13 Names, Dates, People, and Places

Table of contents

This chapter describes a module which may be used for the encoding of names and other phrases descriptive of persons, places, or organizations, in a manner more detailed than that possible using the elements already provided for these purposes in the Core module. In section 3.5 Names, Numbers, Dates, Abbreviations, and Addresses it was noted that the elements provided in the core module allow an encoder to specify that a given text segment is a proper noun, or a referring string, and to specify the kind of object named or referred to only by supplying a value for the type attribute. The elements provided by the present module allow the encoder to supply a detailed sub-structure for such referring strings, and to distinguish explicitly between names of persons, places, and organizations.

This module also provides elements for the representation of information about the person, place, or organization to which a given name is understood to refer and to represent the name itself, independently of its application. In simple terms, where the core module allows one simply to represent that a given piece of text is a name, this module allows one further to represent a personal name, to represent the person being named, and to represent the canonical name being used. A similar range is provided for names of places and organizations. The main intended applications for this module are in biographical, historical, or geographical data systems such as gazetteers and biographical databases, where these are to be integrated with encoded texts.

The chapter begins by discussing attributes common to many of the elements discussed in the remaining parts of the chapter (13.1 Attribute Classes Defined by this Module) before discussing specifically the elements provided for the encoding of component parts of personal names (section 13.2.1 Personal Names), place names (section 13.2.3 Place Names) and organizational names (section 13.2.2 Organizational Names). Elements for encoding personal and organizational data are discussed in section 13.3 Biographical and Prosopographical Data. Elements for the encoding of geographical data are discussed in section 13.3.4 Places. Finally, elements for encoding onomastic data are discussed in 13.3.5 Names and Nyms, and the detailed encoding of dates and times is described in section 13.3.6 Dates and Times.

13.1 Attribute Classes Defined by this Module

Most of the elements made available by this chapter share some important characteristics which are expressed by their membership in specific attribute classes. Members of the class att.naming have specialized attributes which support linkage of a naming element with the entity (person, place, organization) being named; members of the class att.datable have specialized attributes which support a number of ways of normalizing the date or time of the data encoded by the element concerned.

13.1.1 Linking Names and their Referents

The class att.naming is a subclass of the class att.canonical, from which it inherits the following attributes:
  • att.canonical provides attributes which can be used to associate a representation such as a name or title with canonical information about the object being named or referenced.
    keyprovides an externally-defined means of identifying the entity (or entities) being named, using a coded value of some kind.
    ref(reference) provides an explicit means of locating a full definition for the entity being named by means of one or more URIs.
As discussed elsewhere, these attributes provide two different ways of associating any sort of name with its referent. For cases where all that is required is to provide some minimal information about the person name, for example their occupation or status, the att.naming class also provides a simple role attribute. It also provides an additional attribute, which allows the name itself to be associated with a base or canonical form:
  • att.naming provides attributes common to elements which refer to named persons, places, organizations etc.
    rolemay be used to specify further information about the entity referenced by this name, for example the occupation of a person, or the status of a place.
    nymRef(reference to the canonical name) provides a means of locating the canonical form (nym) of the names associated with the object named by the element bearing it.
The encoder may use these attributes in combination as appropriate. For example:
That silly man
<name role="politician" type="person">David Paul Brown</name> has suffered ...
The ref attribute should be used wherever it is possible to supply a direct link such as a URI to indicate the location of canonical information about the referent.
That silly man
<name ref="#DPB1" type="person">David Paul Brown</name> has suffered ...
This encoding requires that there exist somewhere a person element with the identifier DPB1, which will contain canonical information about this particular person, marked up using the elements discussed in 13.3 Biographical and Prosopographical Data below. The same element might alternatively be provided by some other document, of course, which the same attribute could refer to by means of a URI, as explained in 16.2 Pointing Mechanisms:
That silly man
<name
  ref="http://www.example.com/personography.xml#DPB1"
  type="person">David Paul Brown</name> has suffered
...
More than one URI may be supplied if the name refers to more than one person. For example, assuming the existence of another person element for Mrs Brown, with identifier EBB1, a reference to ‘the Browns’ might be encoded
That wretched pair
<name ref="#DPB1 #EBB1" type="person">the Browns</name> came to dine
...
The key attribute is provided for cases where no such direct link is required: for example because resolution of the reference is carried out by some local convention, or because the encoder judges that no such resolution is necessary. As an example of the first case, a project might maintain its own local database system containing canonical information about persons and places, each entry in which is accessed by means of some system-specific identifier constructed in a project-specific way from the value supplied for the key attribute.42 As an example of the second case, consider the use of well-established codifications such as country or airport codes, which it is probably unnecessary for an encoder to expand further:
I never fly from <name key="LHR" type="place">Heathrow Airport</name>
to
<name key="FR" type="place">France</name>

However, as explained in 3.5.1 Referring Strings, interchange is improved by use of tag URIs in ref instead of key.

The nymRef attribute has a more specialized use, where it is the name itself which is of interest rather than the person, place, or organization being named. See section 13.3.5 Names and Nyms for further discussion.

Some members of the att.naming class are also members of the att.responsibility class, from which they inherit the following attributes:

  • att.responsibility provides attributes indicating who is responsible for something asserted by the markup and the degree of certainty associated with it.
    resp(responsible party) indicates the agency responsible for the intervention or interpretation, for example an editor or transcriber.
    cert(certainty) signifies the degree of certainty associated with the intervention or interpretation.

This enables an encoder to record the agency responsible for a given assertion (for example, the name) and the confidence placed in that assertion by the encoder. Examples are given below.

13.1.2 Dating Attributes

Members of the att.datable class share the following attributes:

  • att.datable provides attributes for normalization of elements that contain dates, times, or datable events.
    periodsupplies a pointer to some location defining a named period of time within which the datable item is understood to have occurred.
  • att.datable.w3c provides attributes for normalization of elements that contain datable events using the W3C datatypes.
    whensupplies the value of the date or time in a standard form, e.g. yyyy-mm-dd.
    notBeforespecifies the earliest possible date for the event in standard form, e.g. yyyy-mm-dd.
    notAfterspecifies the latest possible date for the event in standard form, e.g. yyyy-mm-dd.
    fromindicates the starting point of the period in standard form, e.g. yyyy-mm-dd.
    toindicates the ending point of the period in standard form, e.g. yyyy-mm-dd.

The when attribute is used to specify a normalized form for any temporal expression, independently of how it is represented in the text, as in the following example:
<date when="1807-06-09">June 9th</date> The
period is approaching which will terminate my present
copartnership. On the <date when="1808-01-01">1st Jany.</date> next,
it expires by its own limitation.
The period attribute provides a convenient way of associating an event or date with a named period. Its value is a pointer which should indicate some other element where the period concerned is more precisely defined. A convenient location for such definitions is the taxonomy element in the classDecl (classification declaration) in the encodingDesc of a TEI Header. A taxonomy may contain simply a bibliographic reference to an external definition for it. More usefully, it may also contain a series of category elements, each with an identifier and a description. The identifier can then be used as the target for a period attribute. For example, a taxonomy of named periods might be defined as follows:
<taxonomy xml:id="greekperiods">
 <category xml:id="tyranny">
  <catDesc>Before 510 BC</catDesc>
 </category>
 <category xml:id="classical">
  <catDesc>Between 510 and 323 BC</catDesc>
 </category>
 <category xml:id="hellenistic">
  <catDesc>
   <ref
     target="http://www.wikipedia.com/wiki/Hellenistic">Hellenistic</ref>. Commonly treated as
  <date notBefore="-0323" notAfter="-0031">from
       the death of Alexander to the Roman conquest.</date>
  </catDesc>
 </category>
 <category xml:id="roman">
  <catDesc>
   <ref
     target="http://www.wikipedia.com/wiki/Roman_Empire">Roman</ref>
  </catDesc>
 </category>
 <category xml:id="christian">
  <catDesc> The Christian period technically starts at the
     birth of Jesus, but in
     practice is considered to date from the conversion of Constantine
     in <date when="0312">312 AD</date>. </catDesc>
 </category>
</taxonomy>
With these definitions in place, any datable event may be associated with a specific period:
<placeName period="#christian">Stauropolis</placeName>
The other dating attributes provided by this class support a wide range of methods of specifying temporal information in a normalized form. Some simple examples follow:
<birth when="1857-03-15">15 March 1857.</birth>
<birth notBefore="1857-03-01" notAfter="1857-04-30">Some time
in March or April of 1857.</birth>
<residence from="1857-03-01" to="1857-04-30">In March and April of 1857.</residence>
<residence from="1857-03-01" notAfter="1857-04-30">From the 1st of March to
some time in April of 1857.</residence>

Normalization of date and time values permits the efficient processing of data (for example, to determine whether one event precedes or follows another). These examples all use the W3C standard format for representation of dates and times. Further examples, and discussion of some alternative approaches to normalization are given in section 13.3.6.3 More Expressive Normalizations below.

13.2 Names

13.2.1 Personal Names

The core rs and name elements can distinguish names in a text but are insufficiently powerful to mark their internal components or structure. To conduct nominal record linkage or even to create an alphabetically sorted list of personal names, it is important to distinguish between a family name, a forename and an honorary title. Similarly, when confronted with a string such as ‘John, by the grace of God, king of England, lord of Ireland, duke of Normandy and Aquitaine, and count of Anjou’, the analyst will often wish to distinguish amongst the various constituent elements present, since they provide additional information about the status, occupation, or residence of the person to whom the name belongs. The following elements are provided for these and related purposes:

  • persName (personal name) contains a proper noun or proper-noun phrase referring to a person, possibly including one or more of the person's forenames, surnames, honorifics, added names, etc.
  • surname contains a family (inherited) name, as opposed to a given, baptismal, or nick name.
  • forename contains a forename, given or baptismal name.
  • roleName contains a name component which indicates that the referent has a particular role or position in society, such as an official title or rank.
  • addName (additional name) contains an additional name component, such as a nickname, epithet, or alias, or any other descriptive phrase used within a personal name.
  • nameLink (name link) contains a connecting phrase or link used within a name but not regarded as part of it, such as van der or of.
  • genName (generational name component) contains a name component used to distinguish otherwise similar names on the basis of the relative ages or generations of the persons named.

In addition to the att.naming attributes mentioned above, all of the above elements are members of the class att.personal, and thus share the following attributes:

  • att.personal (attributes for components of names usually, but not necessarily, personal names) common attributes for those elements which form part of a name usually, but not necessarily, a personal name.
    fullindicates whether the name component is given in full, as an abbreviation or simply as an initial.
    sortspecifies the sort order of the name component in relation to others within the name.

The persName element may be used in preference to the general name element irrespective of whether or not the components of the personal name are also to be marked. The element persName is synonymous with the element <name type="person">, except that its type attribute allows for further subcategorization of the personal name itself, for example as a married, maiden, pen, pseudo, or religious name. Consequently the following examples are equivalent:
That silly man
<rs ref="tag:projectname.org,2012:DPB1" type="person">David Paul Brown</rs> has suffered the
furniture of his office to be seized
the third time for rent.
That silly man
<rs ref="tag:projectname.org,2012:DPB1" type="person">
 <name>David Paul Brown</name>
</rs> has suffered ...
That silly man
<name ref="tag:projectname.org,2012:DPB1" type="person">David Paul Brown</name> has suffered ...
That silly man
<persName ref="tag:projectname.org,2012:DPB1">David Paul Brown</persName> has suffered ...

The persName element is more powerful than the rs and name elements because distinctive name components occurring within it can be marked as such.

