ArticlePDF Available

¿Así me escucho cuando hablo?: Attitudes towards Spanish, English, and Code-Switching in Two Texas Border Towns.



Content may be subject to copyright.
is is a contribution from Spanish in Context 12:2
© 2015. John Benjamins Publishing Company
is electronic file may not be altered in any way.
e author(s) of this article is/are permitted to use this PDF file to generate printed copies to be
used by way of oprints, for their personal use only.
Permission is granted by the publishers to post this file on a closed server which is accessible
only to members (students and faculty) of the author’s/s’ institute. It is not permitted to post
this PDF on the internet, or to share it on sites such as Mendeley, ResearchGate,
Please see our rights policy on
For any other use of this material prior written permission should be obtained from the
publishers or through the Copyright Clearance Center (for USA:
Please contact or consult our website:
John Benjamins Publishing Company
Spanish in Context 12:2 (2015), 177198. doi 10.1075/sic.12.2.01ran
issn 15710718 / e-issn 15710726 © John Benjamins Publishing Company
“Is that what I sound like when I speak?”
Attitudes towards Spanish, English, and
code-switching in two Texas border towns*
Natalie Rangel, Verónica Loureiro-Rodríguez and
María Irene Moyna
is study investigates language attitudes towards English, Spanish and code-
switching in two Texas border cities (Laredo and Edinburg) by means of a
matched guise test with three attribute dimensions (solidarity, status, and
personal appeal). It was found that there were no signicant overall dierences
between attitudes in the two cities. As anticipated, code-switching received the
lowest ratings in all dimensions; English and Spanish were matched for status,
and Spanish received the highest scores for solidarity. When the variable of
gender was considered (both for raters and speakers), dierences in ratings
emerged, evidence that in both cities the three varieties play dierent roles in
mediating gender relations.
Keywords: language attitudes, matched-guise, Texas-Mexico border
1. Introduction
e 2,000-mile border between Texas and Mexico is characterized by long-stand-
ing language contact between Spanish and English and pervasive individual and
societal bilingualism (Bills, Hernández Chávez, and Hudson 1995). Studies con-
ducted in the Rio Grande Valley (Anderson-Mejías 2005, Mejías and Anderson
1988, Mejías, Anderson-Mejías, and Carlson 2003) show that both English and
Spanish play an important role in the daily lives of people throughout the region.
Yet, beyond these general common trends, there may be regional dierences
worth exploring. In order to ascertain whether such dierences exist, the present
study compares two border cities, Edinburg, located in Hidalgo County within the
RGV, and Laredo, located 150 miles west of the valley and not a part of the same
geographic and economic area.
© 2015. John Benjamins Publishing Company
All rights reserved
178 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
Laredo is one of the fastest growing cities in the nation, with a predomi-
nantly Hispanic population (95.6%) of about 250,000 (US Census Bureau 2010).
With four international bridges, it is the most important inland port along the
United States-Mexico border, and NAFTA’s primary port and trade corridor
(International Trade 2012). e city is closely allied to its namesake and sister city
on the Mexican side.
e Río Grande Valley (RGV), dened by the census as a Metropolitan
Statistical Area and made up of four counties (Starr, Hidalgo, Cameron, and
Willacy), comprises over 60 communities and cities linked to two major metro ar-
eas, Brownsville-Harlingen and McAllen-Edinburg-Mission. As a whole, the RGV
has a population of 1,264,091, 90% of whom are Hispanic (U.S. Census Bureau
2010). Although traditionally rural, today the area’s economy is based on gov-
ernment, education, health care, retail, and international trade over eight bridges
(Alonzo 1998, 53). A land of contrasts, the RGV is also home to the poorest com-
munities in the nation, or colonias, unincorporated settlements that lack adequate
water resources, paved roads, and housing (Arispe y Azevedo 2009).
With so many similarities, one could expect language behaviors and attitudes
in Laredo and the RGV to be the same. However, our starting hypothesis, par-
tially conrmed by our study, is that there are dierences between the perceptions
of speakers in the two locations towards the three varieties under consideration.
Since the hypothesis is based on the diverse socio-historical factors at play in each
location, we briey present these in the next section.
2. Historical background
2.1 e Borderlands and U.S. Annexation
South Texas is dened by the presence of the Río Grande, also known as the Río
Bravo, which runs from southwestern Colorado to the Gulf of Mexico. From the
earliest human settlement in the area, the river has been a lifeline, not only as a re-
liable source of water in this hot dry region, but also as a means of transportation.
Before the arrival of Europeans, and as early as ca. A.D. 1300, these lands were
already inhabited by Indian tribes belonging to the Coahuiltecan linguistic family
and organized in small autonomous bands engaged in trade (Adams 2008, 2).
