Criteria for the Construction of a Corpus for a Mexican Spanish Dictionary of SexuaHty

Abstract

This paper describes • • criteria to buiId an electronic Corpus for a Mexican Spanish Dictionary of Sexuality. The Corpus will be used, among other things, for the extraction of the field's terminology. Needless to say, automatic extraction of terms presupposes a representative Corpus. Thus, we sketch our strategies to assure such representativity. ni short, we address the need for a balanced Corpus; that is, one containing texts belonging to all thematic areas ofthe field. Then we describe the sort oftexts that must be included. Also, we discuss our approach to deal with the important sociolinguistic dimension. Finally, we examine the need of an articulated set of concepts; that is, the conceptual system used by sexologists, sexual educators and other specialists (physicians, psychologists, social workers, etc.). • this way, we expect our Dictionary ofSexuality to have a set ofentries which will guarantee the broadest possible coverage ofthis specialty field's knowledge.

Full text

REPORTS
ON
LEXICOGRAPHICAL
AND
LEXICOLOGICAL
PROJECTS
Criteria
for
the
Construction
of
a
Corpus
for
a
Mexican
Spanish
Dictionary
of
SexuaHty
Alfonso
Medina,
Gerardo
Sierra
bistituto
de
higenieria
Universidad
Nacional
Autónoma
de
México
Apartado
Postal
70-472
04510
DelegacionCoyoacan,DF
MEXICO
{amedinau,
gsierram}@iingen.unam.mx
Abstract
This
paper
describes
•
•
criteria
to
buiId
an
electronic
Corpus
for
a
Mexican
Spanish
Dictionary
of
Sexuality.
The
Corpus
will
be
used,
among
other
things,
for
the
extraction
of
the
field's
terminology.
Needless
to
say,
automatic
extraction
of
terms
presupposes
a
representative
Corpus.
Thus,
we
sketch
our
strategies
to
assure
such
representativity.
ni
short,
we
address
the
need
for
a
balanced
Corpus;
that
is,
one
containing
texts
belonging
to
all
thematic
areas
ofthe
field.
Then
we
describe
the
sort
oftexts
that
must
be
included.
Also,
we
discuss
our
approach
to
deal
with
the
important
sociolinguistic
dimension.
Finally,
we
examine
the
need
of
an
articulated
set
of
concepts;
that
is,
the
conceptual
system
used
by
sexologists,
sexual
educators
and
other
specialists
(physicians,
psychologists,
social
workers,
etc.).
•
this
way,
we
expect
our
Dictionary
ofSexuality
to
have
a
set
ofentries
which
will
guarantee
the
broadest
possible
coverage
ofthis
specialty
field's
knowledge.
1.
Introduction
Researchers
and
educators
rely
upon
knowledge
representation
tools
that
provide
consensus,
integration
and
clarity
of
concepts
and
definitions
within
their
specialty
fields.
One
very
old
type
of
this
kind
of
tool
is
the
dictionary.
And
one
field
of
distinct
complexity,
especially
in
countries
like
Mexico,
is
that
of
Sexuality
Studies
and
Sexual
Education.
Perhaps
the
main
problem
for
a
dictionary
for
this
field
is
the
dispersion
of
terminology.
Although
some
scattered
glossaries
and
vocabularies
do
exist,
there
is
no
single
reference
work
in
Spanish
which
provides
the
terminological
consensus
needed
for
Sexual
Education;
in
fact,
not
even
for
communication
among
Sexology
specialists.
Dictionaries
in
book
form
do
exist,
but
they
are
not
meant
either
for
Sexual
Education
or
for
sexological
research
consensus.
Some
are
marketing
products
intended
to
promote
the
image
of
media
celebrities.
1
Others
are
encyclopedic
works
portraying
distant
realities
and
therefore
failing
to
fulfill
the
needs
of
a
Mexican
audience.
2
A
dictionary
tailored
to
the
needs
of
such
an
audience
presupposes,
on
the
one
hand,
some
articulated
set
of
concepts
and
definitions
representing
the
up
to
date
sexological
knowledge
related
to
important
educational
issues
like
Health
and
Human
Rights:
the
conceptual
system
deemed
by
experts
to
be
most
relevant
in
the
field
of
Sexology.
On
the
other
hand,
the
set
of
terms
used
by
regular
Mexican
people
to
refer
to
their
sexual
experiences
should
also
be
considered.
Whether
linguistic
communities
are
informed
or
not
about
sexological
facts,
the
reality
is
that
most
people
do
use
their
language
to
taUc,
secretly
391