Many cultures distinguish between a family or inherited surname and additional personal names, often known as given names. These should be tagged using the surname and forename elements respectively and may occur in any order:
<persName>
 <surname>Roosevelt</surname>,
<forename>Franklin</forename>
 <forename>Delano</forename>
</persName>
<persName>
 <forename>Franklin</forename>
 <forename>Delano</forename>
 <surname>Roosevelt</surname>
</persName>
The type attribute may be used with both forename and surname elements to provide further culture- or project-specific detail about the name component, for example:
<persName>
 <forename type="first">Franklin</forename>
 <forename type="middle">Delano</forename>
 <surname>Roosevelt</surname>
</persName>
<persName>
 <forename type="given">Margaret</forename>
 <forename type="unused">Hilda</forename>
 <surname type="maiden">Roberts</surname>
 <surname type="married">Thatcher</surname>
</persName>
<persName type="religious">Muhammad Ali</persName>
<persName>
 <forename>Norman</forename>
 <surname type="complex">St John Stevas</surname>
</persName>
Values for the type attribute are not constrained, and may be chosen as appropriate to the encoding needs of the project. They may be used to distinguish different kinds of forename or surname, as well as to indicate the function a name component fills within the whole. In this example, we indicate that a surname is toponymic, and also point to the specific place name from which it is derived:
<persName>
 <forename>Johan</forename>
 <surname type="toponymic" ref="#dystvold">Dystvold</surname>
</persName>
<!-- ... -->
<placeName xml:id="dystvold">Dystvold</placeName>
The value complex was suggested above for the not uncommon case where the whole of a surname is composed of several other surname elements. These nested surnames may be individually tagged as well, together with appropriate type values:
<persName>
 <forename>Kara</forename>
 <surname type="complex">
  <surname type="paternal">Hattersley</surname>-
 <surname type="maternal">Smith</surname>
 </surname>
</persName>
The full attribute may be used to indicate whether a name is an abbreviation, initials, or given in full:
<persName>
 <forename full="abb">Maggie</forename>
 <surname>Thatcher</surname>
</persName>
These elements may be applied as the encoder considers appropriate, including cases where phrases or expressions are used to stand for surnames or forenames, as in the following:
<s>
 <persName>
  <forename>Peter</forename>
  <surname>son of Herbert</surname>
 </persName> gives the king 40 m. for
having custody of the land and heir of <persName>
  <forename>John</forename>
  <surname>son of Hugh</surname>
 </persName>...
</s>
Similarly, patronymics may be treated as forenames, thus:
... but it remained for
<persName>
 <forename>Snorri</forename>
 <forename>Sturluson</forename>
</persName>
to combine the two traditions in cyclic form.
When a patronymic is used as a surname, however (e.g. by an individual who otherwise would have no surname, but lives in a culture which requires surnames), it may be tagged as such:
Even <persName>
 <forename>Finnur</forename>
 <surname>Jonsson</surname>
</persName>
acknowledged the artificiality of the procedure...
Alternatively, it may be felt more appropriate to mark a patronymic as a distinct kind of name, neither a forename nor a surname, using the addName element:
<persName>
 <forename>Egill</forename>
 <addName type="patronym">Skallagrmsson</addName>
</persName>
In the following example, the type attribute is used to distinguish a patronymic from other forenames:
<persName ref="tag:projectname.org,2012:pn9">
 <forename sort="2">Sergei</forename>
 <forename sort="3" type="patronym">Mikhailovic</forename>
 <surname sort="1">Uspensky</surname>
</persName>

This example also demonstrates the use of the sort attribute common to all members of the model.persNamePart class; its effect is to state the sequence in which forename and surname elements should be combined when constructing a sort key for the name.

Some names include generational or dynastic information, such as a number, or phrases such as ‘Junior’, or ‘the Elder’; these qualifications may also be used to distinguish similarly named but unrelated people. In either case, the genName element may be used to distinguish such labels from other parts of the name, as in the following examples:
<persName ref="tag:projectname.org,2012:HEMA1">
 <surname>Marques</surname>
 <genName>Junior</genName>,
<forename>Henrique</forename>
</persName>
<persName>
 <forename>Charles</forename>
 <genName>II</genName>
</persName>
<persName>
 <forename>Rudolf</forename>
 <genName>II</genName>
 <surname type="dynasty">Hapsburg</surname>
</persName>
<persName>
 <surname>Smith</surname>
 <genName>Minor</genName>
</persName>
It is also often convenient to distinguish phrases (historically similar to the generational labels mentioned above) used to link parts of a name together, such as ‘von’, ‘of’, ‘de’ etc. It is often a matter of arbitrary choice whether such components are regarded as part of the surname or not; the nameLink element is provided as a means of making clear what the correct usage should be in a given case, as in the following examples:
<persName ref="tag:projectname.org,2012:DUDO1">
 <roleName type="honorific" full="abb">Mme</roleName>
 <nameLink>de la</nameLink>
 <surname>Rochefoucault</surname>
</persName>
<persName>
 <forename>Walter</forename>
 <surname>de la Mare</surname>
</persName>

Finally, the addName and roleName elements are used to mark all name components other than those already listed. The distinction between them is that a roleName encloses an associated name component such as an aristocratic or official title which exists in some sense independently of its bearer. The distinction is not always a clear one. As elsewhere, the type attribute may be used with either element to supply culture- or application- specific distinctions. Some typical values for this attribute for names in the Western European tradition follow:

nobility
An inherited or life-time title of nobility such as Lord, Viscount, Baron, etc.
honorific
An academic or other honorific prefixed to a name e.g. Doctor, Professor, Mrs., etc.
office
Membership of some elected or appointed organization such as President, Governor, etc.
military
Military rank such as Colonel.
epithet
A traditional descriptive phrase or nick-name such as The Hammer, The Great, etc.

Note, however, that the role a person has in a given context (such as witness, defendant, etc. in a legal document) should not be encoded using the roleName element, since this is intended to mark roles which function as part of a person's name, not the role of the person bearing the name om general. Information about roles, occupations, etc. of a person are encoded within the person element discussed below in 13.3 Biographical and Prosopographical Data.

Here are some further examples of the usage of these elements:
<persName ref="tag:projectname.org,2012:PGK1">
 <roleName type="nobility">Princess</roleName>
 <forename>Grace</forename>
</persName>
<persName ref="tag:projectname.org,2012:GRMO1" type="pseudo">
 <addName type="honorific">Grandma</addName>
 <surname>Moses</surname>
</persName>
<persName ref="tag:projectname.org,2012:SLWICL1">
 <roleName type="office">President</roleName>
 <forename>Bill</forename>
 <surname>Clinton</surname>
</persName>
<persName ref="tag:projectname.org,2012:MOGA1">
 <roleName type="military">Colonel</roleName>
 <surname>Gaddafi</surname>
</persName>
<persName ref="tag:projectname.org,2012:FRTG1">
 <forename>Frederick</forename>
 <addName type="epithet">the Great</addName>
</persName>
A name may have any combination of the above elements:
<persName ref="tag:projectname.org,2012:EGBR1">
 <roleName type="office">Governor</roleName>
 <forename sort="2">Edmund</forename>
 <forename full="init" sort="3">G.</forename>
 <addName type="nick">Jerry</addName>
 <addName type="epithet">Moonbeam</addName>
 <surname sort="1">Brown</surname>
 <genName full="abb">Jr</genName>.

</persName>

Although highly flexible, these mechanisms for marking personal name components will not cater for every personal name, nor for every processing need. Where the internal structure of personal names is highly complex or where name components are particularly ambiguous, feature structures are recommended as the most appropriate mechanism to mark and analyze them, as further discussed in chapter 18 Feature Structures.

13.2.2 Organizational Names

In these Guidelines, we use the term ‘organization’ for any named collection of people regarded as a single unit. Typical examples include institutions such as ‘Harvard College’ or ‘the BBC’ and businesses such as ‘Apple’ or ‘Google’ but also racial or ethnic groupings or political factions where these are regarded as forming a single agency such as ‘the Scythians’ or ‘the Militant Tendency’. Giving a loosely-defined group of individuals a name often serves a particular political or social agenda and an analysis of the way such phrases are constructed and used may therefore be of considerable importance to the social historian, even where the objective existence of an ‘organization’ in this sense is harder to demonstrate than that of (say) a named person. In the case of businesses or other formally constituted institutions, the component parts of an organizational name may help to characterize the organization in terms of its perceived geographical location, ownership, likely number of employees, management structure, etc.

Like names of persons or places, organizational names can be marked up as referring strings or as proper names with the rs or name elements respectively. The element orgName is provided for use where it is desired to distinguish organizational names more explicitly.

  • orgName (organization name) contains an organizational name.

This element is a member of the same attribute classes as persName, as discussed above in 13.1.1 Linking Names and their Referents.

The orgName element may be used to mark up any form of organizational name:
About a year back, a question of considerable
interest was agitated in the
<orgName type="voluntary" ref="tag:projectname.org,2012:PAS1">Pennsyla. Abolition Society</orgName>
This encoding is equivalent to, but more specific than, either of the following representations:
About a year back, a question of considerable
interest was agitated in the <rs ref="tag:projectname.org,2012:PAS1" type="org">
 <name>Pennsyla. Abolition Society</name>
</rs>.
About a year back, a question of considerable
interest was agitated in the
<name ref="tag:projectname.org,2012:PAS1" type="org">Pennsyla. Abolition
Society</name>.
As shown above, like the rs and name elements, the orgName element has a key attribute with which an external identifier such as a database key can be assigned to the organization name, and also a ref attribute which can be used to point directly to an org element containing information about the organization itself (see further 13.3.3 Organizational Data). Its type attribute should be used to characterize the name (rather than the organization), for example as an acronym:
Mr Frost will be able to earn an extra fee from
<orgName type="acronym">BSkyB</orgName>
rather than the
<orgName type="acronym">BBC</orgName>
as a phrase:
The feeling in <country>Canada</country> is one of
strong aversion to the <orgName type="phrase">United
States Government</orgName>, and of
predilection for self-government under
the
<orgName type="phrase">English Crown</orgName>
<orgName>The Justified Ancients of Mu Mu</orgName>
or as a composite of other kinds of name:
<orgName type="partnerNames">
 <surname>Ernst</surname> &amp; <surname>Young</surname>
</orgName>
The components of an organization's name may include place names as well as personal names:
A spokesman from
<orgName type="regional">
 <orgName>IBM</orgName>
 <country>UK</country>
</orgName> said ...
or role names:
THE TICKET which you will receive herewith has been formed by
the <orgName>Democratic Whig <name type="role">party</name>
</orgName> after the most careful deliberation,
with a reference to all the great objects of NATIONAL, STATE,
COUNTY and CITY concern, and with a single eye to the <hi>Welfare and Best Interests of the Community</hi>.
As indicated above, organizational names may also be specified hierarchically particularly where the named organization is itself a department or a branch of a larger organizational entity. ‘The Department of Modern History, Glasgow University’ is an example:
<orgName>
 <orgName>Department of Modern History</orgName>
 <orgName>
  <name type="city">Glasgow</name>
  <name type="role">University</name>
 </orgName>
</orgName>

13.2.3 Place Names

Like other proper nouns or noun phrases used as names, place names can simply be marked up with the rs element, or with the name element. For cartographers and historical geographers, however, the component parts of a place name provide important information about the relation between the name and some spot in space and time. They also provide important evidence in historical linguistics.