During his rst expedition between 1528–1536, Alvar Núñez Cabeza de Vaca
had some contact with the local indigenous peoples, but the history of Tejanos in
South Texas began in earnest in the middle of the 18th century. It was then that
Spanish frontier captains surveyed the land and saw its potential for breeding live-
stock. is prompted the organization of entradas by the Viceroy of New Spain,
© 2015. John Benjamins Publishing Company
All rights reserved
Attitudes towards Spanish, English, and code-switching in two Texas border towns 179
which were led by José de Escandón (Adams 2008, 6; Alonzo 1998, 16) and re-
sulted in the establishment of six settlements in the lower Río Grande. Among the
rst was the Villa de San Agustín de Laredo, founded in 1755 and thus the earliest
of the Texas-Mexico border towns (Adams 2008, 11–12; Alonzo 1998, 31). e city
prospered thanks to its ranching and because it was a good port for the crossing of
goods from the Texan interior (Adams 2008, 1).
e other area of interest for this study is the Rio Grande Valley, located at the
southeastern-most point of the present border between the U.S. and Mexico. e
earliest Spanish settlements, including Santa Ana de Camargo, Lugar de Mier, San
Juan de los Esteros (present-day Matamoros), were established by the Spaniards
on the south bank of the river. RGV cities such as Brownsville, Edinburg, and
McAllen were established much later, in the late 19th and early 20th centuries.
By the time of Mexico’s independence, the population of the villas along the
Río Grande had increased from 5,053 in 1794 to 13,956 in 1823 (Alonzo 1998,
61). e Mexican trade that started along the river eventually drew the interest of
foreign merchants and Anglo American impresarios, who started to arrive in the
1820s and 30s. e opening of the port at Matamoros (now bordering the Texan
city of Brownsville) started international commerce in the RGV (Alonzo 1998,
67–68) and in the 1820s the city prospered thanks to its new status. Laredo, too,
would benet from international trade along the river, as the prime north-south
crossing site for goods and merchandise (Adams 2008, 62). It exported cotton and
tobacco, and imported European-manufactured goods into Mexico, soon surpass-
ing the RGV in volume of commerce (Hickey Cavazos 2012, 19).
Eventually, the colonization eorts by Anglos led to the breaking o of Texas
as an independent Republic, later incorporated into the US. e ensuing disputes
between the US and Mexico led to the Mexican-American war (1846–1848), an
unequal match which resulted in the loss by Mexico of over half of its territory.
Although the treaty that ended the war gave guarantees to former Mexican citi-
zens, Texas was deeply changed by the war, in demographic, legal, and cultural
terms. Nowhere were these changes starker and more traumatic than in South
Texas. Once the Río Grande was declared the separation between the U.S. and
Mexico, the north and south banks, which had been a single community, were
split. In places like Laredo many inhabitants moved to the west bank and founded
Nuevo Laredo in the state of Tamaulipas (Hickey Cavazos 2012, 24). Additional
demographic changes were wrought by the arrival of Anglos, lured by opportuni-
ties for economic growth.
e established Mexican Texans had dierent legal views on land ownership
from the newly arrived Anglos and the government that backed them. For Tejanos,
who followed Spanish laws and customs, land had belonged to the king, but had
been given to the people that could bring the monarch prosperity. Families oen
© 2015. John Benjamins Publishing Company
All rights reserved
180 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
held land grants for generations without titles. On the other hand, for Anglo
Americans, land belonged to the government and was a source of revenue through
sale and property taxes. Written documents were considered the only valid means
to prove ownership. is set the stage for a massive process of land dispossession
in South Texas. However, the populations of the two areas under study reacted dif-
ferently, and this made a big dierence in the outcome.
In Laredo, Tejano elites strategically allied themselves to the newly arrived
Anglos by dividing into dierent partisan groups, each with Anglo allies (Alonzo
1998, 125). ey also intermarried with the newcomer professionals and mer-
chants. Additionally, Laredoans took preemptive measures to prevent loss of
lands. In 1847, they petitioned the Texas legislature for assurances on the protec-
tion of their property rights (Alonzo 1998, 47). us, when a land commission was
established in 1850 to evaluate claims (Alonzo 1998, 152), the citizens of Laredo
succeeded in conrming theirs. ings went dierently for land claims submit-
ted from Hidalgo County, in the RGV. ere, many claims from well-known and
powerful landowners were denied by the commission and/or later lost on appeal.
is led to the purchase of lands by Anglos and the ensuing founding of cities in
those areas, which explains their later dates of establishment (Brownsville in 1848,
Mission in 1908, Edinburg in 1919, and McAllen in 1927).
e status of Spanish, especially in the public domain, also underwent changes
aer Annexation. e local population had to perform a delicate balancing act,
adjusting to the superimposition of English as the ocial language while striv-
ing to maintain their traditions (Monday and Vick 2007, 27). Public education
was an arena where these conicting pressures became evident. Mexican Texans
were very concerned about education, in particular for their children (Alonzo
1998, 126). Laredo and the RGV saw the emergence of educational institutions,
both private and religious, and public schools appeared shortly aer annexation.
Laredo residents in particular considered it important to know how to speak and
write well in Spanish (Hickey Cavazos 2012, 188) and successfully resisted the
imposition of English Only (McKenzie 2004, 82, Calderón 1993, 644).
