Right to the Correct Spelling of Personal Names
TODO: Erkki, please add details for the UNESCO Decision
International law
International law protects language – as a part of the ethnic, cultural, national and personal identity – against discrimination and assimilation:
Charter of the United Nations (1945), Art. 1.3, Art. 13(1) b, Art. 55 c
Universal Declaration of Human Rights (1948), Art. 2(1)
European Convention for the Protection of Human Rights and Fundamental Freedoms (1950), also known as the European Convention on Human Rights (ECHR), Art. 14
International Covenant on Civil and Political Rights (1966), Art. 26, Art. 27
Final Act of the Conference on Security and Cooperation in Europe (CSCE) in Helsinki (1975), Chapter 1, VII
- Undersigning States commit themselves to heed the provisions of the Act in national law-making (Chapter 1, X)
Document of the Copenhagen meeting of the conference on the Human dimension of the CSCE (1990), Nr. 31, Nr. 32
Framework Convention for the Protection of National Minorities (1995), Art. 5.1, 5.2
- UNESCO Decision
Private life and family life – comprising implicitly also names and family names – are likewise protected:
European Convention for the Protection of Human Rights and Fundamental Freedoms, Art. 8(1)+(2)
According to the European Court of Human Rights (ECHR Burghartz vs. Switzerland judgment), article 8 protects carrying the complete family name.
Concrete references to language-specific spelling of names in public administration are made in the law protecting minority languages:
European Charter for Regional or Minority Languages, Art. 10.5:
“The Parties undertake to allow the use or adoption of family names in the regional or minority languages, at the request of those concerned.”
Framework Convention for the Protection of National Minorities, Art. 11.1:
“The Parties undertake to recognise that every person belonging to a national minority has the right to use his or her surname (patronym) and first names in the minority language and the right to official recognition of them […].”
By common sense, these principles should also apply to the languages of other Member States. Indeed, this is the case in the ICCS Convention No. 14:
ICCS Convention No. 14 on the recording of surnames and forenames in civil status registers (1973), Article 2:
Where a record is to be made in a civil status register by an authority of a Contracting State and there is produced for that purpose a copy of or extract from a civil status record or some other document that shows the surnames and forenames in the same characters as those used in the language in which the record is to be made, those surnames and forenames shall be reproduced literally without alteration or translation. Any diacritic marks forming part of such forenames and surnames shall also be reproduced, even if such marks do not exist in the language in which the record is to be made.”
ICCS Convention No. 14 was signed by seven EU Member States (Germany, Austria, Belgium, Greece, Italy, Luxembourg, the Netherlands) and also by the candidate country Turkey. Six more EU countries (Spain, France, Hungary, Poland, Portugal, United Kingdom) as well as non-Member Switzerland and the candidate country Croatia are ICCS members and could join the Convention.
Although the Convention only regulates the spelling in civil registry entries (e.g. birth and marriage), it would make no sense at all to use a different spelling in other fields of public administration (registration, identity documents, health insurance cards, drivers licences, diplomas etc.), as this would compromise the consistency of personal data.
EU law
EU law mirrors international law in many points that are relevant for dealing with names from other languages:
- The Union is founded on the principles of respect for Human Rights (Art. 6.1 of the Treaty on European Union [TEU]).
- The Union shall respect the fundamental rights guaranteed by the European Convention for the Protection of Human Rights and Fundamental Freedoms (Art. 6.2 TEU).
- According to Article 12 of the Treaty establishing the European Community (TEC), discrimination based on nationality is forbidden.
In addition, the Charter of Fundamental Rights of the European Union, reference added to Art. 6 TEU by the (not yet fully ratified) Lisbon Treaty, will guarantee:
- respect for private and family life (Art. 7),
- equality before the law (Art. 20),
- no discrimination based on language (Art. 21) and
- respect for cultural and linguistic diversity (Art. 22).
National law
National law in European countries, however, is quite heterogeneous and is not always marked by the consequent implementation of international law:
Dutch law
In the Netherlands, the “Besluit standaardschrijfwijze persoonsgegevens” (decision on standard spelling of personal data) entered into force as early as in 1993. It referred to Dutch standards NEN 1888 and 5825 (defining an international set of Latin characters, probably MES-1) and set a four-year implementation period to equip all automated systems (not only civil registries) with adequate software by 01/01/1997.
German law
The service regulations for registrars (civil registry) state: “Diacritical marks (accents, hooks etc.) contained in names or other words of foreign origin shall be reproduced without alteration.” (§ 49.1, Dienstanweisung für die Standesbeamten und ihre Aufsichtsbehörden - DA -)
For electronic registration, the “Datensatz für das Meldewesen” (DSMeld, data set for registration) applies. However, this document is inconsistent: first it defines an exclusive German character set (A/a to Z/z plus letters with umlaut and lowercase sharp s) but later in Annex 4 it defines also an international set of ca. 300 Latin characters ("Sonderzeichensatz Meldewesen"), provided that the application supports the Universal Character Set (UCS).
A survey conducted in 2006 revealed that the Universal Character Set was supported only by 3 out of 13 communal registration software applications.
- A streamlined version of the DSMeld (UCS-only) is currently due for approval; the technical implementation period would end in May 2011 .