These Guidelines distinguish three ways of referring to places. A place name (represented using the placeName element) may consist of one or more names for hierarchically-organized geo-political or administrative units (see section 13.2.3.1 Geo-political Place Names). A place named simply in terms of geographical features such as mountains or rivers is represented using the geogName element (see section 13.2.3.2 Geographic Names). Finally, an expression consisting of phrases expressing spatial or other kinds of relationship between other kinds of named place may itself be regarded as a way of referring to a place, and hence as a kind of named place (see section 13.2.3.3 Relative Place Names).

  • placeName contains an absolute or relative place name.
  • geogName (geographical name) identifies a name associated with some geographical feature such as Windrush Valley or Mount Sinai.

As members of the att.naming class, all of these elements bear the attributes key, ref, and nymRef mentioned above. These attributes are primarily useful as a means of linking a place name with information about a specific place. Recommendations for the encoding of information about a place, as distinct from its name, are provided in 13.3.4 Places below.

Like the persName element discussed in section 13.2.1 Personal Names, the placeName element may be regarded simply as an abbreviation for the elements <name type="place"> or <rs type="place">. The following encodings are thus equivalent:43
After
spending some time in our <rs ref="tag:projectname.org,2012:NY1" type="place">modern <name ref="tag:projectname.org,2012:BA1" type="place">Babylon</name>
</rs>, <name ref="tag:projectname.org,2012:NY1" type="place">New York</name>, I have proceeded to
the <rs ref="tag:projectname.org,2012:PH1" type="place">City of Brotherly Love</rs>.
After spending some
time in our <placeName ref="tag:projectname.org,2012:NY1">modern
<placeName ref="tag:projectname.org,2012:BA1">Babylon</placeName>
</placeName>,
<placeName ref="tag:projectname.org,2012:NY1">New
York</placeName>, I have proceeded to the <placeName ref="tag:projectname.org,2012:PH1">City of
Brotherly Love</placeName>.
13.2.3.1 Geo-political Place Names
A place name may contain text with no indication of its internal structure:
<placeName>Rochester,
NY</placeName>
More usually however, a place name of this kind will be further analysed in terms of its constitutive geo-political or administrative units. These may be arranged in ascending sequence according to their size or administrative importance, for example: ‘Rochester, New York’, or as a single such unit, for example ‘Belgium’. These Guidelines provide a hierarchy of generic element names, each of which may be more exactly specified by means of a type attribute:
  • district contains the name of any kind of subdivision of a settlement, such as a parish, ward, or other administrative or geographic unit.
  • settlement contains the name of a settlement such as a city, town, or village identified as a single geo-political or administrative unit.
  • region contains the name of an administrative unit such as a state, province, or county, larger than a settlement, but smaller than a country.
  • country (country) contains the name of a geo-political unit, such as a nation, country, colony, or commonwealth, larger than or administratively superior to a region and smaller than a bloc.
  • bloc (bloc) contains the name of a geo-political unit consisting of two or more nation states or countries.
These elements are all members of the model.placeNamePart class, members of which may be used anywhere that text is permitted, including within each other as in the following examples:
<placeName>
 <settlement type="city">Rochester</settlement>,
<region type="state">New York</region>
</placeName>
<placeName ref="tag:projectname.org,2012:LSEA1">
 <country type="nation">Laos</country>,
<bloc type="sub-continent">Southeast Asia</bloc>
</placeName>
<placeName>
 <district type="arondissement">6ème</district>
 <settlement type="city">Paris, </settlement>
 <country>France</country>
</placeName>
13.2.3.2 Geographic Names
Places may also be named in terms of geographic features such as mountains, lakes, or rivers, independently of geo-political units. The geogName is provided to mark up such names, as an alternative to the placeName element discussed above. For example:
<geogName ref="tag:projectname.org,2012:MIRI1" type="river">Mississippi River</geogName>

In addition to the usual phrase level elements, the geogName element may contain the following specialized element:

  • geogFeat (geographical feature name) contains a common noun identifying some geographical feature contained within a geographic name, such as valley, mount, etc.

Where the geogFeat element is used to characterize the kind of geographic feature being named, the name element will generally also be used to mark the associated proper noun or noun phrase:
<geogName ref="tag:projectname.org,2012:MIRI1" type="river">
 <name>Mississippi</name>
 <geogFeat>River</geogFeat>
</geogName>
A more complex example, showing a variety of practices, follows:
The isolated ridge
separates two great corridors which run from <name ref="tag:projectname.org,2012:GLCO1" type="place">Glencoe</name> into
<geogName ref="tag:projectname.org,2012:GLET1" type="glen">
 <geogFeat>Glen</geogFeat>
 <name>Etive</name>
</geogName>, the
<geogName ref="tag:projectname.org,2012:LAGA1" type="hill">
 <geogFeat xml:lang="gd">Lairig</geogFeat>
 <name>Gartain</name>
</geogName> and the

<geogName ref="tag:projectname.org,2012:LAEI1" type="hill">
 <geogFeat xml:lang="gd">Lairig</geogFeat>
 <name>Eilde</name>
</geogName>

The Gaelic word lairig may be glossed as sloping hill face. The most efficient way of including this information in the above encoding would be to create a separate nym element for this component of the name and then point to it using the nymRef attribute, as further discussed in 13.3.5 Names and Nyms.

13.2.3.3 Relative Place Names

All the place name specifications so far discussed are absolute, in the sense that they define only one place. A place may however be specified in terms of its relationship to another place, for example ‘10 miles northeast of Paris’ or ‘near the top of Mount Sinai’. These relative place names will contain a place name which acts as a referent (e.g. ‘Paris’ and ‘Mount Sinai’). They will also contain a word or phrase indicating the position of the place being named in relation to the referent (e.g. ‘the top of’, ‘north of’). A distance, possibly only vaguely specified, between the referent place and the place being indicated may also be present (e.g. ‘10 miles’, ‘near’).

Relative place names may be encoded using the following elements in combination with either a placeName or a geogName element.
  • offset marks that part of a relative temporal or spatial expression which indicates the direction of the offset between the two place names, dates, or times involved in the expression.
  • measure contains a word or phrase referring to some quantity of an object or commodity, usually comprising a number, a unit, and a commodity name.
Some examples of relative place names are:
<placeName ref="tag:projectname.org,2012:NRPA1">
 <offset>near the top of</offset>
 <geogName>
  <geogFeat>Mount</geogFeat>
  <name>Sinai</name>
 </geogName>
</placeName>
<placeName>
 <measure>20 km</measure>
 <offset>north of</offset>
 <settlement type="city">Paris</settlement>
</placeName>
If desired, the distance specified may be normalized using the unit and quantity attributes of measure:
<placeName ref="tag:projectname.org,2012:Duncan">
 <measure unit="km" quantity="17.7">11 miles</measure>
 <offset>Northwest of</offset>
 <settlement type="city">Providence</settlement>, <region type="state">RI</region>
</placeName>

The internal structure of place names is like that of personal names—complex and subject to an enormous amount of variation across time and different cultures. The recommendations in this section should however be adequate for a majority of users and applications; they may be extended using the mechanisms described in chapter 23.3 Personalization and Customization to add new elements to the existing classes. When the focus of interest is on the name components themselves, as in place name studies for example, the elements discussed in 13.3.5 Names and Nyms may also be of use. Alternatively, the meaning structure itself may be represented using feature structures (18 Feature Structures).

13.3 Biographical and Prosopographical Data

This module defines a number of special purpose elements which can be used to markup biographical, historical, and prosopographical data. We envisage three basic types of users and uses for these elements. The first is the person interested in creating or converting an existing set of biographical records, for example of the type found in a Dictionary of National Biography. The second is the person hoping to create or convert a database-like collection of information about a group of people, possibly but not necessarily the people referenced in a marked-up collection of documents or a text-corpus. The third type would be those interested in the creation or conversion of biographical or CV-like structured texts for use in such applications as Human Resource management.

To cater for this diversity, these Guidelines propose a flexible approach, in which encoders may choose for themselves the degree of prescription appropriate to their needs. If one were interested, for example, in converting existing DNB-type records, and wanted to preserve the text as is, the person element (see 13.3.2 The Person Element) could simply contain the text of an article, placed within p elements, possibly using elements such as name or date to mark up features of that text. For a more structured entry, however, one would extract the data and place information contained by the text, and encode it directly using the more specific elements described in this section.

13.3.1 Basic Principles

Information about people, places, and organizations, of whatever type, essentially comprises a series of statements or assertions relating to:

  • characteristics or traits which do not, by and large, change over time
  • characteristics or states which hold true only at a specific time
  • events or incidents which may lead to a change of state or, less frequently, trait.

‘Characteristics’ or ‘traits’ are typically independent of an individual's volition or action and can be either physical, such as sex or hair and eye colour, or cultural, such as ethnicity, caste, or faith. The distinction is not entirely straightforward, however: while sex is fairly obviously a physical trait, gender should rather be regarded as culturally determined, and the division of mankind into different ‘races’, proposed by early (white European) anthropologists on the basis of physical characteristics such as skin colour, hair type and skull measurements, is by many modern cultural anthropologists now considered to be more a social or mental construct than an objective biological fact. Furthermore, while some characteristics will obviously change over time, hair colour for example, none, in principle—not even sex—is immutable.

‘States’ include, for example, marital status, place of residence and position or occupation. Such states have a definite duration, that is, they have a beginning and an end and are typically a consequence of the individual's own action or that of others.

By ‘changes in state’ are meant the events in a person's life such as birth, marriage, or appointment to office; such events will normally be associated with a specific date or a fairly narrow date-range. Changes in states can also cause or be caused by changes in characteristics. Any statement or assertion on any of these aspects of a person's life will be based on some source, possibly multiple sources, possibly contradictory. Taking all this into account it follows that each such statement or assertion needs to be able to be documented, put into a time frame and be relatable to other statements or assertions of the same or any of the other types.

The elements defined by the module described in this chapter may, for the most part, all be regarded as specializations of one or other of the above three classes. Generic elements for state, trait, and event are also defined:

  • state contains a description of some status or quality attributed to a person, place, or organization often at some specific time or for a specific date range.
  • trait contains a description of some status or quality attributed to a person, place, or organization typically, but not necessarily, independent of the volition or action of the holder and usually not at some specific time or for a specific date range.
  • event (event) contains data relating to any kind of significant event associated with a person, place, or organization.
    whereindicates the location of an event by pointing to a place element
  • listEvent (list of events) contains a list of descriptions, each of which provides information about an identifiable event.

13.3.2 The Person Element

Information about a person, as distinct from references to a person, for example by name, is grouped together within a person element. Information about a group of people regarded as a single entity (for example ‘the audience’ of a performance) may be encoded using the personGrp element. Note however that information about a group of people with a distinct identity (for example a named theatrical troupe) should be recorded using the org element described in section 13.3.3 Organizational Data below.

These elements may appear only within a listPerson element, which groups such descriptions together, and optionally also describes relationships amongst the people listed.

  • listPerson (list of persons) contains a list of descriptions, each of which provides information about an identifiable person or a group of people, for example the participants in a language interaction, or the people referred to in a historical source.
  • listRelation provides information about relationships identified amongst people, places, and organizations, either informally as prose or as formally expressed relation links.

One or more listPerson elements may be supplied within the particDesc (participant description) element in the profileDesc element of a TEI Header (see 2.4 The Profile Description). Like other forms of list, however, the listPerson can also appear within the body of a text when the module defined by this chapter is included in a schema.