2.2 e U.S.-Mexico border in the 20th Century
All through their history, U.S. border cities have been connected to and inuenced
by the Mexican side, in a symbiosis that continues to this day. Over the 20th cen-
tury, Mexico went from being a rural society to a highly industrialized country
with increasingly more and bigger cities. Northern urban centers were especially
aected, as international trade and local economies were boosted by the require-
ments of U.S. wars rst, and then by Mexico’s own growing demand for American
goods (Garza 2002). en, in 1965 the Mexican Border Industrialization Program
© 2015. John Benjamins Publishing Company
All rights reserved
Attitudes towards Spanish, English, and code-switching in two Texas border towns 181
encouraged the establishment of maquiladoras to increase employment and de-
crease poverty in the area; low labor costs, among other advantages, encouraged
multinational rms to establish their plants and manufacture goods for the U.S.
market (Coronado and Phillips 2007, 19). en, in 1994 NAFTA brought about a
reduction in taris, tripling U.S. exports to Mexico, and quadrupling imports from
that country (US Department of State 2010). is expansion led to further popula-
tion growth in Mexican border cities, which are generally larger than their corre-
sponding U.S. twin cities (compare Laredo with pop. 245,000 and Nuevo Laredo,
373,000). In turn, the disposable income of Mexicans has boosted the economy
of the U.S. side and led to one of the lowest unemployment rates in the country
(López 2006). It is estimated that as much as 39% of Laredo’s and 33% of McAllen’s
goods are exported into Mexico through retail purchases (Coronado and Phillips
2007, 30). Additionally, people cross the border frequently in either direction to
work, study, seek services, and socialize. e social and economic integration of
the border has aected both of the areas under study in similar ways, which may
have led to the erasure of earlier dierences.
e continued presence and inuence of Spanish has therefore been pervasive
all along the Texas-Mexico border. To the frequent presence of visitors and shop-
pers one should add the mass media broadcasting of major Mexican television
channels, such as Televisa and TV Azteca. Also, the majority of the radio stations
that air in Laredo and Edinburg come from Mexico. e few local stations are pre-
dominantly in Spanish or broadcast in a mix of Spanish and English.
Yet, the ocial language policies taken in the two areas under study have not
been identical. Laredo United Consolidated School District was the rst to in-
troduce bilingual programs in Texas in 1964. e idea eventually spread to other
areas of Texas, such as El Paso and San Antonio, as well as cities in RGV, like
McAllen, Edinburg, La Joya, and Mission (Rodriguez 2010). Today, bilingual and
ESL programs exist in both Laredo and RGV. However, the goals of the programs
dier between regions. While the bilingual departments of the districts in Laredo
emphasize the development of bi-literate, bilingual, and bicultural students, some
districts in RGV, including Edinburg, give more importance to the mastery of
English in listening, speaking, reading, and writing (Bilingual/ESL Department
2014, Bilingual/ESL/Dual Language/Foreign Language 2014, Bilingual Education
2013, 2014, Bilingual/ESL Oce 2014, Bilingual Education/ESL 2014).
e ultimate objective of our study is to ascertain whether the long-standing
historical dierences between Laredo and Edinburg have had attitudinal conse-
quences, or whether the later equalizing eect of national and international poli-
cies has eliminated these dierences. In the next section we discuss the application
of the matched-guise technique to the study of language varieties in contact.
© 2015. John Benjamins Publishing Company
All rights reserved
182 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
3. Measuring language attitudes: e matched-guise test
Language attitudes, understood as individuals’ predisposition to react favorably
or unfavorably towards certain linguistic varieties (Garrett 2010), are a useful tool
in the eld of sociolinguistics. Understanding individuals’ perceptions of lan-
guages or certain linguistic features, such as accents or code-switching, helps us
bridge the gap between the social and psychological aspects of language (Giles
and Coupland 1991). Measuring language attitudes, however, is not a simple task.
First, in the mentalist framework we adhere to, attitudes are an internal mental
state, and thus not directly observable (Appel and Muysken 1987, Fasold 1984).
Secondly, attitudes have aective, behavioral, and cognitive components (Garrett
2010), which may or may not be present at once in a given evaluation, and are
oen undistinguishable from one another (Bohner and Wanke 2002). Finally, the
cognitive and aective components of an attitude may not match the behavior
component (Garrett 2010); in other words, individuals’ feelings about a linguistic
variety may deviate from their actual use of such variety.
Sociolinguistic research following the mentalist framework investigates lan-
guage attitudes indirectly, seeking to answer the question of how an individual co-
vertly reacts to a linguistic variety and its speakers. e most common technique
to measure these implicit language attitudes is the matched-guise test, pioneered
by Lambert et al. (1960) to elicit covert reactions towards English and French in
Montreal. Lambert and his associates had four bilingual males record the same text
in English and French (i.e., eight ‘guises’), and two additional males, who acted as
llersor distractors by recording just one version of the text each. Participants
were told that the purpose of the study was to ascertain to what extent people
judge a speaker by simply hearing his voice, and were given a rating sheet to evalu-
ate each voice they would hear on the basis of fourteen traits on a 6-point Likert
scale. One of the key features of the matched-guise test is the deception involved
in the experiment. While participants are obviously aware that they are complet-
ing an attitude-rating task, they do not know that they are rating the same speaker
using two (or more) dierent linguistic varieties (Garrett 2010, 41). Lambert et al.