EURALEX2004
PROCEEDWGS
or
not,
about
sexuality,
mdeed,
the
sociolinguistic
dimension
is
a
very
complex
issue
because
regional
and
social
variation
is
inevitable.
Nevertheless,
its
impact
on
Sexuality
Studies
and
Sexual
Education
is
important,
bideed,
new
word
documentation
would
help
educators
monitor
student
attitudes
toward
sexual
issues
and
favor
overall
communication
among
specialists.
The
present
paper
describes
our
attempt
to
deal
with
the
lack
of
such
dictionary.
Namely,
we
describe
our
criteria
to
build
an
electronic
Corpus
for
a
Mexican
Spanish
Dictionary
of
Sexuality.
Once
built,
the
Corpus
will
be
used
for
terminology
extraction.
However,
identifying
the
terminology
of
a
field
is
one
very
specific
task
that
fits
within
a
larger
program
of
compiUng,
publishing
and
distributing
a
dictionary.
For
the
purpose
of
education
the
interest
in
Mexico
for
carrying
this
out
is
considerable.
Mexico's
Sexual
Education
community
is
willing
to
offer
their
expertise
and
validate
the
resulting
material.
And
other
prestigious
research
and
educational
institutions
are
committed
to
the
advancement
and
accomplishment
of
the
larger
project.
Thus,
El
Colegio
de
Mexico,
a
research
institute
and
graduate
school
which
has
pioneered
lexicographical
and
terminographical
investigations
in
Mexico
(Lara
et
al.,
1979;
Lara,
1990;
Pozzi,
1996),
has
an
experienced
team
of
terminologists
committed
to
the
appropriate
formulation
of
the
definitions
of
the
extracted
terms.
Also,
the
mstitute
of
Engineering
of
our
National
Autonomous
University
is
developing
an
electronic
dictionary
3
capable
of
performing
complex
multimedia
functions
and
suitable
for
massive
distribution
(Sierra
et
al.,
2002).
2.
Criteria
for
Building
a
Corpus
Ideally,
terminologists
rely
on
the
opinion
of
experts
to
determine
terms
and
to
define
them.
However,
expert
participation
in
a
terminological
project
can
be
very
expensive.
Hence,
electronic
corpora
(seen
as
written
representations
of
expert
opinion)
are
important
alternatives
to
intensive
participation
by
the
field's
specialists.
Nevertheless,
expert
opinion
remains
for
this
project
an
important
secondary
source
of
information.
Terminology
extraction
from
corpora
can
be
accomplished
automatically
by
means
of
statistical
methods.
For
example,
one
method
of
identifying
a
specialized
field's
terminology
is
to
compare
the
distribution
of
words
from
a
general
purpose
corpus
with
that
of
a
specialized
corpus.
Selection
of
those
word
types
with
higher
frequencies
in
the
latter
yields
a
set
ofpotential
terms
for
the
field.
However,
research
proper
begins
with
the
selection
of
the
sample
texts
that
should
belong
to
the
Corpus.
First,
the
relevant
thematic
areas
of
the
field
in
question
must
be
identified.
Given
the
different
perspectives
from
which
Sexuality
can
be
viewed,
there
is
a
lack
of
consensus
among
specialists.
Some
are
medical
specialists,
some
are
psychologists,
some
are
social
workers,
some
are
Human
Rights
militants,
etc.
The
thematic
structure
of
the
field
varies
from
one
perspective
to
the
other.
We
solved
this
problem
by
referring
to
an
international
institution
considered
a
scientific
authority
by
all.
hi
this
manner,
we
obtained
from
the
Kinsey
fnstitute
the
thematic
areas
oftheir
Library
classification
system.
Second,
the
key
texts
ofthe
field
must
be
included.
Although
widely
read
works
are
originally
written
in
some
language
other
than
Spanish
(mostly
English)
or
in
some
other
dialect
(mostly
Argentinean
and
Peninsular
Spanish),
many
have
greatly
influenced
the
training
of
researchers
and
educators
in
Mexico.
Thus,
it
is
not
surprising
that
many
of
the
392