The type attribute may be used to distinguish lists of people of different kinds where this is considered convenient:
<profileDesc>
 <particDesc>
  <listPerson type="historical">
   <person xml:id="ART1">
    <persName>Arthur</persName>
   </person>
   <person xml:id="BERT1">
    <persName>Bertrand</persName>
   </person>
<!-- ... -->
  </listPerson>
  <listPerson type="mythological">
   <person xml:id="ART2">
    <persName>Arthur</persName>
   </person>
   <person xml:id="BERT2">
    <persName>Bertrand</persName>
   </person>
<!-- ... -->
  </listPerson>
 </particDesc>
</profileDesc>

The person element carries several attributes. First, as a member of att.editLike, which is a subclass of the att.responsibility class, it carries the usual attributes for providing details about the information recorded for that person itself, such as its reliability or source:

  • att.responsibility provides attributes indicating who is responsible for something asserted by the markup and the degree of certainty associated with it.
    cert(certainty) signifies the degree of certainty associated with the intervention or interpretation.
    resp(responsible party) indicates the agency responsible for the intervention or interpretation, for example an editor or transcriber.
  • att.editLike provides attributes describing the nature of an encoded scholarly intervention or interpretation of any kind.
    evidenceindicates the nature of the evidence supporting the reliability or accuracy of the intervention or interpretation.
    sourcecontains a list of one or more pointers indicating sources supporting the given intervention or interpretation.

In addition, a small number of very commonly used personal properties may be recorded using attributes specific to person and personGrp :

  • person provides information about an identifiable individual, for example a participant in a language interaction, or a person referred to in a historical source.
    rolespecifies a primary role or classification for the person.
    sexspecifies the sex of the person.
    agespecifies an age group for the person.
  • personGrp (personal group) describes a group of individuals treated as a single person for analytic purposes.

These attributes are intended for use where only a small amount of data is to be encoded in a more or less normalized form, possibly for many person elements, for example when encoding basic facts about respondents to a questionnaire. When however a more detailed encoding is required for all kinds of information about a person, for example in a historical gazeteer, then it will be more appropriate to use the elements age, sex and others described elsewhere in this chapter.

Note that the age attribute is not intended to record the person's age expressed in years, months, or other temporal unit. Rather it is intended to record into which age bracket, for the purposes of some analysis, the person falls. A simple (perhaps too simple to be useful) binary classification of age brackets would be child and adult. The actual age brackets useful to various projects are likely to be varied and idiosyncratic, and thus these Guidelines make no particular recommendation as to possible values. Instead, individual projects are recommended to define the values they use in their own customization file, using a declaration like the following:
<elementSpec ident="person" module="namesdates" mode="change">
 <attList>
  <attDef mode="replace" ident="age">
   <datatype>
    <rng:ref name="data.enumerated"/>
   </datatype>
   <valList type="closed">
    <valItem ident="child">
     <desc>less than 18 years of age</desc>
    </valItem>
    <valItem ident="adult">
     <desc>18 to 65 years of age</desc>
    </valItem>
    <valItem ident="retired">
     <desc>over 65 years of age</desc>
    </valItem>
   </valList>
  </attDef>
 </attList>
</elementSpec>
The above declaration, were it properly placed in a customization file, establishes that the age attribute of person has only three possible values, child, adult, and retired. For more information on customization see 23.3 Personalization and Customization.

The person element may contain many sub-elements, each specifying a different property of the person being described. The remainder of this section describes these more specific elements. For convenience, these elements are grouped into three classes, corresponding with the tripartite division outlined above: one for traits, one for states and one for events. Each class contains both specific elements for common types of biographical information, and a generic element for other, user-defined, types of information.

All the elements in these three classes belong to the attribute class att.datable, which provides the following attributes:

  • att.datable.w3c provides attributes for normalization of elements that contain datable events using the W3C datatypes.
    whensupplies the value of the date or time in a standard form, e.g. yyyy-mm-dd.
    notBeforespecifies the earliest possible date for the event in standard form, e.g. yyyy-mm-dd.
    notAfterspecifies the latest possible date for the event in standard form, e.g. yyyy-mm-dd.
    fromindicates the starting point of the period in standard form, e.g. yyyy-mm-dd.
    toindicates the ending point of the period in standard form, e.g. yyyy-mm-dd.

as discussed in 13.1 Attribute Classes Defined by this Module above.

13.3.2.1 Personal Characteristics
The model.persStateLike class contains elements describing physical or socially-constructed characteristics, traits, or states of a person. Members of the class comprise the following specific elements:
  • faith specifies the faith, religion, or belief set of a person.
  • langKnowledge (language knowledge) summarizes the state of a person's linguistic knowledge, either as prose or by a list of langKnown elements.
  • nationality contains an informal description of a person's present or past nationality or citizenship.
  • sex specifies the sex of a person.
  • age (age) specifies the age of a person.
  • socecStatus (socio-economic status) contains an informal description of a person's perceived social or economic status.
  • persName (personal name) contains a proper noun or proper-noun phrase referring to a person, possibly including one or more of the person's forenames, surnames, honorifics, added names, etc.
  • occupation contains an informal description of a person's trade, profession or occupation.
  • residence (residence) describes a person's present or past places of residence.
  • affiliation (affiliation) contains an informal description of a person's present or past affiliation with some organization, for example an employer or sponsor.
  • education contains a description of the educational experience of a person.
  • floruit contains information about a person's period of activity.
  • state contains a description of some status or quality attributed to a person, place, or organization often at some specific time or for a specific date range.
  • trait contains a description of some status or quality attributed to a person, place, or organization typically, but not necessarily, independent of the volition or action of the holder and usually not at some specific time or for a specific date range.
All, apart from langKnowledge, allow content of ordinary prose containing phrase-level elements.
<socecStatus ref="tag:projectname.org,2012:AB1">Status AB1 in the RG Classification scheme</socecStatus>

The meanings of concepts such as sex, nationality, or age are highly culturally-dependent, and the encoder should take particular care to be explicit about any assumptions underlying their usage of them. For example, when recording personal age in different cultures, there may be different assumptions about the point from which age is reckoned. A statement of the practice adopted in a given encoding may usefully be provided in the editorialDecl element discussed in 2.3.3 The Editorial Practices Declaration.

The langKnowledge element contains either paragraphs or a number of langKnown elements; it may take a tags attribute, which provides one or more standard codes or ‘tag’s for the languages. The langKnown element must have a tag attribute, which indicates the language with the same kind of ‘language tag’. These ‘language tags’ are discussed in detail in vi.1. Language identification.

Furthermore, the langKnown element also has a level attribute to indicate the level of the person's competence in the language. It is thus possible either to say:
<langKnowledge tags="ff fr wo en">
 <p>Speaks fluent Fulani, Wolof, and French. Some knowledge of English.</p>
</langKnowledge>
or
<langKnowledge>
 <langKnown level="fluent" tag="ff">Fulani</langKnown>
 <langKnown level="fluent" tag="wo">Wolof</langKnown>
 <langKnown level="fluent" tag="fr">French</langKnown>
 <langKnown level="basic" tag="en">English</langKnown>
</langKnowledge>
The sex element carries a value attribute to give the ISO 5218:2004 values (1 for male, 2 for female, 9 for non-applicable, and 0 for unknown).
<sex value="2">female</sex>
As elsewhere, these coded values may be used as an alternative to or normalization of the actual descriptive text contained by the element. The previous example might equally well be given as
<sex value="2"/>
The generic trait and state elements are also a member of this class,
  • trait contains a description of some status or quality attributed to a person, place, or organization typically, but not necessarily, independent of the volition or action of the holder and usually not at some specific time or for a specific date range.
  • state contains a description of some status or quality attributed to a person, place, or organization often at some specific time or for a specific date range.
These element can be used to extend the range of information supplied about an individual's personal characteristics. Either may contain an optional label element, used to provide a human-readable specification for the characteristic concerned and a description of the feature itself supplied within a desc element. These may be followed by or one or more p elements supplying more detailed information about the trait. In either case, these may be followed by one or more notes or bibliographical references. The type, ref, and key attributes may be used to indicate a fuller definition of the combination of feature and value.
<trait type="ethnicity" key="alb">
 <label>Ethnicity</label>
 <desc>Ethnic Albanian.</desc>
</trait>
These elements are provided as a simple means of extending the set of descriptive features available in a standardized way. For example, there are no predefined elements for such features as eye or hair colour. If these are to be recorded, they may simply be added as new types of trait:
<trait type="physical">
 <label>eye colour</label>
 <desc>blue</desc>
</trait>
<trait type="physical">
 <label>hair colour</label>
 <desc>brown</desc>
</trait>

If none of the more specialised elements listed above is appropriate, then a choice must be made between the two generic elements trait and state. If you wish to distinguish between characteristics that are generally perceived to be transient and those which are generally considered unchanging, use state for the former, and trait for the latter. It may also be helpful to note that that traits are typically, but not necessarily, independent of the volition or action of the holder. If the distinction between state and trait is not considered relevant or useful, use state.

The persName element is repeatable and can, like all TEI elements, take the attribute xml:lang to indicate the language of the content of the element, as well as a type attribute to indicate the type of name, whether a nickname, maiden name, alternative form, etc. This is useful in cases where, for example, a person is known by a Latin name and also by any number of vernacular names, many or all of which may have claims to ‘authenticity’. In order to ensure uniformity, the method generally employed in the library world has been to accept the form found in some authority file, for example that of the American Library of Congress, as the ‘base’ or ‘neutral’ form. Feelings can run high on this matter, however, and people are often reluctant to accept as ‘neutral’ an overtly foreign form of the name of their local saint or hero. Within the person element any number of variant forms of a name can be given, with no prioritisation, and hence less likelihood of offence. The Icelandic scholar and manuscript collector Árni Magnússon, to give his name in standard modern Icelandic spelling, is known in Danish as Arne Magnusson, the form which he himself, as a long term resident of Denmark, generally used; there is also a Latinised form, Arnas Magnæus, which he used in his scholarly writings. All three forms can be given, and in any order:
<person xml:id="ArnMag">
 <persName xml:lang="is">Árni Magnússon</persName>
 <persName xml:lang="da">Arne Magnusson</persName>
 <persName xml:lang="la">Arnas Magnæus</persName>
</person>
At the other extreme, a person may be named periphrastically as in the following example:
<person xml:id="simon_son_of_richard2">
 <persName>Simon, son of Richard</persName>
 <residence>
  <placeName>
   <region>Essex</region>
  </placeName>
 </residence>
 <floruit notBefore="1219" notAfter="1223">1219-1223</floruit>
</person>
Alternatively, the generic name element may be used for all of the naming components in a description. For example, a description of the first living held by the Icelandic clergyman and poet Jón Oddsson Hjaltalín might be tagged as follows:
<state type="office" from="1777-04-07" to="1780-07-12">
 <p>Jón's first living — which he apparently accepted rather reluctantly — was at
 <name type="place">Háls í Hamarsfirði</name>, <name type="place">Múlasýsla</name>, to which
   he was presented on 7 April 1777. He was ordained the following
   month and spent three years at Háls, but was never happy there,
   due largely to the general penury in which he was forced to live —
   a recurrent theme throughout the early part of his life. In June
   of 1780 the bishop recommended that Jón
   should <q xml:lang="da">promoveres til andet bedre kald, end det
     hand hidindtil har havt</q>, and on 12 July it was agreed that
   he should exchange livings with
 <name type="person" ref="tag:projectname.org,2012:ThorJon">sr. Þórður Jónsson</name> at
 <name type="place">Kálfafell á Síðu</name>,
 <name type="place">Skaftafellssýsla</name>.</p>
 <bibl>ÞÍ, Stms I.15, p. 733.</bibl>
 <bibl>ÞÍ, Stms I.17, p. 102.</bibl>
</state>
Similarly, the generic state or trait element may be used in preference to the more specific elements listed above:
<state type="nationality" notBefore="2002-01-15">
 <label>Nationality</label>
 <desc>American citizen from 15 January 2002.</desc>
</state>
is the same as:
<nationality notBefore="2002-01-15">American citizen from 15 January 2002.</nationality>
or even:
<nationality notBefore="2002-01-15" key="US"/>
13.3.2.2 Personal Events

The model.persEventLike class contains elements describing specific events in a person's history, for example birth, marriage, or appointment. These are not characteristics of an individual, but often cause an individual to gain such characteristics, or to enter a new state. Members of this class comprise the following elements:

  • birth (birth) contains information about a person's birth, such as its date and place.
  • death (death) contains information about a person's death, such as its date and place.