found that French-speaking participants evaluated the English guises much more
favorably in height, looks, leadership, intelligence, self-condence, dependability,
ambition, sociability, character, and likability. is particular outcome, that is, a
minority language group viewing the majority language and its speakers more fa-
vorably, is common in bilingual societies where there is a hegemonic language
(Romaine 1995, 289).
e matched-guise test has been used in numerous studies since Lambert et al.s
groundbreaking research, and subsequently concerns have been raised regarding
its limitations to measure covert language attitudes (see Garrett 2010 for a detailed
© 2015. John Benjamins Publishing Company
All rights reserved
Attitudes towards Spanish, English, and code-switching in two Texas border towns 183
overview). First, researchers must consider that participants may be reacting not
solely to the linguistic variety, but also to the voice. Moreover, the repetition of
texts may cause an exaggeration of the language variation, making participants
more aware of the distinction between linguistic varieties than they normally
would in their regular environment. Another disadvantage of the matched-guise
test relates to the issue of authentic production of the accent or the mimicking of a
certain variety; if speakers cannot produce all varieties accurately, results will not
be reliable or valid. e use of written texts as vocal stimuli has been criticized on
the basis that they do not render the style of spontaneous speech, and thus par-
ticipants will likely evaluate the reading dierently. Researchers are encouraged to
choose a passage with neutral content, to avoid the risk of participants reacting to
the content instead of the linguistic variety. However, even if the text chosen for
the experiment is assumed to be “factually neutral”, one must be aware that each
individual will interpret it in a dierent way according to pre-existing social sche-
mata (Garrett 2010). Finally, the administrative procedures of the matched-guise
test have also been the object of criticism. Researchers oen administer this test
in group settings, such as classrooms, which reduces the ecological validity of the
method. Moreover, the judgments required of participants may force them to rely
on the properties of the language to a larger extent than what they would normally
do if they had other non-linguistic cues (Solís Obiols 2002). Due to all these issues
concerning the validity of the matched-guise test, a multi-method approach to the
study of language attitudes is recommended (Edwards 1982). erefore, recent
work in this eld employs a combination of matched-guise tests and direct meth-
ods such as interviews, questionnaires, and surveys (Pieras-Guasp 2002, Hoare
2001, Ihemere 2006).
To this date, the matched-guise test has not been employed to investigate at-
titudes in the bilingual area of South Texas. However, a few attitudinal studies
have been carried out using a combination of direct and other indirect methods,
such as the verbal-guise test. is technique diers from the matched-guise test
in that, instead of asking the same speaker to read the dierent versions of the
text, a range of speakers is used to record each passage. e verbal-guise test is
oen used when researchers are unable to obtain authentic-sounding guises from
a single speaker. De la Zerda Flores and Hopper (1975) opted for this technique to
explore attitudes towards standard English, standard Mexican Spanish, accented
English and Tex-Mex (which they dened as “Texas Spanish”) amongst Mexican-
Americans in San Antonio. e authors found that accented English and Tex-Mex
were rated less favorably by all participants, except for those who self-identied as
Chicano/a. Another key nding was that less educated, lower income subjects re-
acted very positively towards both standard varieties and were the least tolerant of
the non-standard varieties. By contrast, participants with some college education
© 2015. John Benjamins Publishing Company
All rights reserved
184 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
reacted more favorably towards accented English than standard English, and rated
non-standard varieties higher than the less educated group. Finally, none of the
groups in De la Zerda Flores and Hopper’s study reacted negatively towards stan-
dard Mexican Spanish.
Using a questionnaire, Mejías and Anderson (1988) and Mejías et al. (2003)
investigated attitudes towards Spanish language among Mexican-American stu-
dents and professionals (mostly lawyers and physicians) in the Lower Rio Grande
Valley (LRGV). is is the rst, and so far the only, longitudinal study of attitudes
in Texas. In their early study, the authors found that students used Spanish more
frequently for communication reasons, that is, for achievement of social ends,
than for instrumental or sentimental reasons. ey also found that professional
females used Spanish for sentimental reasons, while their male counterparts had
a more instrumental perception of their own language use. Finally, participants
whose families have been established in the United States for generations viewed
Spanish language as a value, while newer arrivals did not share this view (405). In
2003 the researchers replicated the study, expecting to nd dierences in attitudes,
especially amongst students. However, despite the 20-year gap, results were very
similar. Studentsmotivations for using Spanish were still primarily communica-
tive, but a slight increase in the dimension of language loyalty was also found.
Mejías et al.s ndings also suggest that in the LRGV, although there are signals that
indicate a potential shi to English, the maintenance of Spanish is overwhelmingly
supported by factors such as the proximity to Mexico and the constant inux of
Spanish monolingual immigrants (149). Dailey et al. (2005), using a verbal-guise
test in Southern California, reached a similar conclusion regarding the relation-
ship between attitudes and linguistic landscape. ey found that Hispanics that
lived in a linguistic landscape that favored Spanish had less positive attitudes to-
wards Anglo-accented speakers, while those who perceived more English in their
landscape had more positive attitudes towards Anglo speakers.
