Reformationes consiliorum civitatis Ragusii: encoding guidelines
Following the example of Women Writers Project Encoding Guide for Early Printed Books, and realizing the importance of ample documentation for research based on the TEI XML standard for the representation of texts in digital form, we list here the details of encoding, the solutions of problems, and the specific elements used for a digital edition of the Libri reformationum civitatis Ragusii.1)
First we present the base text of our digital edition, and then we describe the TEI XML elements and attributes selected for the edition, as well as its conventions.
The digital edition is being designed with later XQuery explorations in mind, so the value of selected solutions will be demonstrated by different XQuery searches made possible by our encoding scheme.
Further search examples can be found on the related BaseX Adventures page.
The base text
In the first, exploratory phase, we use as base text the modern edition of the Reformationes 1390-1392, published in 2005 as Volume 6 of the Monumenta historica Ragusina series: Nella Lonza and Zdravko Šundrica, Odluke dubrovačkih vijeća od 1390. do 1392. Hrvatska akademija znanosti i umjetnosti, Zavod za povijesne znanosti u Dubrovniku, Zagreb-Dubrovnik 2005 (WorldCat record here).
The TEI XML source, documentation, and XQueries are available on Bitbucket.
Text structure
The manuscript Libri reformationum of the Dubrovnik councils, written mostly in Latin by the city chancellors, record three types of official deliberations: those by the Consilium minus (the Minor Council), by the Consilium rogatorum, and by the Consilium maius (the Major Council), the Reformationes include also lists of office-holders on duty.
The councils met on various days of month throughout the year. Furthermore, the original manuscript codices group records of each council in a separate part of a volume.2)
The basic structure of our edition thus consists of four divisions:
div ana=“annus”: this division holds all records of a council from a single year (Latin annus)div ana=“mensis”: contains all records of a council from a single month (Latin mensis)div ana=“dies”: all records made during a single day (Latin dies), or a single meeting (the councils sometimes made pauses, mostly lunch breaks, it seems, with another session in the afternoon)
Each div has a special identifier (@n attribute), in the form councilYYYY(-MM-DD): maius1392, maius1392-12, or maius1392-12-03; five-letter codes for councils are maius, minus, rogat. In case of multiple records for the same council on the same day, we add letters a, b, c… at the identifier's end (maius1392-12-03b).
Examples
1. Year 1390, Consilium minus:
<div ana="annus" n="minor1390">
2. October 1391, Consilium maius:
<div ana="mensis" n="maior1391-10">
3. December 5, 1392, Consilium rogatorum:
<div ana="dies" n="rogat1392-12-05">
XQuery exercises and examples
- Find all records of the Consilium minus sessions, order them chronologically.
- Find a specific session of the Consilium rogatorum.
- How many meetings of the Consilium maius were held in January 1391?
Section titles
The divs for years, months, and days mostly have titles, usually bearing the date of the session, sometimes also the number of councillors present etc. The titles are encoded with head element, and dates with date. We have introduced such elements even where titles are absent, using them to hold empty date and ref/@ana=“rector” elements, referring respectively to dates and rectors on duty.
A fully written out title is encoded like this:
<div ana="dies" xml:id="maius1392-11-28"> <head><date when="1392-11-28">Die XXVIII novembris 1392.</date> In maiori consilio ballote <num>LXXXV</num>.</head>
This is how implied information is made explicit in a title (and implicit in encoding):
<div ana="mensis" xml:id="minus1390-08"> <head><date when="1390-08-01"/></head>
In titles of months we are adding the (implied) rector's ID as well:
<div ana="mensis" n="maius1390-01"> <head> <date when="1390-01"/> <ref ana="rector" target="BONAMAR1390-02"/> </head>
XQuery exercises and examples
- Return all headings with text, ordered alphabetically.
- Return all meetings which record the number of voters present. Order the meetings by the number of voters. Alternatively, order them chronologically.
- Return all heads of meetings ordered by rector's ID.
- Return all meetings with the ID of a specific rector.
Item titles (themes)
During a single session, one or more items were discussed; for the sake of orientation, chancellors presented items as titles, sometimes as marginal notes.