Only two specific elements (birth and death) are proposed. The generic element event has a similar content model to that of state and trait; the chief difference being that it can include a placeName element to identify the name of the place where the event occurred. It is used to describe any event in the life of an individual or organization.

In the following example, we give a brief summary of the wedding of Jane Burden to the English writer, designer, and socialist William Morris, encoded as an event element embedded within the person element used to record data about Morris, though we could equally well have embedded the event within the person element for Burden, or have given it as a freestanding event independent of either person element:
<person xml:id="WM">
<!-- ... -->
 <event type="marriage" when="1859-04-26">
  <label>Marriage</label>
  <desc>
   <name type="person" ref="#WM">William Morris</name> and <name
     type="person"
     ref="http://en.wikipedia.org/wiki/Jane_Burden">Jane Burden</name> were
     married at <name type="place">St Michael's Church, Ship Street, Oxford</name> on
  <date when="1859-04-26">26 April 1859</date>. The wedding was
     conducted by Morris's friend <name type="person" ref="#RWD">R. W.
       Dixon</name> with <name type="person" ref="#CBF">Charles
       Faulkner</name> as
     the best man. The bride was given away by her father,
  <name type="person" ref="#RB">Robert Burden</name>.
     According to the account that <name
     type="person"
     ref="http://en.wikipedia.org/wiki/Edward_Burne-Jones">Burne-Jones</name>
     gave <name type="person" ref="#JWM">Mackail</name>
   <quote>M. said to Dixon beforehand <said>Mind
         you don't call her Mary</said> but he did</quote>. The entry in the
     Register reads: <quote>William Morris, 25, Bachelor Gentleman, 13
       George Street, son of William Morris decd. Gentleman. Jane Burden,
       minor, spinster, 65 Holywell Street, d. of Robert Burden,
       Groom.</quote> The witnesses were Jane's parents and Faulkner. None of
     Morris's family attended the ceremony. Morris presented Jane with a
     plain gold ring bearing the London hallmark for 1858. She gave her
     husband a double-handled antique silver cup.</desc>
  <bibl>J. W. Mackail, <title>The Life of William Morris</title>, 1899.</bibl>
 </event>
</person>
<person xml:id="RB">
 <persName>Robert Burden</persName>
</person>
<person xml:id="RWD">
 <persName>R.W. Dixon</persName>
</person>
<person xml:id="CBF">
 <persName>Charles Faulkner</persName>
</person>
<person xml:id="EBJ">
 <persName>
  <forename>Edward</forename>
  <surname>Burne-Jones</surname>
 </persName>
</person>
<person xml:id="JWM">
 <persName>J.W. Mackail</persName>
</person>
In this example the ref attributes on the various name elements point either to an external source or to a person element within which other information about the person named may be found. As further discussed below (13.3.2.3 Personal Relationships), a relation element may then be used to link them in a more meaningful way:
<relation name="spouse" mutual="#WM #JBM"/>
<relation name="friend" mutual="#WM #RWD"/>
<relation name="parent" active="#RB" passive="#JBM"/>
As mentioned above, all these elements, both the specific and the generic, are members of the att.datable attribute class, which means they can be limited in terms of time. The following encoding, for example, demonstrates that the person named David Jones changed his name in 1966 to David Bowie:
<person xml:id="DB">
 <persName notAfter="1966">David Jones</persName>
 <persName notBefore="1966">David Bowie</persName>
</person>
All the generic elements are also members of the att.responsibility and att.editLike classes. These classes make available the attributes cert, to indicate the degree of certainty, resp, the agency responsible, and evidence, the nature of the evidence used. In this way it is possible, in the case of multiple and conflicting sources, to provide more than one view of what happened, as in the following example:
<event type="birth" resp="#XYZ" cert="high">
 <p>Born in <name type="place">Brixton</name> on 8 January 1947.</p>
</event>
<event type="birth" resp="#ABC" cert="low">
 <p>Born in <name type="place">Berkhamsted</name> on 9 January 1947.</p>
</event>
13.3.2.3 Personal Relationships

When the module defined by this chapter is included in a schema, the following two elements may be used to document relationships amongst the persons, places, or organizations identified:

  • listRelation provides information about relationships identified amongst people, places, and organizations, either informally as prose or as formally expressed relation links.
  • relation (relationship) describes any kind of relationship or linkage amongst a specified group of objects, places, events or people.
    namesupplies a name for the kind of relationship of which this is an instance.
    activeidentifies the ‘active’ participants in a non-mutual relationship, or all the participants in a mutual one.
    mutualsupplies a list of participants amongst all of whom the relationship holds equally.
    passiveidentifies the ‘passive’ participants in a non-mutual relationship.

These elements are both members of the att.typed class, from which they inherit the type and subtype attributes in the usual way. The value specified for either attribute on a listRelation element is implicitly applicable to all of its child relation elements, unless overridden.

A relationship, as defined here, may be any kind of describable link between specified participants. A participant (in this sense) might be a person, a place, or an organization. In the case of persons, therefore, a relationship might be a social relationship (such as employer/employee), a personal relationship (such as sibling, spouse, etc.) or something less precise such as ‘possessing shared knowledge’. A relationship may be mutual, in that all the participants engage in it on an equal footing (for example the ‘sibling’ relationship); or it may not be if participants are not identical with respect to their role in the relationship (for example, the ‘employer’ relationship). For non-mutual relationships, only two kinds of role are currently supported; they are named active and passive. These names are chosen to reflect the fact that non-mutual relations are directed, in the sense that they are most readily described by a transitive verb, or a verb phrase of the form is X of or is X to. The subject of the verb is classed as active; the direct object of the verb, or the object of the concluding preposition, as passive. Thus parents are ‘active’ and children ‘passive’ in the relationship ‘parent’ (interpreted as is parent of); the employer is ‘active’, the employee ‘passive’, in the relationship employs. These relationships can be inverted: parents are ‘passive’ and children ‘active’ in the relationship is child of; similarly ‘works for’ inverts the active and passive roles of ‘employs’.

For example:
<listRelation>
 <relation name="parent" active="#P1 #P2" passive="#P3 #P4"/>
 <relation name="spouse" mutual="#P1 #P2"/>
 <relation
   type="social"
   name="employer"
   active="#P1"
   passive="#P3 #P4"/>
</listRelation>
This example defines the relationships amongst a number of people not further described here; we assume however that each person has been allocated an identifier such as P1, P2, etc. which can be linked to using references such as #P1, #P2 etc. Then the above set of relation elements describe the following three relationships amongst the seven people referenced:
  • P1 and P2 are parents of P3 and P4.
  • P1 and P2 are linked in a mutual relationship called ‘spouse’—that is, P2 is the spouse of P1, and P1 is the spouse of P2.
  • P1 has the social relationship ‘employer’ with respect to P3 and P4.

Relationships within places and organizations are further discussed in the relevant section below. Relationships between for example organizations and places, or places and persons, may be handled in exactly the same way.

13.3.3 Organizational Data

The org and listOrg elements are used to store data about an organization such as its preferred name, its locations, or key persons within it.

  • org (organization) provides information about an identifiable organization such as a business, a tribe, or any other grouping of people.
  • listOrg (list of organizations) contains a list of elements, each of which provides information about an identifiable organization.

These elements are intended to be used in a way analogous to the place and person elements discussed elsewhere in this chapter, that is to provide a unique wrapper element for information about an entity, distinct from references to that entity which are typically encoded using a naming element such as <name type="org"> or orgName. The content of a naming element will represent the way an organization is named in a given context; the content of an org represents the information known to the encoder about that organization, gathered together in a single place, and independent of its textual realization.

An organization is not the same thing as a list or group of people because it has an identity of its own. That identity may be expressed solely in the existence of a name (for example ‘The Scythians’), but is likely to consist in the combination of that name with a number of events, traits, or states which are considered to apply to the organization itself, rather than any of its members. For example, a sports team might be defined in terms of its membership (a listPerson), its fixtures (a listPlace), its geographical affiliation (a placeName), or any combination of these. It will also have properties which may be used to categorize it in some way such as the kind of sport played, whether the team is amateur or professional, and so on: these are probably best dealt with by means of the type attribute. However, it is the name of the sports team alone which identifies it.

The content model for org permits any mixture of generic state, trait, or event elements: the presence of the orgName element described in 13.2.2 Organizational Names is however strongly recommended.

In other respects, the org element is used in much the same way as place or person. An organization may have different names at different times:
<org>
 <orgName notAfter="1960">The Silver Beetles</orgName>
 <orgName from="1960-08">The Beatles</orgName>
</org>
The names of the people making up an organization can also change over time, (if they are known at all). For example:
<org>
 <orgName notAfter="1960">The Silver Beetles</orgName>
 <orgName notBefore="1960">The Beatles</orgName>
 <state type="membership" from="1960-08" to="1962-05">
  <desc>
   <persName>John Lennon</persName>
   <persName>Paul McCartney</persName>
   <persName>George Harrison</persName>
   <persName>Stuart Sutcliffe</persName>
   <persName>Pete Best</persName>
  </desc>
 </state>
 <state type="membership" notBefore="1963">
  <desc>
   <persName>John Lennon</persName>
   <persName>Paul McCartney</persName>
   <persName>George Harrison</persName>
   <persName>Ringo Starr</persName>
  </desc>
 </state>
</org>
An org may contain subordinate orgs:
<org>
 <orgName>Oxford University Computing Services</orgName>
 <org>
  <orgName>Information and Support Group</orgName>
 </org>
 <org>
  <orgName>Infrastructure Group</orgName>
  <org>
   <orgName>Networking Team</orgName>
  </org>
  <org>
   <orgName>System Development Team</orgName>
  </org>
 </org>
 <org>
  <orgName>Learning Technologies Group</orgName>
 </org>
</org>
The following example demonstrates the use of the listOrg element to group together a number of org elements, each of which is defined solely by means of an informal description, itself containing other names.
<p>The TEI institutional hosts are: <listOrg>
  <org xml:id="bu">
   <orgName>Brown University</orgName>
   <desc>The host contribution is made jointly by the <name type="project">Brown University Women Writers Project</name> and the
   <orgName>Brown University Library's Center for Digital
         Initiatives</orgName>.</desc>
  </org>
  <org xml:id="na">
   <orgName>Nancy</orgName>
   <desc>Hosting is provided by a group of institutions located in
       Nancy, France, coordinated by <orgName>Loria</orgName> and also
       including <orgName>ATILF</orgName> and <orgName>INIST</orgName>.</desc>
  </org>
  <org xml:id="ou">
   <orgName>Oxford University</orgName>
   <desc>Hosting is provided by the <orgName>Research Technologies
         Service</orgName> at <orgName>Oxford University Computing
         Services</orgName>.</desc>
  </org>
  <org xml:id="uv">
   <orgName>University of Virginia</orgName>
   <desc>Virginia's host support comes jointly from the
   <orgName>Institute for Advanced Technology in the
         Humanities</orgName> and the <orgName>University of Virginia
         Library</orgName>.</desc>
  </org>
<!-- from http://www.tei-c.org/About/hosting.xml-->
 </listOrg>
</p>
In a more elaborated version of this example, the organizational names tagged using orgName might be linked using the key or ref attribute to a unique org element elsewhere.

