Terminology Management 6 August 2026
Multilingual Term Matching: Aligning a Concept Across Languages
Matching the same concept in two languages is not always one to one. With examples, we look at how multilingual term matching is done, at the concept anchor and at the gaps.
Matching a concept in two languages looks at first sight like placing two words side by side, but the job is rarely that simple. Multilingual term matching is the work of aligning the equivalents of the same concept in different languages, taking the concept itself as the shared anchor. The aim is for the term in every language to be attached to the same concept.
Contents
Hann’s dictionary, as a bilingual term resource, is a classic example of this kind of matching: every German term is given together with its English equivalent, its subject field code and its gender. In this article we look at how multilingual term matching is done, why the concept is the anchor and how equivalents that are not one to one are managed.
What Is Multilingual Term Matching?
Multilingual term matching means attaching the terms of different languages not to one another but to a shared concept. Instead of matching two languages directly, the equivalent in each language is placed under the same concept. When a third or a fourth language is added, the new term is attached to that same concept as well; the pairings no longer have to be built one by one.
The strength of this approach lies in the way it scales. In a five-language termbase, if you work from the concept you hold five equivalents for each concept; if you work from the word you have to build dozens of pairings between languages. The concept anchor brings order to that tangle. Each language is attached to the shared concept by a single link.
Matching aligns not only the equivalents but also their context. The subject field, the definition and the usage of the same concept in each language sit together. A nuance present in one language is carried across to another. Good matching manages not terms but the identity of the concept in all its languages.
The practical benefit of this structure is multilingual consistency. If an organisation’s product is translated into five languages, the same concept stays tied to the same definition in all five; no single language shifts the concept on its own. In a word-anchored arrangement, by contrast, each language keeps its own list and over time the languages drift apart. The concept anchor prevents that drift by bringing the languages together around a common centre.
Why Is the Concept the Shared Anchor?
Words do not overlap one to one across languages, but the concept is shared. That is why the anchor for matching should be the concept and not the word. A single word in one language may correspond to two separate words in another; putting the concept at the centre keeps that branching in order.
Hann’s dictionary illustrates the principle. German Achse is a single word, but in English it becomes either “axle” or “axis” depending on the concept. Had the matching been word-anchored, one German word would have blurred two separate English concepts. In a concept-anchored structure, “axle” and “axis” are separate concepts and each is attached to its own German equivalent.
Anchoring on the concept also frees the matching from having a direction. Word-anchored matching usually runs from one source language towards one target language and does not work in reverse. In a concept-anchored structure every language is equal; the matching can be read from any language to any other. This symmetry is the foundation of multilingual work: rather than treating one language as the source and another as the target, it keeps all languages in an equal position around the shared concept.
Concepts Without a One-to-One Equivalent
The hardest part of matching is that not every concept has a one-to-one equivalent in every language. A concept may be covered in one language by a single term and in another by several terms or by an explanation. This mismatch arises from the different ways languages divide up the world, and it cannot be concealed.
| Situation | Typical pattern | Matching decision |
|---|---|---|
| Full equivalent | One concept, one term | Match directly |
| Splitting | One concept, several terms | Note the condition |
| Gap | No equivalent | Add an explanation |
The way to manage these situations is to document them rather than hide them. If a concept has more than one equivalent in the target language, the record states which one is used under which condition. If there is no equivalent at all, a short explanation or a loan term is proposed. Making the gap visible steers the translator towards the right decision; concealing it leaves them at a loss.
There is a mismatch in the opposite direction too: two separate concepts in one language may be covered by a single term in another. A single target term then carries two different concepts, and without context there is no telling which one is meant. Keeping the matching at concept level prevents the confusion by holding the two concepts in separate records. Word-level matching, by contrast, squeezes them into one line and produces ambiguity.
Lessons from Hann’s Bilingual Structure
Hann’s dictionary shows the finer points of bilingual matching. Every German term is set out with its English equivalent and separated by subject field code; where the same German word has different English equivalents in different fields, each is given separately. This is proof that the matching is done at concept level and not at word level.
The dictionary also notes regional differences. British and American usage for the same concept are marked separately; one equivalent holds in British English, another in American English. This shows that even a single target such as “English” contains matching decisions of its own. A language divides more finely than is usually assumed.
