What Is Subject-Field Coding? The Subject Label That Ties a Term to the Right Meaning

The quickest indication of which sense a term carries is usually a short label attached beside it. Subject-field coding means giving every term this label, which states the subject field it belongs to (automotive, chemistry, nuclear and so on). The code reduces the dozens of possible equivalents of a polysemous term to the single correct one.

Contents

Hann’s dictionary is built on this logic from beginning to end. Abbreviations such as PHYS, MATS, CHEM, ELEC, MECH and NUCL say which engineering world each equivalent belongs to. This article looks at the decisive role of subject-field coding and subject classification in terminology management, at its limits, and at the mistakes most commonly made in practice. Our aim is to show how coding can be made a convenience rather than a burden.

What Does Subject-Field Coding Solve?

Subject-field coding fixes the meaning of a term before context does. When you see the entry “Achse” in a dictionary, the word on its own can mean both “axis” and “axle”. But with a PHYS code beside it, it means “axis of rotation”; with an AUTO code, “rear axle”. The code takes the first decision on the translator’s behalf.

This is in effect a disambiguation step. A term without a code leaves the translator alone with the question “which meaning?”; a coded term already carries the answer. Because every German compound in Hann’s dictionary is listed with its own subject-field code, the reader can tell apart dozens of terms formed from the same root without confusing them. The code turns the dictionary from a list into a decision tool.

Another function of the subject-field code is that it speeds up searching. In a base holding thousands of terms, a translator who wants to filter out only the equivalents in one particular field reaches an answer within seconds thanks to the code. Without a code, the same search means wading through dozens of irrelevant equivalents. Coding is thus one of the rare techniques that raise accuracy and speed at the same time.

Main Fields and Related Fields

Hann’s system arranges fields in layers. The main fields come first: Physics, Materials Science, Chemistry, Electrical, Mechanical and Nuclear Engineering. Their related fields are then listed: Aeronautics, Automotive, Electronics, Construction, Computing and the like.

Code	Field	Main field it belongs to
PHYS	Physics	Main field
AERO	Aeronautics	Physics/Mechanics
AUTO	Automotive	Mechanical
SEMI	Semiconductor	Electronics
NUCL	Nuclear	Main field

This arrangement ties a term not only to a single box but also to the relations between boxes. An automotive term also leans on the conceptual world of mechanical engineering.

Rather than covering every main field, Hann’s list picks the smallest set that is enough to tell terms apart. It is the minimum set of fields needed to separate the concepts of the Polyseme Dictionary, the largest of the dictionaries. This measured stance matters: as the number of fields rises the system becomes more detailed, but it also becomes more fragile. A good field list aims at discrimination, not coverage.

Subfields: Why Does Classification Grow Deeper?

Main fields are often not enough on their own. In Hann’s dictionary, Physics is divided into subfields such as Acoustics, Electrostatics, Fluid Dynamics, Magnetism, Optics and Wave Mechanics. Each subfield has its own abbreviation: ACU, ELSC, FLUD, MAGN, OPT, WAVE.

The reason for this deepening is practical. In the Polyseme Dictionary, the largest of them, fine distinctions are needed to keep terms apart; whether a wave term is used in acoustics or in optics changes the equivalent completely. The smaller dictionaries in the book, by contrast, use only the main field codes for the sake of simplicity. The depth of the classification is therefore proportional to the size of the dictionary and to how polysemous its terms are. A classification deeper than necessary only adds pointless recording work.

There is a practical test: a subfield should be opened only if it genuinely helps to tell two terms apart. The distinction between “optics” and “acoustics” is necessary because it separates different equivalents of the same wave term. A subfield that distinguishes no term at all merely makes the decision harder at the moment of recording. Classification should therefore grow as the need arises, and should not be built in as much detail as possible from the outset.

Descriptors: Relation Information Beyond the Code

A subject-field code says where a term is used, but it does not show the relations between terms. Hann uses separate descriptors for that. What a term is a “part” of, what it is a “type” of, and which terms it stands in “contrast” with are all marked with short symbols.

The dictionary entry dynamo, p: bicycle, t: generator illustrates this: a dynamo is a “part” of a bicycle and a “type” of generator. In the same way, the disc brake entry records its contrasting term alongside it with the note “ct: drum brake”. Combined with the subject-field code, these descriptors both position the term and relate it to its neighbours.