The application guidelines for identity documents (passport / ID card) are also inconsistent: they define an exclusive German character set (A/a to Z/z plus letters with umlaut and lowercase sharp s) but then also allow for other characters (e.g. accents) “for legal reasons”, though without specifying any of the aforementioned documents of international law (Leitfaden zum Ausfüllen eines Antrages auf Ausstellung eines Reisepasses / Leitfaden zum Ausfüllen eines Antrages auf Ausstellung eines Personalausweises).
Administrative regulations state that in case of doubt, spelling like in the civil registry entry shall prevail (6.2.2.1, Allgemeine Verwaltungsvorschriften zur Durchführung des Passgesetzes [Bundesanzeiger 179/2000, p. 18859]).
Moreover, a court ruling in 1992 clarified that altering of diacritics in official documents is de jure a change of name and thus prohibited (Bundesverwaltungsgericht 29/09/1992, reference mark 1 C 41/90, Juris database entry WBRE310554902).
Today, identity documents are de facto issued with all Latin special characters mentioned in the DSMeld.
For representation and exchange of digital data on the federal level, the document SAGA 4.0 (standards and architectures for eGovernment applications) applies. It makes the use of the Universal Character Set obligatory:
“Character sets / Obligatory: Unicode v4.x UTF-8 / In order to have enough code points for the various letters, cyphers and symbols existing worldwide, ISO 10646:2003 (also known as Unicode v4.x) in UTF-8 format should be used as character set for documents”.
The Polish-German Treaty
The bilateral treaty on good neighbourly relations and friendly cooperation of 17 June 1991 makes specific provisions regarding the spelling of names:
Art. 20 § 1+3: […] persons having German nationality in the Federal Republic of Germany who are of Polish origin or affirm that they belong to the community of Polish language, culture or tradition […] shall in particular have the right […] to express their given and family names in the mother tongue form […].
The same applies to Polish citizens with names of German origin. The most prominent linguistic example is the Polish EU Commissioner Danuta Hübner, featuring a non-Polish umlaut.
Austrian law
According to § 11(1) of the law on civil registry and §5(3) of the regulation on civil registry, names in Latin script have to be reproduced literally and with all diacritics.
Registration in Austria is a rare case of technological modernisation preceding law-making rather than following it: although neither registration law nor the corresponding executive order lays down the spectrum of permitted characters, the electronic central registry “Zentrales Melderegister” (2002) was programmed with state-of-the-art technologies (Java, XML) supporting the Universal Character Set. When the Austrian government drafted an E-Government law in 2003, the association of Austrian social insurance carriers (who process large amounts of personal data) wrote in an opinion:
“By regulation, a standard character set should be specified the use of which will be obligatory for electronic data exchange between the Federal level, regions, towns, municipalities and all public bodies. This character set should contain at least the special characters of the national minorities but also of the neighbouring countries, of the most common international commercial languages […] and some foreign languages occurring frequently in Austria (Turkish, Polish).”
When the E-Government law entered into force in 2004, it did not contain any references to character sets. But its Article 5 modified registration law by legalising the exchange of data between registries and civil registries, thus subjecting registries to § 11 of the law on civil registry. The Austrian E-Government ABC (2006) states:
“The present legal situation envisions mandatory use of diacritical marks in civil registries. Applications derived from this (e.g. registries) have to support diacritical marks in any case. All other applications should change over quickly in order to avoid inconsistencies.”
To support authorities and their service providers during the implementation, the Austrian Ministry of the Interior has issued a special recommendation named “Handbuch diakritische Zeichen” (manual on diacritical marks).
Swedish law
In most Swedish government agencies, the amount of possible characters for data processing was too limited to handle the either the increased number of foreign personal names, a consequence of strong immigration and free movement of workers, or the Sami minority language, though Sweden had ratified the European Charter for Regional or Minority Languages and obliged itself to “allow the use or adoption of family names in the regional or minority languages”. Moreover, within the framework of the EU and the Schengen Agreement, the exchange of information about persons between Swedish agencies and equivalent agencies in other European countries was increasing, requiring a common set of characters.
In 2005, the “Riktlinjer för teckenmängd för namnskrivning i svensk elektronisk förvaltning” (guidelines on the character set for writing of names in Swedish eGovernment) were established. They define all characters that government agencies and its staff shall be able to apply when typing, storing, processing and exchanging data containing the names of people, companies and geographic places. The guidelines comprise an Annex 1 with 284 Latin characters, for example characters to support the Sami and Roma minority languages, and an Annex 2 with 96 Latin characters from the Vietnamese language. For data exchange between public authorities, only signs that are specified in Annex 1 are allowed. Encoding shall be applied according to ISO/IEC 10646:2003 in transformation format UTF-8, also when the data is made available to individuals or companies. Advice on how to handle older registrations of names (i.e. where the diacritical marks were deleted) is “desirable”.
The document is not binding, but Annex 1 must be followed by those who claim that they follow the guidelines. The guidelines may later be transferred into mandatory regulations.
Swiss law
Switzerland has not signed the ICCS Convention No. 14 despite being an ICCS Member. In its “directives on spelling of names of foreign nationals”, the Federal Department of Justice and Police states that diacritical marks have to be adopted only if they are also present in one of the four official Swiss languages (German, French, Italian, and Romansh). Other diacritical marks are rigorously cut off (č --> c). These directives apply to all federal, cantonal and communal authorities.