By means of interview and conversational data, Galindo (1995) investigated
the linguistic attitudes of Chicano adolescents in two bilingual communities in
Austin, Texas. She found that females who preferred to use Spanish or both Spanish
and English, and came from households where the use of Spanish was encour-
aged, had more positive attitudes and a higher sense of loyalty towards Spanish
than males (95). Galindo also found conicting attitudes among her participants.
While some of them had negative perceptions about Spanish and its speakers, they
also expressed their desire to maintain the language alive for future generations,
as they perceived it as an essential part of their heritage. Furthermore, these ado-
lescents perceived that speaking English with a Spanish accent could prevent an
individual from gaining access to certain social groups (96).
© 2015. John Benjamins Publishing Company
All rights reserved
Attitudes towards Spanish, English, and code-switching in two Texas border towns 185
Following a similar qualitative approach, Velázquez (2009) explored language
attitudes and language maintenance among a group of adults from El Paso, Texas,
a border community where Spanish has a strong presence in the public domain.
e author interviewed and observed ve middle-class families with parents who
self-identied as Hispanic and were native speakers of either Chicano or Mexican
Spanish (74). Velázquez found that these parents had the perception that neither
standard Mexican Spanish nor standard English corresponded with the varieties
spoken in El Paso and, similarly to Galindo’s participants, they believed that speak-
ing English without a Spanish accent was crucial to avoid discrimination and also
for class mobility. ese ndings, Velázquez points out, need to be understood “as
a consequence of the underlying tensions present in a community such as El Paso,
where intense linguistic and interethnic contact takes place every day” (81–82).
4. Methodology
4.1 Research questions
To investigate attitudes towards standard Mexican Spanish, standard English and
code-switching in the Texas/USA border, we administered a matched-guise test
among bilingual college students in Edinburg and Laredo. ese are the specic
questions we sought to answer:
1. Are bilingual speakers perceived dierently depending on the linguistic vari-
ety they speak and their gender?
2. Is code-switching regarded positively among bilingual speakers?
3. Do bilinguals from Edinburg and Laredo react dierently towards standard
Mexican Spanish, standard English and code-switching?
4. Do bilingual males hold dierent attitudes from bilingual females?
4.2 Matched-guise test
For our matched-guise test, we prepared three versions (standard Mexican
Spanish, standard English, and code-switching) of the same passage to use as lis-
tening stimuli (Figure 1). e original text contained English/Spanish switches
and was obtained from the Spanish in Texas website (http://www.coerll.utexas.
edu/), a project established by the Center for Open Educational Resources and
Language Learning (COERLL) at the University of Texas. Before preparing the
standard English and Mexican Spanish versions of the text, we edited them for
length and conducted a grammatically judgment test among 10 bilinguals to en-
sure that participants would react to the code-switching nature of the text and not
© 2015. John Benjamins Publishing Company
All rights reserved
186 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
to its grammatical and syntactic irregularities (Toribio 2001). Aer incorporating
the comments from the bilingual judges, the following passages were used:
Code-switched passage:
Iba a todas mis classes, y luego a lunch, then I’d wait for Julie to get there. Siempre la espe-
raba en la library. Ahí era donde nos juntábamos, and then we’d go together to, ya sea a la
cafeteria o a una store que estaba ahí, to eat. I’d wait for the bus y nos llevaba a la casa. O
si no, I’d call my mom pa’ que viniera por mí. If not, me quedaba para band practice o for
some activity que tenía que hacer.
Standard Mexican Spanish passage:
Iba a toda mis clases, y luego seguía la comida, y luego esperaba a que Julie llegara. Siempre
la esperaba en la biblioteca. Ahí era donde nos juntábamos, y luego nos íbamos juntas a, ya
sea la cafeteria o a una tiendita que estaba ahí para comer. Esperaba al camión y nos llevaba
a la casa. O si no le hablaba a mi mamá para que viniera por mí. Si no, me quedaba a la
práctica del grupo de banda o para alguna actividad que tenía que hacer.
Standard English passage:
I would go to all of my classes, and then it was lunch. en I’d wait for Julie to get there; I’d
always wait for her at the library. at’s where we’d get together, and then we’d go together
to either the cafeteria or to a small store that was there to eat. I’d wait for the bus and it’d
take us home. Or if not, I’d call my mom to pick me up. If not, I’d stay for band practice or
for some other activity that I had to do.
Figure 1. Texts used as listening stimuli.
Four English/Mexican Spanish bilingual speakers (2 males and 2 females) were
recorded reading the three passages. Two other bilinguals (1 male, 1 female) were
used as llers and were asked to read the English and Spanish texts only. A total
of sixteen recordings (12 guises and 4 llers) were generated. e six bilinguals
recorded were rst generation Mexican-Americans who had lived in Texas most
of their lives. eir oral production in English and Spanish was native-like and
undistinguishable from that of monolinguals from the United States and Mexico.
eir age and education level closely matched those of the raters, in that they were
between 21 and 25, and were attending university.
