right in your word processor, and, unless you’re still using an early version of some obscure software application, you can usually exchange files with others. If you’re ambitious, you can add a third column for client comments or corrections.

Excel hell:

Some of what passes for terminology gets stored in spreadsheets such as Microsoft Excel. Now, personally I have nothing against Excel or its makers, but it was never designed to store text, much less terminological data. It simply makes it easy to create big tables without going through the trouble of sizing and formatting them. But it was designed to crunch numbers, not words. And it’s not a database, so there is no organizational structure, no way to search or specify a field or category, no way to create entities and attributes. Although both methods (tables and spreadsheets) are common in translation today, neither can be recommended for serious terminological research. They produce fairly large files in proprietary formats generated by applications designed for purposes that have nothing to do with terminology. If these methods are commonly used, it is because they are readily accessible and there are few terminology management programs being marketed to the individual translator.

Makeshift solutions:

There are other problems with such makeshift solutions, including the lack of management tools, inadequate search and retrieval tools, impoverished content (the record is reduced to term pairs), lack of efficiency, lack of good import/export tools, etc. But the biggest problem, and the one that can’t be so easily overcome, is the fact that such methods fail to reflect the most basic requirements of termbase design. Not only that, they fail to reflect the most fundamental premises of terminology.


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s