REPORTS
ONLEXICOGRAPHICAL
AND
LEXICOLOGICAL
PROJECTS
Mexican
terms
for
sexual
affairs
are
in
fact
loans
from
other
languages
or
other
terminologies
from
other
dialectal
systems.
Third,
we
will
include
the
written
works
of
renowned
Mexican
researchers
and
educators.
The
Sexual
Education
and
Sexological
Research
community
in
Mexico
is
small
but
very
productive.
The
Mexican
Federation
of
Sexual
Education
and
Sexology
(Federación
Mexicana
de
Educación
Sexual
y
Sexología,
FEMESS)
has
members
of
very
diverse
disciplines,
among
them:
Medicine,
Sexual
Education,
Sexual
Health,
Culture
and
Ideology
of
Sexuality
and
Politics
of
Sexuality.
Throughout
the
years
its
members
have
built
their
own
frameworks
and
teaching
methods.
Much
of
their
written
work
has
not
been
published,
so
we
have
requested
that
they
provide
us
with
their
notes
and
other
unpublished
materials.
Regarding
sociolinguistic
research,
the
Corpus
will
contain
results
from
a
poll
designed
to
obtain
the
terms
used
by
people
to
refer
to
sexual
matters
(anatomic
parts,
sexual
stereotypes,
contraceptives,
sexual
transmissible
diseases,
derogatory
words
with
sexual
connotations,
etc).
4
Due
to
the
complexity
ofthe
sociolinguistic
dimension,
further
steps
can
be
added
at
a
later
stage
to
this
strategy.
A
reasonable
goal
for
this
proposal
is
to
apply
such
inquiry
to
a
variety
of
small
groups
of
society
members
(workers,
university
students,
adults
over
40,
etc.).
5
Needless
to
say,
sociolinguistic
research
is
radically
different
from
the
compilation
of
experts'
texts.
Our
Corpus
will
contain
both
sources
of
information;
so,
both
types
of
discourse
wiU
be
subjected
to
the
same
statistical
investigations.
However,
some
aspects
of
the
poll
may
not
allow
us
to
elicit
appropriate
contexts
of
term
use
(consider,
for
instance,
one
word
or
enumerative
answers
to
certain
poll
questions).
Nevertheless,
we
can
obtain
important
quantitative
data
by
including
poll
results
(in
a
specific
format)
in
the
Corpus.
And
we
do
expect
the
poll
to
yield,
after
all,
some
data
about
the
terms'
uses
and
typical
contexts.
Another
strategy
we
decided
to
implement
was
to
collect
electronic
material
with
sexual
connotations
available
in
the
biternet
(chats,
groups,
lists).
Although
questions
may
be
raised
about
how
representative
this
is
of
Mexican
Speech,
it
does
represent
a
sample
of
spontaneous
discourse
among
a
segment
ofyoung
people.
bi
order
to
illustrate
how
the
final
dictionary
may
profit
from
this
variety
of
information
sources,
we
have
provided
in
Table
1
some
terms
related
to
the
notion
of
bisexuality.
The
first
two
are
representative
ofthe
colloquial
register
and
the
last
one
stands
for
the
formal
register.
The
last
column
exhibits
some
typical
contexts
in
which
the
terms
occur.
Notice
how
the
contexts
ofuse
ofthe
colloquial
terms
resemble
spoken
conversation.
We
expect
our
sociolinguistic
research
to
yield
this
kind
of
information:
colloquial
terms
and
their
contexts
will
come
mostly
from
the
polls.
Also,
examine
the
different
meanings
displayed
for
the
formal
term.
Different
text
sources,
representing
distinct
schools
of
thought,
may
conceive
phenomena
differently.
Their
points
of
view
may
be
reflected
by
listing
the
definitions
which
best
describe
them.
Thus,
whether
being
bisexual
is
a
matter
of
preference,
attraction
or
biology
will
not
be
a
matter
of
discussion;
these
points
of
view
will
simply
be
presented
as
the
possible
senses
of
the
term.
Meanwhile,
a
truly
prescriptive
dictionary
would
present
one
point
of
view
as
the
correct
one.
393