4.3 Data collection
e matched-guise test was administered among bilingual students taking Spanish
classes at Texas A&M International University in Laredo and at the University of
Texas-Pan American in Edinburg. Because listening to all sixteen guises would
have likely caused participant fatigue, we divided the recordings into two groups
of eight, so each group of subjects listened to one set only. During the administra-
tion of the test, we made sure that an equivalent number of participants in both
cities listened to each set of recordings.
© 2015. John Benjamins Publishing Company
All rights reserved
Attitudes towards Spanish, English, and code-switching in two Texas border towns 187
In the classrooms, participants were told that they would listen to eight dif-
ferent speakers reading the same text in dierent linguistic varieties. To reduce
the possibility that participants would notice the repetition of speakers, speak-
ers’ guises were maximally spaced apart. Also, the llers were randomly placed
amongst the other recordings to further enhance the deception of the test. For our
analysis, we only considered the linguistic guises of the four speakers who read
all three versions of the passage. In order to assure data quality, we only included
responses from participants who were born and grew up in Laredo and Edinburg.
Data from students who le one or more responses blank (either in the rating
sheet or in the sociolinguistic questionnaire) or showed evidence of malingering
(e.g., all responses were ‘5’ or ‘1’) were discarded (n = 17).
Before conducting the matched-guise experiment, participants were given a
sociolinguistic questionnaire that included items about their age, sex, language
use, and parents’ nationality and occupation. en, they were presented with the
auditory segment of the experiment. Aer listening to each of the eight record-
ings, participants rated the speaker on 16 personal attributes on a 5-point Likert
scale (1 = Not at all, 5 = Very). Attributes were grouped a priori into three dimen-
sions and ordered randomly on the rating sheet. e dimensions and attributes
used were: solidarity (honest, kind, likeable, friendly, sincere, and religious), status
(rich, old-fashioned, educated, high class, and urban), and personal appeal (in-
telligent, hardworking, good-looking, funny, and open-minded). e quantitative
score for each dimension is the average of its contributing attributes.
4.4 Statistical methods
A multi-step analysis was performed to analyze our data, including an analysis to
determine if the linguistic variety was signicant, a matched pairs analysis using
the data from each individual student, and a matched pairs analysis between 30
students from Laredo and 30 from Edinburg. Each dimension score was modeled
with linear regression with xed eects of location (Laredo or Edinburg), par-
ticipant gender (male or female), speaker gender (male or female), and speaker
language (Spanish, English or code-switching). A matched pairs t-test analysis was
used to determine dierences within and between locations. To compare dier-
ences between locations, participants were paired up by matching their responses
from the sociolinguistic questionnaire. at is, subjects from either location were
matched by considering the number of identical answers that they gave for each
question and by exactly or closely matching their age. In total, 60 participants
(30 from each location) were used to determine dierences between Laredo and
Edinburg. e Shapiro-Wilk test was performed to ascertain the dierences for
© 2015. John Benjamins Publishing Company
All rights reserved
188 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
normality in each of the matched pairs tests. All data manipulations and statistical
analysis were carried out using Statistical Analysis Soware (SAS).
5. Results
5.1 Descriptive statistics
Participants (n = 187) ranged in age between 18 and 35 years, and they all attended
Spanish classes at Texas A&M International University in Laredo (n = 96) and at
the University of Texas-Pan American in Edinburg (n = 91). Eighty-one percent
of the participants were female (n = 152) and nineteen percent were male (n = 35)
(see Figure 2). For the present study, students’ language use and their parents’ ori-
gin and occupation were not included as variables.
0Laredo Edinburg
Figure 2. Participants by location and gender.
5.2 Statistical analysis
5.2.1 Location
No statistically signicant dierences at a 5% level were found for the interaction
between location and linguistic variety. In other words, contrary to our prediction,
location by itself was not a variable in determining participants’ attitudes towards
Spanish, English and code-switching.
5.2.2 Linguistic variety
While no statistically signicant dierences were found when comparing Spanish
and English, statistically signicant dierences at a 5% level were found when
comparing Spanish and English with code-switching. For the three dimensions,
code-switching was rated lower than both Spanish and English (see Figure 3).
© 2015. John Benjamins Publishing Company
All rights reserved
Attitudes towards Spanish, English, and code-switching in two Texas border towns 189
Solidarity Status Personal Appeal
Figure 3. Dimension average ratings for Spanish, English and code –switching.
5.2.3 Solidarity
English and code-switching guises. In Laredo and Edinburg, male participants gave
similar ratings to English and code-switching, regardless of the gender of the
speaker. Female participants in Laredo gave higher ratings to the English guises,
regardless of the gender of the speaker (p-value = 0.0034 for female speakers and
p-value = 0.0151 for male speakers). Female participants in Edinburg gave higher
ratings to English when the speaker was male (p-value = 0.0005) (see Figure 4).
Spanish and code-switching guises. In Laredo, participants gave higher ratings
to Spanish guises overall, regardless of the gender of the speaker (p-values = 0.0369
and 0.0001 for male and female raters respectively). In Edinburg, male participants
gave higher ratings to Spanish for both male and female speakers (p-values = 0.0134
and 0.0166 respectively), and female participants also rated Spanish guises higher
regardless of the gender of the speaker (p-value = 0.0001) (see Figure 4).