13.3.4 Places

In 13.2.3 Place Names we discuss various ways of naming places such as towns, countries, etc. In much the same way as these Guidelines distinguish between the encoding of names for people and the encoding of other data about people, so they also distinguish between the encoding of names for places and the encoding of other data about places. In this section we present elements which may be used to record in a structured way data about places of any kind which might be named or referenced within a text. Such data may be useful as a way of normalizing or standardizing references to particular places, as the raw material for a gazetteer or similar reference document associated with a particular text or set of texts, or in conjunction with any form of geographical information system.

The following elements are provided for this purpose:

  • listPlace (list of places) contains a list of places, optionally followed by a list of relationships (other than containment) defined amongst them.
  • place contains data about a geographic location

The model.placeStateLike class contains elements describing characteristics of a place which have a definite duration, such as its name. Any member of the model.placeNamePart may be used for this purpose, since a place element will usually contain at least one, and possibly several, placeName-like elements indicating the names associated with it, by different people, in different languages, or at different times.

For example, the modern city of Lyon in France was in Roman times known as Lugdunum. Although the modern and the Roman city are not physically co-extensive, they have significant areas which overlap, and we may therefore wish to regard them as the same place, while supplying both names with an indication of the time period during which each was current.

A place is defined, however, by its physical location, which does not typically change over time. Locations may be specified in a number of ways: as a set of coordinates defining a point or an area on the surface of the earth, or by providing a description of how the place may be found, usually in terms of other place names. For example, we can identify the location of the Canadian city of London, either by specifying its latitute and longitude, or by specifying that we mean the city called London located in the province called Ontario within the country called Canada.

In addition we may wish to supply a brief characterization of the place identified, for example to state that it is a city, an administrative area such as a country, or a landmark of some kind such as a monument or a battlefield. If our typology of places is simple, the open ended type attribute is the easiest way to represent it: so we might say <place type="city">, <place type="battlefield"> etc.

Within the place element, the following elements may be used to provide more information about specific aspects of the place in a structured form:

  • placeName contains an absolute or relative place name.
  • location defines the location of a place as a set of geographical coordinates, in terms of other named geo-political entities, or as an address.

13.3.4.1 Varieties of Location

A location may be specified in one or more of the following ways:

  1. by supplying a string representing its coordinates in some standardized way within a geo element, as shown below
  2. by supplying one or more place name component elements (e.g. country, settlement etc.) to place it within a geo-political context
  3. by supplying a postal address, e.g. using the address element
  4. by supplying a brief textual description, e.g. using the desc element
  5. by using a non-TEI XML vocabulary such as the Geography Markup Language

We give examples of all of these methods in the remainder of this section.

The simplest method of specifying a location is by means of its geographic coordinates, supplied within the geo element. This may be used to supply any kind of positional information, using one of the many different geodetic systems available. Such systems vary in their format, in their scope or coverage, and more fundamentally in the reference frame (the ‘datum’) used for the coordinate system itself. The default recommended by these Guidelines is to supply a string containing two real numbers separated by whitespace, of which the first indicates latitude and the second longitude according to the 1984 World Geodetic System (WGS84); this is the system currently used by most GPS applications which TEI users are likely to encounter.44We might therefore record the information about the place known as ‘Lyon’ as follows:
<place xml:id="LYON1" type="city">
 <placeName notBefore="1400">Lyon</placeName>
 <placeName notAfter="0056">Lugdunum</placeName>
 <location>
  <geo>45.769559 4.834843</geo>
 </location>
</place>
Identifying Lyon by its geo-political status as a settlement within a country forming part of a larger political entity, we might represent the same ‘place’ as follows:
<place xml:id="LYON2">
 <placeName notBefore="1400">Lyon</placeName>
 <placeName notAfter="0056">Lugdunum</placeName>
 <location>
  <bloc>EU</bloc>
  <country>France</country>
 </location>
</place>
Elements such as bloc are specialized forms of placeName, as discussed in 13.2.3.1 Geo-political Place Names.
We may use the same procedure to represent the location of smaller places, such as a street or even an individual building:
<place type="building">
 <placeName>Brasserie Georges</placeName>
 <location>
  <country key="FR"/>
  <settlement type="city">Lyon</settlement>
  <district type="arrondissement">IIème</district>
  <district type="quartier">Perrache</district>
  <placeName type="street">Cours de Verdun</placeName>
 </location>
</place>
Note the use of the type attribute to categorize more precisely both the kind of place concerned (a building) and the kind of name used to locate it, for example by characterizing the generic district as an ‘arrondissement’, or a ‘quartier’.
We may also treat imaginary places in the same way:
<place type="imaginary">
 <placeName>Atlantis</placeName>
 <location>
  <offset>beyond</offset>
  <placeName>The Pillars of <persName>Hercules</persName>
  </placeName>
 </location>
</place>
A location sometimes resembles a set of instructions for finding a place, rather than a name:
<place xml:id="MYF">
 <placeName notAfter="1969">Yasgur's Farm</placeName>
 <placeName notBefore="1969">Woodstock Festival Site</placeName>
 <location>
  <measure>one mile</measure>
  <offset>north west of</offset>
  <settlement>Bethel</settlement>
  <region>New York</region>
 </location>
</place>
The element address may also be used to identify a location in terms of its postal or other address:
<place type="cemetery">
 <placeName>Protestant Cemetery</placeName>
 <placeName type="official" xml:lang="it">Cimitero Acattolico</placeName>
 <location type="geopolitical">
  <country>Italy</country>
  <settlement>Rome</settlement>
  <district>Testaccio</district>
 </location>
 <location type="address">
  <address>
   <addrLine>Via Caio Cestio, 6</addrLine>
   <addrLine>00153 Roma</addrLine>
  </address>
 </location>
</place>
When, as here, the same place is given multiple locations, the type attribute should be used to characterize the kind of location, as a means of indicating that these are alternative ways of identifying the same place, rather than that the place is spread across several locations.

The location element may thus identify a place to a greater or lesser degree of precision, using a variety of means: a name, a set of names, or a set of coordinates. The geo element introduced earlier is by default understood to supply a value expressed in a specific (and widely used) notation. If a location contains more than one geo, this is interpreted as being really the same place in the universe, but with different systems used to refer to it. If there is a lack of consensus about the location (of, for example, Camelot), more than one location should be used, each with its own geo.

By default, the content of geo is interpreted as following the standard known as the World Geodetic System (WGS). This may be modified, however, in two ways.

Firstly, the content of the geo element can be expressed some other way, that is, according to some different geodetic system. The decls attribute is used point to a geoDecl element defined in the document header, which describes a different datum.

Secondly, the element geo may be redefined to contain markup from a different XML vocabulary which is specifically designed to represent this kind of information. This technique is used throughout the Guidelines where specialized markup is required, for example to embed mathematical expressions or vector graphics, and is further described and exemplified in 23.3.4 Examples of Modification . For geographic information, suitable non-TEI vocabularies include:

  • the OpenGIS Geography Markup Language (GML) being defined by the OGC45
  • the Keyhole Markup Language (KML) used by Google Maps46

In the following example, we have defined the location of the place ‘Lyon’ using GML and indicated the two names associated with it at different times:
<place type="city">
 <placeName notBefore="1400">Lyon</placeName>
 <placeName notAfter="0056">Lugdunum</placeName>
 <location>
  <geo>
   <gml:Polygon>
    <gml:exterior>
     <gml:LinearRing> 45.256 -110.45 46.46 -109.48 43.84 -109.86 45.8 -109.2
           45.256 -110.45 </gml:LinearRing></gml:exterior></gml:Polygon>
  </geo>
 </location>
</place>

A bibl element may be used within location to indicate the source of the location information.

<location>
 <geo>53.226658 -0.541254</geo>
 <bibl>
  <title>Roman Inscriptions of Britain</title>, <idno>262</idno>
 </bibl>
</location>
13.3.4.2 Multiple Places
A place may contain other places. This containment relation can be directly modelled in XML: thus we can say that the towns of Vilnius and Kaunas are both in a place called Lithuania (or Lietuva) as follows:
<place>
 <country>Lithuania</country>
 <country xml:lang="lt">Lietuva</country>
 <place>
  <settlement>Vilnius</settlement>
 </place>
 <place>
  <settlement>Kaunas</settlement>
 </place>
</place>

This does not, of course, imply that Vilnius and Kaunas are the only places constituting Lithuania; only that they are within it. A separate place element may indicate that it is a part of Lithuania by supplying a relation element, as discussed below (13.3.4.4 Relations Between Places).

As a further example, the islands of Mauritius, Réunion, and Rodrigues are collectively known as the Mascarene Islands. Grouped together with Mauritius there are also several smaller offshore islands, with rather picturesque French names. These offshore islands do not however constitute an identifiable place as a whole. One way of representing this is as follows:
<place type="islandGroup">
 <placeName>Mascarene Islands</placeName>
 <placeName>Mascarenhas Archipelago</placeName>
 <place type="island">
  <placeName>Mauritius</placeName>
  <listPlace type="offshoreIslands">
   <place>
    <placeName>La roche qui pleure</placeName>
   </place>
   <place>
    <placeName>Île aux cerfs</placeName>
   </place>
  </listPlace>
 </place>
 <place type="island">
  <placeName>Rodrigues</placeName>
 </place>
 <place type="island">
  <placeName>Réunion</placeName>
 </place>
</place>
Here is a more complex example, showing the variety of names associated at different times and in different languages with a set of hierarchically grouped places—the settlement of Carmarthen Castle, within the town of Carmarthen, within the administrative county of Carmarthenshire, Wales.
<place xml:id="wales" type="country">
 <placeName xml:lang="cy">Cymru</placeName>
 <placeName xml:lang="en">Wales</placeName>
 <placeName xml:lang="la">Wallie</placeName>
 <placeName xml:lang="la">Wallia</placeName>
 <placeName xml:lang="fro">Le Waleis</placeName>
 <place xml:id="carmarthenshire" type="region">
  <region type="county" xml:lang="en" notBefore="1284">Carmarthenshire</region>
  <place xml:id="carmarthen" type="settlement">
   <placeName xml:lang="en">Carmarthen</placeName>
   <placeName xml:lang="la" notBefore="1090" notAfter="1300">Kaermerdin</placeName>
   <placeName xml:lang="cy">Caerfyrddin</placeName>
   <place xml:id="carmarthen_castle" type="castle">
    <settlement>castle of Carmarthen</settlement>
   </place>
  </place>
 </place>
</place>

As noted previously, country, region, and settlement are all specializations of the generic placeName element; they are not specializations of the place element. If it is desired to distinguish amongst kinds of place this can only be done by means of the type attribute as in the above example.

