What it is
The section opened with the distinction between starting from a concept and starting from a form. It closes with what that distinction looks like once somebody has to build a database.
Three principles carry it, all three defined and numbered in the standards. Concept orientation is the principle whereby a concept entry describes a single concept. Term autonomy is the principle whereby all terms in an entry are treated as independent sub-units, so that each designation carries its own status, source and notes rather than hanging off another. Data elementarity is the principle whereby a data field contains only one data element, which rules out the common shortcut of putting a full form and its abbreviation in one field.
The three are not housekeeping. Concept orientation makes an entry answerable to a concept system, term autonomy makes a rating meaningful, and data elementarity makes a field searchable and exchangeable. Break any one and the entry still displays; it just stops being evidence.
What belongs on a record follows: designations, definition, context, notes, grammatical and subject labels, language and source identifiers. Working systems set a smaller mandatory core and leave the rest optional, which is where the trouble starts.
Where the authorities stand
The principles are standardized and stable, and the exchange format for term bases states plainly that the information it represents must be concept-oriented, deferring the other three definitions to the vocabulary and design standards.
The gap the record should end on is one field. The translation-oriented terminography standard defines a transfer comment as a note providing information on the degree of equivalence, directionality or other special features affecting equivalence between a designation in one language and another in a second. That is precisely what the rest of this section has been describing: the position on the equivalence scale, the direction in which a pair holds, and the caveat that makes it usable. It has a name, a definition and a number, and most working term bases do not have the field.
Kardos makes the argument from the inside, examining a large institutional term base and finding equivalence information scattered rather than recorded: the level of equivalence can be determined only through careful study of the references or through conceptual analysis, which is to say the user does the work again. Her proposal is concrete: indicate the system-specific nature of a term by putting the relevant country or organization in an origin field, and add the transfer comment as an optional data category.
The empirical picture supports the complaint. A survey of translators at international organizations, with an examination of their term banks' records, found that supporting notes and additional information for decision-making remained marginal, that notes were limited and sources often unreliable, and that several terminological solutions frequently coexisted inside one entry.
Sandrini supplies the position the section can close on, and it is not the obvious one. A comparative terminological entry should provide the information a decision requires while leaving the decision to the translator: the product provides the necessary information, but the translator takes the actual decisions. A descriptive resource should show how concepts and terms are used in their respective legal environments and should not seek artificial one-to-one equivalences. A record that hands over an insertable equivalent has answered the question the translator should have been helped to answer.
Consequences for translation
- One concept, one entry. Where a record has accumulated two concepts, every equivalent on it is ambiguous, and the ambiguity is invisible to whoever reads it next.
- Every designation carries its own status and source. A rating attached to an entry rather than to a term tells nobody which term was rated.
- Record directionality. A pair validated in one direction should say so: the reverse reading is often wrong and nothing on a symmetric-looking record warns against it.
- Where the transfer comment field is missing, its content still has to go somewhere. A free-text note doing that job beats another candidate equivalent.
Examples
EN>FR
- Source: notary public
- Target: notaire public
- Comment: Concept orientation failing quietly. A term bank record carries the English and French designations with a general definition, but the common-law officer authorized to administer oaths and the civil-law notary of Quebec are two concepts with different professional roles. The distinction survives only in a free-text observation, where nothing indexes it, instead of producing two entries.
FR>EN
- Source: fiducie
- Target: trust
- Comment: The directionality case, and the clearest argument for a transfer comment field. The French designation here is standardized common-law French, settled for rendering the common-law trust, and the record says so. It does not say the pair holds in one direction only, so a translator working out of a Quebec civil-law source will read fiducie, find trust, and write a concept the source did not use.
References
- Sandrini, P. (1999). "Legal Terminology: Some Aspects for a New Methodology." Hermes 22, 101 to 111, at pp. 102, 108 and 109.
- ISO 30042:2008. Systems to manage terminology, knowledge and content: TermBase eXchange. Geneva: International Organization for Standardization, clause 6.1.
- ISO 26162-1:2019. Management of terminology resources: Terminology databases, Part 1, Design. Geneva: International Organization for Standardization, entries 3.2.13, 3.2.14 and 3.2.16.
- ISO 12616-1:2021. Terminology work in support of multilingual communication: Part 1, Fundamentals of translation-oriented terminography. Geneva: International Organization for Standardization, entries 3.19, 3.24, 3.25, 3.28 and 3.34.
- Kardos, O. H. (2024). "Conceptual Equivalence in Multilingual EU Terminology: Challenges and Possible Solutions." Terminologija 31, 97 to 116, at p. 113.
- Prieto Ramos, F. and Guzmán, D. (2023). "Translating Legal Terms at International Organisations: Do Institutional Term Banks Meet Translators' Needs?" In J. Zhao, D. Li and V. L. C. Lei (eds), New Advances in Legal Translation and Interpreting, 243 to 264. Singapore: Springer, at pp. 250, 251 and 257.
- Translation Bureau of Canada, TERMIUM Plus, records for notary public and fiducie.