Spanish and English guises. In Laredo, male participants gave similar ratings to
Spanish and English, but the male speakers received higher ratings when speak-
ing Spanish (p-value = 0.0107). Female participants rated Spanish guises higher
regardless of the gender of the speaker (p-value = 0.0001). In Edinburg, male
2CS fs ENG fs SPAN fs CS ms ENG ms
Laredo males
Laredo females
Edinburg males
Edinburg females
Figure 4. Solidarity average ratings for code-switching, English and Spanish female and
male speakers
© 2015. John Benjamins Publishing Company
All rights reserved
190 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
participants gave higher ratings to Spanish overall (p-value = 0.0268); no statisti-
cally signicant dierence was found when the speakers were male, but the female
speakers were rated higher in their Spanish guise (p-value = 0.0056). Female par-
ticipants in Edinburg rated Spanish guises higher regardless of the gender of the
speaker (p-value = 0.0006) (see Figure 4).
5.2.4 Status
English and code-switching guises. Male participants in Laredo gave higher ratings
to English overall (p-value = 0.0089); no statistically signicant dierences were
found when the speaker was male, but the female speakers were rated higher when
speaking in English (p-value = 0.0351). eir Edinburg counterparts also gave
higher ratings to English overall (p-value = 0.0001), and statistically signicant dif-
ferences between guises were found for male and female speakers (p-value = 0.0009
and 0.0002 respectively). In both locations, female participants rated English high-
er regardless of the gender of the speaker (p-value = 0.0001) (see Figure 5).
Spanish and code-switching guises. In Laredo, male and female participants gave
higher ratings to Spanish overall (p-value = 0.0003 for males and p-value = 0.0001
for females), regardless of the gender of the speaker. e same was true for partici-
pants in Edinburg (p-value = 0.0001 for males and females) (see Figure 5).
Spanish and English guises. In Laredo, male participants gave similar scores to
Spanish and English, regardless of the gender of the speaker. eir female coun-
terparts gave higher ratings to Spanish overall (p-value = 0.0077); no statistically
signicant dierences were found when the speaker was female, but females in
Laredo gave higher ratings to male speakers when using Spanish (p-value = 0.0077).
In Edinburg, male participants gave similar scores to Spanish and English, but
there is a small statistically signicant dierence for the female speakers (p-val-
ue = 0.0418), who were perceived as having higher status when speaking English.
Female participants in Edinburg also rated Spanish and English similarly, but gave
Laredo males
Laredo females
Edinburg males
Edinburg females
CS fs ENG fs SPAN fs CS ms ENG ms SPAN ms
Figure 5. Status average ratings for code-switching, English and Spanish female and male
© 2015. John Benjamins Publishing Company
All rights reserved
Attitudes towards Spanish, English, and code-switching in two Texas border towns 191
higher scores to the female speakers when speaking Spanish (p-value = 0.0360)
(see Figure 5).
5.2.5 Personal appeal
English and code-switching guises. In Laredo, male participants gave similar scores
to English and code-switching, regardless of the gender of the speaker. Female
participants gave higher ratings to English overall, regardless of the gender
of the speaker (p-value = 0.001). In Edinburg, male participants rated English
higher overall (p-value = 0.0010); this dierence was statistically signicant for
both the female (p-value = 0.0259) and the male speakers (p-value = 0.0090) (see
Figure 6). Female participants in Edinburg rated English higher overall as well (p-
value = 0.0001), and found the female speakers more personally appealing when
using English (p-value = 0.0028), whereas dierences for male speakers were not
statistically signicant (see Figure 6).
Spanish and code-switching guises. In Laredo, male participants gave similar
scores to Spanish and code-switching, regardless of the gender of the speaker.
Female participants gave higher ratings to Spanish overall, regardless of the gen-
der of the speaker (p-value = 0.001). In Edinburg, male participants rated Spanish
higher than code-switching overall (p-value = 0.0007), and found the female
speaker more personally appealing when using Spanish (p-value = 0.0042), where-
as dierences for male speakers were not statistically signicant. Female partici-
pants in Edinburg gave higher ratings to Spanish guises regardless of the gender of
the speaker (p-value = 0.0001) (see Figure 6).
Spanish and English guises. In Laredo, male participants gave similar scores to
Spanish and English, regardless of the gender of the speaker. Laredo female par-
ticipants rated Spanish higher overall (p-value = 0.0001), and rated both the female
and the male speaker higher in their Spanish guises (p-values = 0.0373 and 0.0004
respectively). In Edinburg, male participants gave similar scores to Spanish and
Laredo males
Laredo females
Edinburg males
Edinburg females
CS fs ENG fs SPAN fs CS ms ENG ms SPAN ms
Figure 6. Personal appeal average ratings for code-switching, English and Spanish female
and male speakers.
© 2015. John Benjamins Publishing Company
All rights reserved
192 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
English, and found the male speaker more personally appealing when speaking
English (p-value = 0.0138). Female participants in Edinburg gave higher scores to
Spanish guises (p-value = 0.0202), and gave higher scores to the female speakers
when using Spanish (p-value = 0.0269) (see Figure 6).