This use of multiple place elements should be distinguished from the (possibly simpler) case where a number of places with some property in common are being grouped together for convenience, for example, in a gazetteer. The listPlace element is provided as a means of grouping places together where there is no implication that the grouped elements constitute a distinct place. For example:
<place type="county">
 <placeName>Herefordshire</placeName>
 <listPlace type="villages">
  <place>
   <placeName>Abbey Dore</placeName>
   <location>
    <geo>51.969604 -2.893146</geo>
   </location>
  </place>
  <place>
   <placeName>Acton Beauchamp</placeName>
  </place>
<!-- etc -->
 </listPlace>
 <listPlace type="towns">
  <place>
   <placeName>Hereford</placeName>
  </place>
  <place>
   <placeName>Leominster</placeName>
  </place>
<!-- etc -->
 </listPlace>
</place>
13.3.4.3 States, Traits, and Events

There are many different kinds of information which it might be considered useful to record for a place in addition to its name and location, and the categories selected are likely to be very project-specific. As with persons therefore these Guidelines make no claim to comprehensiveness in this context. Instead, the generic state, trait, and event elements defined by this module should be used. Each of these may be customized for particular needs by means of their type attribute. These are complemented by a small number of predefined elements of general utility:

  • population contains information about the population of a place.
  • climate (climate) contains information about the physical climate of a place.
  • terrain contains information about the physical terrain of a place.

These are all specializations of the generic trait element. This element may be used for almost any kind of event in the life of a place; no specialized version of this element is proposed, nor do we attempt to enumerate the possible values which might be appropriate for the type attribute on any of these generic elements.

Here is an example, showing how the specific and generic elements may be combined:
<place xml:id="IS">
 <placeName xml:lang="en">Iceland</placeName>
 <placeName xml:lang="is">Ísland</placeName>
 <location>
  <geo>65.00 -18.00</geo>
 </location>
 <terrain>
  <desc>Area: 103,000 sq km</desc>
 </terrain>
 <state type="governance" notBefore="1944">
  <p>Constitutional republic</p>
 </state>
 <state type="governance" notAfter="1944">
  <p>Part of the kingdom of <placeName key="DK">Denmark</placeName>
  </p>
 </state>
 <event type="governance" when="1944-06-17">
  <desc>Iceland became independent on 17 June 1944.</desc>
 </event>
 <state type="governance" from="1944-06-17">
  <p>An independent republic since June 1944</p>
 </state>
</place>
In the following example, the climate example is used to provided a detailed discussion of this particular aspect of the information available about a particular place:
<place xml:id="greece">
 <placeName>Greece</placeName>
 <climate>
  <desc>Greece's climate is divided into three well defined
     classes:</desc>
  <climate>
   <label>Mediterranean</label>
   <desc>It features mild, wet winters and hot, dry
       summers. Temperatures rarely reach extremes, although snowfalls do
       occur occasionally even in <placeName>Athens</placeName>,
   <placeName>Cyclades</placeName> or <placeName>Crete</placeName>
       during the winter.</desc>
  </climate>
  <climate>
   <label>Alpine</label>
   <desc>It is found primarily in <placeName>
     <offset>Western</offset>
         Greece</placeName> (<placeName>Epirus</placeName>,
   <placeName>
     <offset>Central</offset> Greece</placeName>,
   <placeName>Thessaly</placeName>,
   <placeName>
     <offset>Western</offset> Macedonia</placeName> as well
       as central parts of <placeName>Peloponnesus</placeName> like
   <placeName>Achaea</placeName>, <placeName>Arcadia</placeName> and
       parts of <placeName>Laconia</placeName> where the Alpine range pass
       by)</desc>
  </climate>
  <climate>
   <label>Temperate</label>
   <desc>It is found in <placeName>
     <offset>Central</offset> and
    <offset>Eastern</offset> Macedonia</placeName> as well as in
   <placeName>Thrace</placeName> at places like
   <placeName>Komotini</placeName>, <placeName>Xanthi</placeName> and
   <placeName>
     <offset>northern</offset> Evros</placeName>. It features
       cold, damp winters and hot, dry summers.</desc>
  </climate>
 </climate>
</place>
As the above example shows, state and trait elements, and others of the same class, can be nested hierarchically within each other. When this is done, values for the type attribute are to be understood as cumulatively inherited, as elsewhere in the TEI scheme (for example on category or linkGrp). In the following example, the outermost population element concerns the squirrel population between the dates given. This is then broken down into red and gray squirrel populations, and within that into male and female:
<population
  type="squirrel"
  notBefore="1901"
  notAfter="1902-01-11"
  resp="#strabo">
 <population type="red" when="1901-01-10">
  <population type="female">
   <desc>12</desc>
  </population>
  <population type="male">
   <desc>15</desc>
  </population>
 </population>
 <population type="gray" when="1902-01-10" cert="high">
  <population type="female">
   <desc>23</desc>
  </population>
  <population type="male" cert="low" resp="#biber">
   <desc>45</desc>
  </population>
 </population>
</population>
The dating and responsibility attributes here behave slightly differently from the type attribute: responsibility is not an additive property, and therefore an element either states it explicitly, or inherits it from its nearest ancestor. Dating is slightly different again, in that a child element may specify a date more precisely than its parent, as in the example above
Events may also be subdivided into other events. For example, a two part meeting might be represented as follows:
<event type="meeting" when="2007-05-29">
 <desc>All day meeting to resolve content models</desc>
 <event type="preamble" notAfter="13:00:00">
  <desc>first part</desc>
 </event>
 <event type="conclusions" notBefore="13:00:00">
  <desc>second part</desc>
 </event>
</event>

An event element is usually used to record information about a place, or a person; for this reason the element usually appears as content of a place or person. However, it is also possible to describe events independently of either a person or a place. This may be useful in such applications as chronologies, lists of significant events such as battles, legislation, etc.

The listEvent element is a member of the model.listLike class, and may therefore appear wherever lists are permitted, in the same way as the listPerson, listPlace etc. elements described elsewhere in this chapter.

<listEvent>
 <event
   when="1713"
   ref="http://eco.canadiana.ca/view/oocihm.9_01832">
  <label>Treaty of Utrecht</label>
  <desc>France ceded to Great Britain its claims to the <orgName>Hudson's Bay
       Company</orgName> territories in <placeName>Rupert's Land</placeName>,
  <placeName>Newfoundland</placeName>, and
  <placeName>Acadia</placeName> and recognized British suzerainty over <orgName type="tribe">the Iroquois</orgName> but retained its other pre-war
     North American possessions, including
  <placeName key="PEI">Île-Saint-Jean</placeName> (now <placeName key="PEI">Prince Edward
       Island</placeName>)...</desc>
 </event>
 <event when="1774" key="14-GeoIII-c83">
  <label>Quebec Act</label>
  <desc>This act of the British Parliament guaranteed free practice of
     the Catholic faith and restored use of the French Civil Code for
     private matters throughout the Province of Quebec, which had been
     expanded in territory following the <ref>Treaty of Paris</ref>.</desc>
 </event>
 <event
   when="1778"
   ref="http://avalon.law.yale.edu/18th_century/del1778.asp">
  <label>Treaty of Fort Pitt</label>
  <desc>Also known as the <name type="event">Treaty with the
       Delawares</name>, this was the first written treaty between the newly
     formed <orgName>United States</orgName> and any Native American people, in this
     case, the <orgName type="tribe">Lenape</orgName> or Delawares.</desc>
 </event>
</listEvent>
13.3.4.4 Relations Between Places
The relation element may also be used to express relationships of various kinds between places, or between places and persons, in much the same way as it is used to express relationships between persons alone. Returning to the Mascarene Islands example cited above, we might define the island group and its constituents separately, but indicate the relationship by means of a relation element:
<listPlace>
 <place xml:id="MASC">
  <placeName>Mascarene islands</placeName>
  <placeName>Mascarenhas Archipelago</placeName>
 </place>
 <place xml:id="MRU">
  <placeName>Mauritius</placeName>
<!-- ... -->
 </place>
 <place xml:id="ROD">
  <placeName>Rodrigues</placeName>
 </place>
 <place xml:id="REN">
  <placeName>Réunion</placeName>
 </place>
 <relation name="contains" active="#MASC" passive="#ROD #MRU #REN"/>
</listPlace>
This ‘stand-off’ style of representation has the advantage that we can now also represent the fact that a place may be a ‘part of’ more than one other place; for example, Réunion is part of France, as well as part of the Mascarenes. If we add a declaration for France to the list above:
<place type="country" xml:id="FRA">
 <placeName>France</placeName>
</place>
we can now model this dual allegiance by means of a relation element:
<relation name="partOf" active="#REN" passive="#FRA #MASC"/>

13.3.5 Names and Nyms

So far we have discussed ways in which a name or referring string encountered in running text may be resolved by considering the object that the name refers to: in the case of a personal name, the name refers to a person; in the case of a place name, to a place, for example. The resolution of this reference is effected by means of the key or ref attributes available to all elements which are members of the att.naming class, such as persName or placeName and their more specialized variants such as forename or country. However, names can also be regarded as objects in their own right, irrespective of the objects to which they are attached, notably in onomastic studies. From this point of view, the names John in English, Jean in French, and Ivan in Russian might all be regarded as existing independently of any person to which they are attached, and also independently of any variant forms that might be attested in different sources (such as Jon or Johnny in English, or Jehan or Jojo in French). We use the term nym to refer to the canonical or normalized form of a name regarded in such a way, and provide the following elements to encode it:

  • listNym (list of canonical names) contains a list of nyms, that is, standardized names for any thing.
  • nym (canonical name) contains the definition for a canonical name or name component of any kind.

Any element which is a member of the att.naming class may use the attribute nymRef to indicate the nym with which it corresponds. Thus, given the following nym for the name Antony:
<listNym>
 <nym xml:id="N123">
  <form>Antony</form>
 </nym>
<!-- other nym definitions here -->
</listNym>
an occurrence of this name in running text might be encoded as follows:
<forename nymRef="#N123">Tony</forename> Blair
Note that this association (between "Tony" and "Antony") has nothing to do with any individual who might use the name.
The person identified by this particular Tony may however be indicated independently using the ref attribute, either on the forename or on the whole name component:
<forename nymRef="#N123" ref="#BLT">Tony</forename>
....