Another lesson is that grammatical information belongs in the matching too. Hann’s dictionary records the gender of German terms, because without it the term cannot be used correctly. Multilingual matching has to carry not only the equivalent but also the linguistic information needed to use that equivalent correctly. Matching aligns usage, not merely words. The material in our knowledge centre illustrates these details.
Putting the Matching on Record
Multilingual matching is held most cleanly in a concept-oriented record structure. The concept and its definition sit at the centre; beneath them come the preferred term for each language, its synonyms, its subject field and its grammar. Adding a new language calls not for a new concept tree but for one more equivalent under each existing concept.
The record can also note how reliable the matching is. Where one equivalent is complete and verified while another is provisional and contested, that difference is valuable to a translator. Carrying these uncertainties into the record gives the next user an honest picture. The degree of certainty is information just as much as certainty itself.
A well-recorded matching multiplies an organisation’s language assets. A concept network built once becomes more valuable with every new language, because each language is added on top of the same concepts. Rather than scattered bilingual glossaries, a single concept-centred termbase is the solid ground of multilingual production. Matching is the operation that lays that ground; established early, it pays off again and again for every language added later.
Managing the Gaps Between Languages
In every multilingual termbase, some concepts are left without an equivalent in some languages. These gaps are not a defect but a consequence of the nature of languages. What matters is not to ignore a gap but to mark it plainly and propose an interim solution. A marked gap is far safer than a hidden one.
One way of managing gaps is to use a descriptive equivalent: where no single term exists, the concept is conveyed by a short definition. Another is to adopt a loan term provisionally and wait for an equivalent to settle over time. Whichever is chosen, the decision has to go on record so that every translator deals with the same gap in the same way.
Gaps can also close over time. A concept with no equivalent today may acquire an established term in the target language within a few years. Gaps should therefore be reviewed again during regular maintenance. Matching is not a frozen table but a process that lives alongside the languages. Marking a gap honestly today also makes the update easier when the gap closes tomorrow.
The Türkiye context
For Turkish, multilingual matching calls for particular care because of the structure of the language. An English or German technical term is often rendered in Turkish by a noun compound; instead of a one-to-one word there is a compound. This makes it essential to match at concept level rather than at word level. A concept-anchored structure brings order to this flexibility of Turkish.
Another local reality is that the Turkish equivalent of many concepts has not yet settled, which means gaps between languages appear frequently on the Turkish side. A concept of English origin is sometimes borrowed in Turkish, sometimes translated and sometimes circulates in both forms at once; which of them is to be preferred has to be decided in the matching.
At our office in Ankara Kızılay we build the matching around a concept anchor on multilingual projects, and where the Turkish equivalent is uncertain we mark the gap plainly and agree an interim solution with the client. The termbase then stays in an honest structure that can be updated as the Turkish equivalent settles.
What does this mean for you?
If you are an organisation producing in more than one language, or about to, the structure of your term matching decides the fate of your multilingual consistency. A word-anchored start collapses as the languages multiply. For sound matching:
- Build the matching around the concept rather than the word; attach every language to the shared concept.
- Do not hide concepts that have no one-to-one equivalent; document them with an explanation.
- Add regional and grammatical differences to the record.
- Review the gaps between languages again during regular maintenance.
On multilingual projects we build the matching around a concept anchor and document the gaps honestly; our translation services cover this work. To have us assess your multilingual terminology needs, get in touch through the quote form.
Frequently asked questions
What is multilingual term matching?
It is the work of aligning the equivalents of the same concept in different languages, taking the concept as the shared anchor. The aim is for the term in every language to be attached to the same concept and for consistency between languages to be secured.
Does every concept have a one-to-one equivalent in every language?
No. Some concepts are covered in one language by a single term and in another by several terms or by an explanation. These gaps are the part of matching that demands the most care.
Why is the concept the anchor in matching?
Because words do not overlap one to one across languages, whereas the concept is shared. A structure that puts the concept at the centre attaches the equivalent in each language to the same concept and so produces consistent matching.
How are gaps between languages managed?
The gap is noted plainly in the record; where needed, an explanation, several equivalents or a condition of use is added. Documenting a gap rather than hiding it steers the translator towards the right decision.
Sources
- Hann, M. (1992). The Key to Technical Translation, Vol. 2. John Benjamins.
- Wüster, E. (1979). Einführung in die allgemeine Terminologielehre. Springer.
- ISO 1087:2019. Terminology work and terminology science – Vocabulary.