ATLAS LINGÜÍSTICO (Y ETNOGRÁFICO) DE CASTILLA - LA MANCHA
página principal
AUTOMATIC CARTHOGRAPHY AND GEOLINGUISTIC DATA BASES
Francisco Moreno-Fernández
José Enrique Moreno-Fernández
Antonio García de las Heras
University of Alcalá, Spain
Introduction
Since 1996, the University of Alcalá has been making a remarkable effort
to obtain a computer application capable of mapmaking, in an automatic way,
materials previously registered in one or more databases. This effort has allowed
us to incorporate our work into a systematic form of research which has been
very much developed in other research centers, both in Europe and in America:
we can refer to the computer systems developed for the Atlante linguistico del
ladino dolomitico/centrale e dialetti limitrofi, of the University of Salzburg
(Austria), in the computerization of the Atlante linguistico de la Sicilia,
accomplished in the “Centro di Studi Filologici e Linguistici Siciliani”
(Palermo, Italia), or in the work of the Linguistic Atlas of the Middle and
South Atlantic States, of the University of Georgia (USA).
In the Spanish speaking countries, the first work which has been done within
the automation field has been the Atlas Lingüístico y Etnográfico
de Cantabria, directed by Manuel Alvar, though there is also news of the information
processing in some of the materials of the Atlas Lingüístico de
Hispanoamérica. The application developed in the University of Alcalá
is called Atlante, and it has been designed for editing any linguistic atlas
just with a basic adaptations.The purpose of this paper is to make a brief description
of Atlante’s usefulness and characteristics.
Objectives of Atlante
The general objective of the Atlante program is the automation of cabinets work
that lead to the attainment of a linguistic atlas, as well as, maximizing the
use of all the linguistic information which is contained in a piece of work
with these characteristics. To fulfill such an objective, it is necessary a
database in which all the materials remain registered, a database which permits
a rapid search of whatever information the researcher requires and the elaboration
of several kinds of indexes. Furthermore, a mechanism is needed to permit us
to transfer to a map the linguistic data contained in the base, either in the
form of phonetic transcription, or in ordinary writing or through symbols.
The purpose of the Atlante application is not only to get to a computerized
draft of a linguistic atlas that is can be represented on paper, but to offer
the possibility of working and manipulating the linguistic information directly
by computerized means. In other words, a mean has been sought to automate the
conventional elaboration of linguistic atlases and permits us to present an
atlas in printed paper or in computer supports, as the CD-ROM or on-line consultation.
When deciding the type of atlas we would wish to achieve, we have used as a
reference one of the richest and complex models of those we know: the model
of the Spanish linguistic atlases. These atlases, as it is known, include linguistic
maps, ethnographic maps and linguistic-ethnographic maps, and the linguistic
information is offered, some times through symbols - as it is usual in other
computerized works - but most times, in phonetic transcription. The answers
have a very variable extension at each point with the data distributed within
the maps, outside the maps (on the margins), and in complementary plates. At
the same time, different types of letters and characters of various nature alternate.
The phonetic transcription is also presented following the alphabet of the Revista
de Filología Española.
The characteristics of the work which has served us as a model, demanded the
preparation of an application extremely flexible, capable of registering materials
in phonetic transcription and in ordinary writing. This application must alternate
alphabets in the same text, editing automatically very dense maps, but giving
the researcher the opportunity of adjusting the definitive presentation of the
maps according to his/her needs and wants. All of this is complemented with
the development of the mentioned phonetic alphabet, but enriched with a long
list of phonetic characters, indispensable for the transcription of numerous
features of Spanish speeches.
The databases should be capable of allowing very diverse uses and manipulations,
from the simple consultation of any chain of characters, to the indexes elaboration
- including indexes of inverse forms- the incorporation of sound files, the
incorporation of graphs, and the calculation of linguistic distances between
survey points (dialectometry).
Main characteristics of Atlante
Atlante is a 32 bits operating system application designed for Windows 95. It
has been developed using Visual Basic version 4.0. Creation and management of
the Atlante’s databases have been made with the “Microsoft Access
Jet Engine”. For running Atlante in a personal computer, we recommend
at least the following hardware specifications: 32 meg. RAM and Pentium processor
100 MH or faster and hard drive enough to store the databases and the bmp 256
color files generated by Atlante.
The application uses three specific databases and their graphical projection on the map. These databases are the following:
Survey points or Community Database
This option is available from the main menu, once the application has been opened.
Here it is where you store the name, the id code, the speakers’ name and
the coordinates of each point on the map.
Speakers Database
This option is available from the main menu. In this base we will store the
data concerning the speakers (profession, age, the parents’ born place,
etc.)
Linguistic Information Database (RESPUESTAS)
This option is also available from the main menu. The linguistic data stored
here appear in ordinary writing as well as in phonetic transcription. In this
transcription it is possible to use any font installed in “Windows”.
The University of Alcalá has developed the font afe.ttf for representing
the phonetics symbols established in the Revista de Filología Española.
In any case you may use any available “International Phonetic Alphabet”
font.
From the databases detailed, the most complex, from a technical point of view,
and the richest, from a linguistic perspective, is “RESPUESTAS”
(answers). Technically it is not a single database, but a set of bases, which
will be more or less extensive depending on the contents you want to include:
a general linguistic atlas can not be the same as a phonetic atlas, and it is
not the same a 300 map linguistic atlas as one with 3000 maps. When we talk
about “RESPUESTAS” database (answers) we just pretend to make the
program’s structure a bit easier to understand.
The databases included in the “RESPUESTAS” database could be so
many as linguistic levels in the atlas. For an atlas with just phonetic-phonological
and lexical levels, only two databases should be necessary. On the other hand,
every single database could be composed of tables that we call “Fields”.
For example, the “Phonetic Database” could be divided into the “Vowel
Phonetics” and “Consonant Phonetics” fields, and the lexical
database could be divided into fields like “Human body”, “Farming
Life”, “Atmospheric Phenomenon”, etc. Inside each field, the
answers relating to certain objects, actions or concepts have been inscribed.
The larger the number of items that make up each base, the more necessary the
subdivision in tables or fields is, to insure a very fast and simple handling
of the application.
The introduction of data into the “RESPUESTAS”
database will be made using the form in a window: some data must be stored in
writing and the rest in phonetic transcription. To introduce data in each field
you must choose among the different options in the form (point, speaker, field).
Every other different possible selections are automatically showed. This selection
is possible because inside the form we are provided with an update list of every
survey points and speakers from our databases.
The linguistic information introduced in this form is stored in ordinary writing
as well as phonetic transcription. The main reason to do it this way is to make
searches for strings of characters easier. According to verb conjugation, it
is also possible to have a verb form list for consulting and updating.
To introduce linguistic information, that is to say, the speakers answers in
the questionnaire, we have two different helps: on the one hand, the form provides
a list of answers stored previously; on the other hand, we have graphic help
about the phonetic alphabet by pressing F1 and F2 keys.
To sum up, the forms and windows from where it is possible to update and consult
the answers in the geolinguistic questionnaire provides updated information
about survey points, speakers, the linguistic level and the field in which the
answers are stored, in written form and in phonetic transcription.