EURALEX2004
PROCEEDINGS
Terms
Definitions
Uses
and
contexts
Bi
(s.
y
adj.;
coloquial)
-»Bisexual
(apócope).
[..].
Te
dije
que
aquí
sólo
viene
gente
•
.
Bicicleta
(s.
y
adj.;
coloquial).
-»Bisexual.
[..].
No
te
imaginas,
como
bicicleta
que
era,
bien
que
le
pedaleaba
tanto
a
Rosita
como
a
Jorge.
Bisexual
(s.
y
adj.)
a)
(formal)
-»Preferencia
sexual
de
las
personas
que
no
sienten
mayor
atracción
por
personas
de
algún
sexo
o
género
en
particular
[..].
b)
(formal)
Atracción
sexual
por
personas
de
ambos
sexos
y/o
que
tiene
relaciones
sexuales
indistintamente
con
ellas
[..]
z)
(coloquial,
peyorativo)
—»Homosexual
no
asumido
f..1.
El
bisexual
masculino
propone
y
representa
el
goce
como
una
forma
de
ascesis.
El
paciente
se
asumió
bisexual
al
principio
de
la
consulta.
Los
pinches
bisexuales
son
locas
de
clóset.
Table
1:
Illustration
ofterms
and
their
contexts
Lastly,
technical
terms
may
have
become
part
ofpeople's
common
vocabulary.
And
the
typical
uses
of
those
terms
may
reflect
the
people's
negative
attitude
towards
certain
sexual
matters
such
as
bisexuality
(see
sense
'z)'
ofthird
term
in
table
1).
We
have
in
fact
noticed
in
preliminary
polls
that
well
over
haLf
the
terms
we
have
elicited
are
classified
as
negative
by
the
very
same
people
polled.
The
marking
of
terms
as
pejorative
will
constitute
a
portrait
of
regular
people's
attitudes
towards
sexuality.
Hence,
it
would
be
desirable
to
improve
the
sociolinguistic
poll
and
to
apply
it
periodically
in
order
to
reflect
changes
of
attitude,
and
the
emergence
of
neologisms.
This
does
not
exclude
the
addition
of
expert
materials
to
the
Corpus.
•
short,
this
project
can
grow
indefinitely.
Later
versions
ofthe
dictionary
may
even
include
encyclopedic
information.
But
in
this
paper
it
would
be
very
premature
to
sketch
more
than
what
has
been
dealt
with.
To
summarize,
the
following
points
are
the
key
ideas
mentioned
above:
1.
hi
the
face
of
lack
of
consensus
among
specialists,
regarding
the
thematic
areas
of
then-
field,
find
a
neutral
authority
respected
by
all
specialists.
2.
biclude
in
the
Corpus
translations
of
widely
read
books
in
the
field.
It
is
important
to
select
those
translations
read
by
the
experts;
especially,
those
used
in
the
training
of
researchers
and
educators.
Also,
important
works
from
other
Spanish
speaking
countries
should
be
considered.
394

REPORTS
ONLEXlCOGRAPHlCAL
AND
LEXICOLOGICAL
PROJECTS
3.
biclude
in
the
Corpus
the
work
of
renowned
experts
in
the
field
written
in
Spanish,
published
or
not.
A
balanced
representation
of
the
local
community
deserves
to
appear
in
the
Corpus.
4.
Design
a
strategy
for
obtaining
sociolinguistic
data.
One
way
to
accomplish
this
is
by
designing
and
applying
a
poll
in
order
to
obtain
lexical
information.
A
possible
complement
is
the
inclusion
ofbiternet
materials
in
the
Corpus.
3.
Closing
Remarks
The
gathering
of
representative
discourse
of
the
field
is
the
first
step
towards
constructing
the
electronic
Corpus.
Some
materials
are
ah^eady
available
in
electronic
form.
Most
of
them
are
being
scanned.
The
target
size
for
the
Corpus
is
two
million
tokens.
This
should
be
an
adequate
size
for
obtaining
a
terminology
of
2,500-5,000
types.
The
Corpus
will
be
available
on
the
biternet
for
researchers
and
educators
to
use
as
a
reference
tool.
Term
contexts
and
definitions
found
there
will
be
consulted
for
the
writing
ofthe
target
definitions
of
the
final
dictionary.
A
concordancing
tool
will
naturally
be
provided
for
this
and
any
general
purpose
exploration
ofthe
Corpus.
Once
a
set
of
terms
has
been
extracted,
it
will
be
necessary
to
determine
how
each
term
relates
to
each
other.
More
importantly,
grouping
them
according
to
their
lexical
relations
and
building
with
them
a
hierarchical
conceptual
structure
will
give
an
idea
ofwhat
is
missing.
Text
sampling
errors
can
be
repaired
and
missing
terms
can
be
detected
by
building
such
a
structure.
Then,
specific
research
can
be
conducted
to
fill
the
empty
spaces
in
the
conceptual
system
represented
by
the
terms.
The
set
of
terms
resulting
from
this
will
also
be
available
on
the
biternet.
The
idea
is
to
build
a
data
base
that
can
host
them,
their
definitions
and
other
relevant
lexicographical
data.
Finally,
given
the
nature
of
language,
it
is
important
to
notice
that
both
the
Corpus
and
the
terminology
extracted
are
expected
to
change
with
time.
That
is
why
the
writing
of
the
definitions
and
of
future
versions
of
the
dictionary
will
require
that
both
remain
in
the
hiternet
for
an
indefinite
length
oftime.
Endnotes
'For
example,
radio
and
television
commentator
Anabel
Ochoa's
La
palabra
común,
Diccionario
erótico
México-España,
Mexico,
Colofón,
2002.
2
For
instance,
the
Chilean
Osvaldo
Quijada's
Diccionario
integrado
de
sexología,
Madrid,
•
•
•
•
•
•
•
,
1983.
3
Electronic
dictionaries
offer
many
advantages.
For
instance,
very
sophisticated
searching
possibiUties
of
semasiological
and
onomasiological
types
(Sierra
&
McNaught,
2000).
Also,
multimedia
and
game
components
make
them
very
attractive.
4
Labov's
conversation
modules
are
a
desirable
method
to
be
considered
for
this
poll
0^abov,
1984).
Also,
research
experience
previously
conducted
in
Mexico
to
collect
lexical
items
from
children
should
be
taken
into
account
(López
Chávez,
1993).
'Naturally,
this
can
only
be
a
first
step
towards
a
wider
investigation
across
the
regional
and
social
boundaries
ofthe
country.
Moreover,
this
does
not
exclude
the
possible
inclusion
in
the
corpus
of
other
sexual
research
polls
previously
conducted
in
the
country.
395

