Imprestige | The Orphan of Translation Practices
Languages
Tweets from Imprestige
You currently have access to a subset of Twitter API v2 endpoints and limited v1.1 endpoints (e.g. media post, oauth) only. If you need access to this endpoint, you may need a different access level. You can learn more here: https://developer.twitter.com/en/portal/product

The Orphan of Translation Practices

2014-10-07
  • Dictionary - (C) Jane M Sawyer (http://www.morguefile.com/creative/cohdra)

The translation industry requires all parties—the clients, the agencies and the translators—to pay attention to many things and factors, including business aspects like offering competitive recycling-based discount models, domain-specific aspects like researching context and performing QA tasks, as well as technical aspects like string length limitations and content conversions. One of the major aspects, however, lives its life as an orphan, since no one seems to nurse it, and this aspect is called quality terminology management.

Translation projects often include defining some sources in the form of databases, Excels or web applications to use for terminology look-up. Let’s call them collectively termbases. Basically there are two types of termbases to distinguish between for our purposes: managed and plain termbases. I call a termbase plain if there is no dedicated terminology management tool (which the most CAT vendor offer) behind it: an Excel sheet with a Source and Target column is a good example. And I call a termbase managed if CAT tools are used to create and maintain it.

Managed termbases uses way more columns than just a Source and a Target: they give room to define the part of speech, its gender, number, and administrative properties like status of approval and so on. Oh, and the most important field: definition. This most important field is very often not populated with definitions in managed termbases, this most important field is often missing from plain termbases, and this most important field is mostly ignored by many of the translators. But why is this field so important?

Because it is central to the philosophy of termbases. Termbases, in the ideal world, are not just lists of source-target term pairs, but the written representations of concepts in two or more languages. Can we say the correct translation of term in Hungarian is kifejezés? No, we can not give any answer until consulting the concept expressed in the source language with the sequence of characters making up the word term. Actually term may represent the concept of a standardized phrase for a specific domain (terminus, szakkifejezés, in Hungarian). But it can also represent the concept of a time period (időszak). Or that of a provision set forth in a contract (kikötés, rendelkezés). Or that of an agreement being effective for a definite amount of time (határozott idejű). Now look, how many translations we listed for a short and simple-looking word. And only the often missing definition describing the represented concept could tell good and bad translations of the word apart.

In my experience, translators mostly simply do not care concepts. They are bound to use termbases, and they write that translation of the source word which comes first from the termbase. If this is kifejezés, they will use a word meaning phrase, expression even when the term represents a period in time. It is true that many termbases simply miss the definition, but I don’t think the correct answer is to translate words instead of concepts. And now we come to a fundamental statement, worth writing twice:

Translators should translate concepts, not words, and quality inspectors should evaluate translations accordingly.

And it is very common that a termbase, be it extremely large, does not contain a concept. It was simply not entered before. This should never be seen as an instruction to use one other of the translations of the same source word just because it is listed.

Let’s care more for concepts and definitions in termbases, the two orphaned fields which could give significant help in improving translation quality.

(Cover photo by Jane M Sawyer)

About

Professional with a vision
Gusztáv, the founder and manager of Imprestige Ltd. is an experienced Business Analyst, IT Consultant, Quality Assurance Specialist and Professional Translator, sharing his time between IT projects and translation/localization projects.

Committed to professionalism and quality, in his IT roles he is an evangelist of well-organized software development ensuring agility, productivity and outstanding ROI during the application lifecycle. He is also passionate for creating great user experiences and highly intuitive solutions.

As a translator, Gusztáv belongs to those stimulating and pioneering the creation of clearly speaking translations, realizing that intelligibility and concise phrasing is inevitable in order to enable vendors to exploit the values embedded in their solutions.

For a detailed resume and reference information, please visit: http://gusztav.janvari.name