The practical contribution of descriptors to translation is that they make it easier to tell close concepts apart. A translator who knows the contrasting term will not slip accidentally between “disc brake” and “drum brake”. Subject classification thus turns from a flat list of labels into a meaningful network of concepts. Current data category standards define similar relation fields as well; the principle has not changed since 1992, only the medium that carries it.

Applying Coding in a Termbase

In an electronic termbase, subject-field coding usually appears as a field called “subject field”. When this field is filled in on every record, the different senses of the same term are separated into distinct records and searches can be narrowed by code.

The most critical point in practice is not to leave field values to free text. If the translator types the subject field by hand each time, variants such as “automotive”, “Automotive” and “auto” multiply and the code loses its function. A predefined, closed list should be used instead, and whoever creates the record simply picks from it. This small restriction keeps the base consistently searchable even years later.

A well-designed subject classification gains value as an organisation grows. A new translator looking at the base does not have to guess which product line or technical field a term belongs to. In our knowledge centre, the material on setting up a termbase explains with examples how to define subject fields. Coding takes some effort at the initial set-up, but it repays that effort on every search.

The Limits of Coding and Common Mistakes

Subject-field coding is a powerful tool, but not a limitless one. The most frequent mistake is building a classification more detailed than necessary. Even Hann’s dictionary uses all the subfields only in the largest dictionary in practice; the smaller ones come down to the main field codes alone for simplicity. The lesson is clear: a classification should be exactly fine-grained enough to separate terms in reality, and no more.

The second common mistake is forcing the hierarchy. As the appendix to the book notes, a hierarchical code becomes confusing where the subordinate terms do not stand in the same kind of relation to the superordinate term. A term cannot be both a “part” and a “type” of a broader concept; if the coding ignores that distinction, the base becomes unreliable. The third mistake is letting field names drift over time: calling a field “electronics” today and “embedded systems” tomorrow makes the older records invisible. The field list should therefore change more slowly than the terms, and every change should be documented.

The Türkiye context

Most organisations doing technical translation in Türkiye use subject-field coding as scattered notes rather than as a conscious system. Yet the polysemous technical words of Turkish need coding at least as much as German does. Words such as “iletken”, “gerilim”, “yük” and “alan” become entangled in a base without a field label. Subject classification is the cheapest way of preventing that confusion.

A local difficulty is that Turkish field names have never been standardised. One organisation may say “otomotiv” where another says “taşıt teknolojisi”, and this inconsistency renders the codes useless. At our office in Ankara Kızılay, we recommend agreeing a short, fixed field list with the client at the start of a project. A plain list of fifteen to twenty fields meets the needs of most organisations and makes sure that everyone uses the same label with the same meaning. Once this list is settled, every later project proceeds on the same ground and a shared terminological language grows up inside the organisation.

What does this mean for you?

For anyone who has had a termbase built, or is about to, subject-field coding decides whether the base will be long-lived. An uncoded base becomes muddled after a few hundred terms; a coded one stays searchable even at thousands. For a solid start:

1. Draw up a short list of the technical fields your organisation actually works in.
2. Give each field one fixed label name, and merge synonymous field names.
3. Arrange the fields in two layers: main field and, where needed, subfield.
4. Never leave the subject field blank on a new term record.

These four steps keep the base searchable even when it reaches thousands of terms. On corporate projects we design this field list together with you, and as part of our translation services we deliver your termbase coded and searchable. To have us review your existing list, you can write to us through the contact page.

Frequently asked questions

What is subject-field coding?

It means attaching to a term record a short label that states the subject field the term belongs to (automotive, chemistry or nuclear, for example). This label fixes the correct equivalent for polysemous terms.

Is there a difference between a subject-field code and subject classification?

A subject-field code is a single label; subject classification is a hierarchical arrangement of those labels. Main fields are divided into subfields, and the term takes a place in that tree.

Why should subject-field codes be hierarchical?

Because a broad field such as physics divides into subfields such as acoustics, optics and mechanics. A hierarchy positions the term at both the general and the specific level, which allows more precise searching.

Is a subject-field code needed in a small dictionary?

It is if the terms are polysemous. In a narrow, single-field dictionary the burden of coding may be unnecessary; but in a mixed base covering more than one technical field, the code is the cheapest and most durable disambiguation tool available.

References

• Hann, M. (1992). The Key to Technical Translation, Vol. 2. John Benjamins.
• ISO 12620:2019. Management of terminology resources – Data categories.
• Sager, J. C. (1990). A Practical Course in Terminology Processing. John Benjamins.