EURALEX2004
PROCEEDDiGS
References
Biber,
D,
Conrad,
S.
and
Reppen,
R.
1998.
Corpus
Linguistics.
Cambridge:
Cambridge
UP.
Cabré,
M.
T.
1993.
La
terminología.
Teoría,
metodología,
aplicaciones.
Barcelona:
Antartica.
Cabré,
M.
T.
2000.
'Elements
for
a
Theory
of
Terminology:
Towards
an
Alternative
Paradigm.'
Terminology,
mternational
Journal
of
Theoretical
and
Applied
Issues
in
Specialized
Communication
6:l,pp.
35-57.
Labov,
W.
1984.
'Field
Methods
ofthe
Project
on
Linguistic
Change
and
Variation.'
bi:
Baugh,
J.
and
Sherzer,
J.
(eds.)
Language
in
Use:
Readings
in
Sociolinguistics.
EngIewood
Cliffs,
NJ:
Prentice-Hall,
pp.
28-53.
Lara,
L.
F.,
Ham
Chande,
R.
and
Garcia
Hidalgo,
M.
I.
1979.
Investigaciones
lingüísticas
en
lexicografia.
Mexico:
El
Colegio
de
Mexico.
Lara,
L.
F.
1990.
Dimensiones
de
la
lexicografia.
A
propósito
del
Diccionario
del
Español
de
México.
Mexico:
El
Colegio
de
México.
López
Chávez,
J.
and
Meza
Canales,
R.
M.
1993.
Léxico
disponible
de
preescolares
mexicanos.
Mexico:
Facultad
de
Filosofía
y
Letras,
UNAM.
McEnery,
T.
and
Wilson,
A.
1996.
Corpus
Linguistics.
Edinburgh:
Edinburgh
UP.
Medina,
A.
2001.
'Diccionario
de
terrmnología
de
la
sexualidaden
México,
rV
Congreso
Nacional
de
Educación
Sexual
y
Sexología.
Congress
organized
by
the
Mexican
Federation
ofSexual
Education.
Veracruz,
Veracruz,
Mexico,
March,
2001.
Medina,
A.
and
Sierra,
G.
2003.
'Construcción
de
un
sistema
lexicográfico
como
apoyo
a
la
educación
sexual',
V
Congreso
Nacional
de
Educación
Sexual
y
Sexología.
Congress
organized
by
the
Mexican
Federation
of
Sexual
Education.
Morelia,
Michoacán,
Mexico,
August,
2003.
Pozzi,
M.
1996.
'BTMEX:
A
flexible
Terminological
Data-Management
System.'
Terminology,
mternational
Journal
of
Theoretical
and
Applied
Issues
in
Specialized
Communication
3:1,
pp.
111-124.
Sager,
J.
C.
1990.
A
Practical
Course
in
Terminology
Processing.
Amsterdam/
Filadelfia:
John
Benjamins.
Sierra,
G.
and
McNaught,
J.
2000.
'Design
of
an
Onomasiological
Search
System:
A
Concept-Oriented
Tool
for
Terminology.'
Terminology,
mternational
Journal
of
Theoretical
and
Applied
Issues
in
SpeciaHzed
Communication
6:l,pp.
1-34.
Sierra,
G.,
Medina,
A.,
Moreno,
M.,
Garduño,
G.
and
Castillo,
G.
2002.
'Diccionario
básico
de
sexualidad'.
Paper
presented
by
Marlene
Moreno
in
the
Congreso
Latinoamericano
de
Multimedieros
Universitarios
2002.
Congress
organized
by
the
University
Association
for
Multimedia,
National
Autonomous
University
of
Mexico,
Mexico
City,
November,
2002.
396