The titles in text are marked as p elements with @ana attributes whose value is thema:
<p ana="thema">De doanis vendendis</p>
However, sometimes the themes were noted in margins; this is marked as a note inside the p element:
<p ana="thema"> <note type="margin" resp="notar"> Pro <persName ref="BUCHIGNMM1390-02">ser Marino de Bucignolo</persName> pro partibus Stagni </note> </p>
XQuery exercises and examples
- Find only the themes written in margins, order them alphabetically.
- Display only the paragraphs with themes, but show the dates of sessions as well.
Candidate lists
Lists of candidates for offices are wrapped in a list element, with a special head/@ana=“thema” heading. In case of elections for the next month's rector, the item containing the future rector's name has a special @ana=“rector-futurus” attribute.
Voting for rector is of potential interest, because sometimes such elections had to go through several rounds of casting votes. Identifying people involved in such situations could offer a starting point for studying power relations in 1390's Dubrovnik.
<list> <head ana="thema">Rector mensis februarii proxime venturi:</head> <item><persName>Ser Give de Volczo</persName> ballote <num value="0">0</num>.</item> <item ana="rector-futurus"><persName ref="BABALMICH1390-01">Ser Michael de Baballio</persName> ballote <num value="52">LII</num>.</item> <item><persName ref="BODACMAR1390-05">Ser Marinus de Bodaça</persName> ballote <num value="32">XXXII</num>.</item> </list>
Names
Using the Index nominum, we have created a small prosopography file where every individual has a unique @xml:id attribute. The value of this attribute is then used as value of the ref attribute (appearing in persName elements), or of the target attribute in the ref element (see above the case of implied rectors), in the main Reformationes XML file.
For example:
<person xml:id="GONDULBENED1390-02"> <persName>Gondula (Gondulla) Benedictus 68, 179</persName> </person>
(References to pages of the print editions have been kept, just in case.)
In the main file:
<persName ref="BENESSMS1390-03">Ser Matheus Syme de Benessa</persName> et <persName ref="GONDULBENED1390-02">ser Benedictus de Gondulla</persName> camerarii comunis Ragusii fuerunt confessi recepisse a <persName ref="GOCEALU1390-01">ser Aluisio de Goçe</persName> pro parte solucionis dicti salis quod ipse cum fratribus receperunt in credenciam <measure unit="perper" quantity="730">perperos septemcentum et triginta</measure>.
Names of places have not been encoded yet!
(Should probably use URNs…)
XQuery exercises and examples
- Produce an enhanced Index nominum, enabling us to follow individuals over time and sessions, rather than over pages of the printed edition (as is the case with the existing index).
- Find constellations – people mentioned together in the same sessions.
- Find all rectors in the text.
- Study candidates for offices.
Measures and values
Votes
Number of votes cast (the voting outcome) is marked by a measure element with the attribute @type=“vota” and @unit=“balota” (votes are called “balotae” in the original Latin text). The measure element envelops the whole sentence (or clause) which records the voting outcome.
Inside the measure element, the number is marked by num element with the value attribute giving the number in Arabic numerals.
Example from the main file:
<measure type="vota" unit="balota"> captum fuit per <num value="8">VIII</num> ballotas </measure>
Money
Money values recorded are marked as measure elements (ideally, with @type=“summa” attribute) with @unit attribute, which at the moment contains five currency types: ducat, perper, soldus, folar, grossus. If the currency is abbreviated, the abbreviation is marked by the abbr element. The number is (ideally) marked by num with @value attribute (and values in Arabic numerals).
Examples from the main file:
<measure unit="perper">yperperorum <num value="10">decem</num> </measure> <measure unit="grossus" quantity="1.5"> uno grosso cum dimidio </measure> <measure unit="grossus"> grossorum <num value="12">XII</num> </measure> <measure unit="ducat" quantity="30"> <abbr>duc.</abbr> <num value="30">XXX</num> </measure>
Weights
At the moment, only weights (in libres) are encoded, using the measure element with @type=“libra” attribute, and the amount recorded given as value of the @quantity attribute. When the record mentions one libre, it does not usually use numerals; we express them as @quantity=“1” attribute of the measure element.
Numbers are marked as above.
Example from the main file:
<measure unit="libra" quantity="500"> usque ad libras <num value="500">VC</num> </measure> <measure unit="libra" quantity="1">libra</measure>