<person xml:id="BLT">
 <persName>Tony Blair</persName>
 <occupation>politician</occupation>
</person>
The nym element may be thought of as providing a specialized kind of dictionary entry. Like a dictionary entry, it may contain any element from the model.entryPart class, such as form, etym, etc. For example, we may show that the canonical form for a given nym has two orthographic variants in this way:
<nym xml:id="J451">
 <form>
  <orth xml:lang="en-US">Ian</orth>
  <orth xml:lang="en-x-Scots">Iain</orth>
 </form>
</nym>
Because a schema intending to make use of the nym or listNym element must include the dictionaries module as well as the namesdates module, many other elements are available in addition to form. For example, to provide a more complex etymological decomposition of a name, we might use the existing etym element, as follows:
<nym xml:id="XYZ">
 <form>Bogomil</form>
 <etym>Means <gloss>favoured by God</gloss> from the
 <lang>Slavic</lang> elements <mentioned xml:lang="ru">bog</mentioned>
  <gloss>God</gloss> and <mentioned xml:lang="ru">mil</mentioned>
  <gloss>favour</gloss>
 </etym>
</nym>
Where it is necessary to mark the substructure of nyms, this may be done by seg elements within the form:
<nym xml:id="ABC">
 <form>
  <choice>
   <seg type="morph">
    <seg>Bog</seg>
    <seg>o</seg>
    <seg>mil</seg>
   </seg>
   <seg type="morph">
    <seg>Bogo</seg>
    <seg>mil</seg>
   </seg>
  </choice>
 </form>
</nym>
The seg element used here is provided by the TEI linking module, which would therefore also need to be included in a schema built to validate such markup. Other possibilities for more detailed linguistic analysis are provided by elements included in that and the analysis (see 17 Simple Analytic Mechanisms) or iso-fs modules (see 18 Feature Structures).
Alternatively, each of the constituents of Bogomil might be regarded as a nym in its own right:
<nym xml:id="B1" type="part">
 <form>bog</form>
</nym>
<nym xml:id="M1" type="part">
 <form>mil</form>
</nym>
Within running text, a name can specify all the nyms associated with it:
...<name nymRef="#B1 #M1">Bogomul</name>...
Similarly, within a nym, the attribute parts is used to indicate its constituent parts, where these have been identified as distinct nyms:
<nym xml:id="BM1" parts="#B1 #M1">
 <form>Bogomil</form>
</nym>
The nym element may also combine a number of other nym elements together, where it is intended to show that they are all regarded as variations on the same root. Thus the different forms of the name John, all being derived from the same root, may be represented as a hierarchic structure like this:
<nym xml:id="J45">
 <form xml:lang="la">Iohannes</form>
 <nym xml:id="J450">
  <form xml:lang="en">John</form>
  <nym xml:id="J4501">
   <form>Johnny</form>
  </nym>
  <nym xml:id="J4502">
   <form>Jon</form>
  </nym>
 </nym>
 <nym xml:id="J455">
  <form xml:lang="ru">Ivan</form>
 </nym>
 <nym xml:id="J453">
  <form xml:lang="fr">Jean</form>
 </nym>
</nym>
The nym element may be used for components of geographical or organizational names as well. For example:
<geogName ref="tag:projectname.org,2012:LAEI1" type="hill">
 <geogFeat xml:lang="gd" nymRef="#LAIRG">Lairig</geogFeat>
 <name>Eilde</name>
</geogName>
...

<nym xml:id="LAIRG">
 <form xml:lang="gd">lairig</form>
 <def>sloping hill face</def>
</nym>
...

As noted above, use of these elements implies that both the dictionaries and the namesdates modules are included in a schema.

13.3.6 Dates and Times

The following elements for the encoding of dates and times were introduced in section 3.5.4 Dates and Times:

  • date contains a date in any format.
  • time contains a phrase defining a time of day in any format.

The current module namesdates provides a mechanism for more detailed encoding of relative dates and times. A relative temporal expression describes a date or time with reference to some other (absolute) temporal expression, and thus may contain an offset element in addition to one or more date or time elements:

  • offset marks that part of a relative temporal or spatial expression which indicates the direction of the offset between the two place names, dates, or times involved in the expression.

As members of the att.datable and att.duration classes, which in turn are members of att.datable.w3c and att.duration.w3c respectively, the date and time elements share the following attributes:

  • att.datable.w3c provides attributes for normalization of elements that contain datable events using the W3C datatypes.
    whensupplies the value of the date or time in a standard form, e.g. yyyy-mm-dd.
  • att.duration.w3c provides attributes for recording normalized temporal durations.
    dur(duration) indicates the length of this element in time.

13.3.6.1 Relative Dates and Times

As noted above, relative dates and times such as ‘in the Two Hundredth and First Year of the Republic’, ‘twenty minutes before noon’, and, more ambiguously, ‘after the lamented death of the Doctor’ or ‘an hour after the game’ have two distinct components. As well as the absolute temporal expression or event to which reference is made (e.g. ‘noon’, ‘the game’, ‘the death of the Doctor’, ‘[the foundation of] the Republic’), they also contain a description of the ‘distance’ between the time or date which is indicated and the referent expression (e.g. ‘the Two Hundredth and First Year’, ‘twenty minutes’, ‘an hour’); and (optionally) an ‘offset’ describing the direction of the distance between the time or date indicated and the referent expression (e.g. ‘of’ implying after, ‘before’, ‘after’).

The ‘distance’ component of a relative temporal expression may be encoded as a temporal element in its own right using either date or time, or with the more generic measure element. A special element, offset, is provided by this module for encoding the ‘offset’ component of a relative temporal expression. The absolute temporal expression contained within the relative expression may be encoded with a date or time element; in turn, those elements may of course be relative, and thus contain date or time elements within themselves. This allows for deeply nested structures such as ‘the third Sunday after the first Monday before Lammastide in the fifth year of the King's second marriage ... ’ but so does natural language.

In the following examples, the when and dur attributes have been used to simplify processing of variant forms of expression:
<date when="1786-12-11">
 <date dur="P14D">A fortnight</date>
 <offset>before</offset>
 <date when="1786-12-25" type="holiday">Christmas 1786</date>
</date>
I reached the station <time when="14:15:00">
 <time dur="PT30M0S">precisely half an hour</time>
 <offset>after</offset>
 <time when="13:45:00" type="occasion">the departure of the afternoon train to Boston</time>
</time>
In the following example, a nested date element is used to show that ‘my birthday’ and the cited date are parts of the same temporal expression, and hence to disambiguate the phrase ‘A week before my birthday on 9th December’:
<date when="--12-02">
 <date>A week</date>
 <offset>before</offset>
 <date when="--12-09">
  <date type="occasion">my birthday</date>
   on <date>9th December</date>
 </date>
</date>
The alternative reading of this phrase could be encoded as follows:
<date when="--12-09">
 <date>A week</date>
 <offset>before</offset>
 <date type="occasion" when="--12-16">my birthday</date>
on <date>9th December</date>
</date>
Where more complex or ambiguous expressions are involved, and where it is desirable to make more explicit the interpretive processes required, the feature structure notation described in chapter 18 Feature Structures may be used. Consider, for example, the following temporal expression which occurs in the Scottish Temperance Review of August 1850, referring to the summer holiday known in Glasgow simply as ‘the Fair’:
Not only is the city, <date ana="#gf50">during the Fair</date>, a horrible nucleus of immorality
and wickedness; it sends our multitudes to pollute and demoralize the
country.
For the definition of the ana attribute, see chapter 17 Simple Analytic Mechanisms (in particular 17.2 Global Attributes for Simple Analyses). It is used here to link the temporal phrase with an interpretation of it. Like most traditional fairs and market days, the Glasgow Fair was established by local custom and could vary from year to year. Consequently, in order to provide such an interpretation, it is necessary to draw upon additional information which may or may not be located in the particular text in question. In this case, it is necessary at least to know the spatial and temporal context (year and place) of the fair referred to. These and other features required for the analysis of this particular temporal expression may be combined together as one feature structure of type date-analysis:
<fs xml:id="gf50" type="date-analysis">
 <f name="event">
  <string>the Fair</string>
 </f>
 <f name="place">
  <string>Glasgow</string>
 </f>
 <f name="year">
  <numeric value="1850"/>
 </f>
 <f name="from-value">
  <string>1850-08-08</string>
 </f>
 <f name="to-value">
  <string>1850-09-19</string>
 </f>
</fs>
For further discussion of feature structure representation see chapter 18 Feature Structures.
13.3.6.2 Absolute Dates and Times

The following are examples of absolute temporal expressions.

The university's view
of American affairs produced a stinging attack by Edmund Burke in the
Commons debate of <date when="1775-10-26">26 October 1775</date>
<date when="1993-05-14">Friday, 14 May 1993</date>
It may be useful to categorize a temporal expression which is given in terms of a named event, such as a public holiday, or a named time such as ‘tea time’ or ‘matins’:
In New York, <date type="occasion" when="--01-01">New Year's Day</date> is the quietest of
holidays, <date when="--07-04" type="occasion">Independence Day</date>
the most turbulent.
Absolute temporal expressions denoting times which are given in terms of seconds, minutes, hours, or of well-defined events (e.g. ‘noon’, ‘sunset’) may similarly be represented using the time element.
The train leaves for Boston at
<time type="twentyfourHour" when="13:45:00">13:45</time>
At <time type="occasion">sunset</time> we walked to the beach.
The train leaves for Boston at
<time xml:lang="en-US" type="descriptive" when="13:45:00-05:00"> a quarter of two
</time>
13.3.6.3 More Expressive Normalizations

The attributes for normalization of dates and times so far described use a standard format defined by XML Schema Part 2: Datatypes Second Edition. This format is widely accepted and has significant software support. It is essentially a profile of ISO 8601 Data elements and interchange formats — Information interchange — Representation of dates and times. The full ISO standard provides formats not available in the W3C recommendation, for example, the capability to refer to a date by its ordinal date or week date, or to refer to a century. It also provides ways of indicating duration and range.

When this module is included in a schema, the following additional attributes are provided:
  • att.datable.iso provides attributes for normalization of elements that contain datable events using the ISO 8601 standard.
    when-isosupplies the value of a date or time in a standard form.
    notBefore-isospecifies the earliest possible date for the event in standard form, e.g. yyyy-mm-dd.
    notAfter-isospecifies the latest possible date for the event in standard form, e.g. yyyy-mm-dd.
    from-isoindicates the starting point of the period in standard form.
    to-isoindicates the ending point of the period in standard form.
  • att.duration.iso provides attributes for recording normalized temporal durations.
    dur-iso(duration) indicates the length of this element in time.
These attributes may be used in preference to their W3C equivalent when it is necessary to provide a normalized value in some form not supported by the W3C attributes. For example, a century date in the W3C format must be expressed as a range, using the from together with either the to attribute or the dur (duration) attribute:
<date from="1301" to="1400">fourteenth century</date>
<date from="1301" dur="P1Y">fourteenth century</date>
With the attribute when-iso, however, it is possible to express the same normalized value in any of the following ways:
<date when-iso="13">fourteenth century</date>
<date when-iso="1301/1400">fourteenth century</date>
<date when-iso="1301/P1Y">fourteenth century</date>

13.4 Module for Names and Dates

The module described in this chapter makes available the following components:

Module namesdates: Names and dates

The selection and combination of modules to form a TEI schema is described in 1.2 Defining a TEI Schema.

Notes
42
In the module described by chapter 22 Documentation Elements a similar method is used to link element descriptions to the modules or classes to which they belong, for example.
↵
43
Strictly, a suitable value such as figurative should be added to the two place names which are presented periphrastically in the second version of this example. This would preserve the distinction indicated by the choice of rs rather than name to encode them in the first version of this example.
↵
44
See http://earth-info.nga.mil/GandG/wgs84/index.html. The most recent revision of this standard is known as the Earth Gravity Model 1996.
↵
45
The OGC is an international voluntary consensus standards organization whose members maintain the Geography Markup Language standard. The OGC coordinates with the ISO TC 211 standards organization to maintain consistency between OGC and ISO standards work. GML is also an ISO standard (ISO 19136:2007).
↵

[English] [Deutsch] [Español] [Italiano] [Français] [日本語] [한국어] [中文]

TEI Consortium | Feedback

TEI material can be licensed differently depending on the use you intend to make of it. Hence it is made available under both the CC+BY and BSD-2 licences. The CC+BY licence is generally appropriate for usages which treat TEI content as data or documentation. The BSD-2 licence is generally appropriate for usage of TEI content in a software environment. For further information or clarification, please contact the TEI Consortium.

  1. http://creativecommons.org/licenses/by/3.0/
  2. http://www.opensource.org/licenses/BSD-2-Clause

TEI Guidelines Version 2.3.0. Last updated on 17th January 2013. This page generated on 2013-01-17T21:36:55Z