6. Discussion
e present study conrmed the verbal guise research of De la Zerda Flores &
Hopper (1975) with respect to the low favorability ratings of code switching. is
is not surprising, given the common tendency for minority language speakers to
accept the stigma placed on their language variety by the socially dominant ma-
jority (Romaine 1995, 289), even in areas such as Laredo and the RGV, where
these speech styles are general in the local community. Still, it is noteworthy that
the stigma of code-switching has persisted aer almost forty years since the rst
study, and to a similar extent in cities closer to the border, where this speech style
is socially pervasive. One possible explanation is that, even in the minds of ha-
bitual code-switchers, ‘ideal’ bilinguals are supposed to behave linguistically like
monolinguals, and thus all switching is assumed to be motivated by a linguistic
deciency (cf. Zentella 1985, 53, for explicit references to this assumption among
Puerto Ricans in New York) .
One of the most interesting ndings of the present study was that standard
Mexican Spanish was evaluated more favorably for solidarity than English in both
locations. is was a departure from De la Zerda Flores & Hopper (1975), where
both standard varieties received similar evaluations. Yet, the earlier study antici-
pates the present results, in the sense that the authors found that participants who
had close ties with Mexico favored Spanish more than others. Given the location
of both Edinburg and Laredo, immediate ties may very well be behind the positive
views towards Spanish. A second possibility could be change over time. One gen-
eration aer the earlier study, attitudes towards Spanish may have become more
positive. Be that as it may, despite the centuries-old superimposition of English,
Spanish maintains a stronghold in the Mexican American culture, admired as a
symbol of these participants’ heritage (De la Zerda Flores & Hopper 1975).
One of the main innovations of this study is the consideration of both female
and male guises, a factor not considered in De la Zerda Flores & Hopper (1975).
is detail proved to be critical to show dierences within locations. For example,
in Laredo, male raters evaluated Spanish and English dierently by gender, favor-
ing Spanish when the speaker was male. Laredo males also evaluated all three
linguistic varieties the same under personal appeal. ese males then appear to
view the non-standard about the same as the standard varieties, revealing a covert
© 2015. John Benjamins Publishing Company
All rights reserved
Attitudes towards Spanish, English, and code-switching in two Texas border towns 193
acceptance of code-switching. On the other hand, female raters exhibited prefer-
ences independently of speaker gender, preferring Spanish across the board. For
their part, all raters in Edinburg favored Spanish to English when the speaker was
a female. ese dierences seem to reect a more general trend for women to be
linguistically status-conscious (Trudgill 1972), using or perceiving standard va-
rieties as the most favorable. In the border areas under study, these tendencies
could be reinforced by cultural and educational practices, given that women are
much more likely to enroll in higher education (17,433 females vs. 14,433 males
in Laredo; 75,673 females vs. 66,355 males in the RGV, according to the United
States Census Bureau). In other words, local speakers seem to have the reasonable
expectation that women will use standard prestige forms, while the bar is set lower
for men (cf. Loureiro-Rodríguez et al. 2013 for a similar nding for Galicia) .
Speaker/rater gender was also a determining factor for dierences between
Laredo and Edinburg, with solidarity exhibiting the largest gap. One of the most
interesting ndings was that female students from Edinburg were slightly more ac-
cepting towards code-switching than those in Laredo, particularly in male speak-
ers. e long and rich cultural history of Laredo and the high social prestige ac-
corded to Spanish in that city may be behind the stigmatization of code-switching.
For example, the early existence of Spanish newspapers in Laredo in the nineteenth
century (Hickey Cavazos 2012, 104–108), may have created stronger prescriptive
views towards the use of standard Mexican Spanish that endure to this day.
7. Conclusions
In this section we consider certain methodological and theoretical issues worth
bearing in mind in future studies of this kind. e rst issue to note is that, in
an experimental study on location like the one described here, it is not always
possible to completely control conditions. For example, the acoustic properties of
the rooms used to collect data may aect rating scores. In the present study, one
of the rooms used in Laredo was larger than any of those available in Edinburg.
Although the same audio equipment was used, the dierences in size aected
acoustics and may have been behind lower scores for Laredo across all varieties. It
should be noted that these lower scores aected all varieties equally and were thus
not statistically signicant, but they point to the need for more careful control of
the testing environment.
A second issue that straddles methodological and theoretical concerns is
related to the characteristics of the code-switched text. First, the use of a doc-
tored code-switched text as part of the read recorded sample may have inuenced
the ratings negatively. Since code-switching normally occurs spontaneously, it is
© 2015. John Benjamins Publishing Company
All rights reserved
194 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
possible that the slightly articial delivery of the text may have aected the evalu-
ations of this variety. As we stated earlier, this has been identied as one of the
disadvantages of the matched-guise (Garrett 2010, 58–57), since it is dicult to
obtain a code-switched text that is both spontaneous-sounding and neutral in
content. Another characteristic of the code-switched text that may have aected
ratings is the frequency of switches. is feature was not controlled for in any way,
other than through the grammaticality judgments. However, it may be the case
that more or less frequent switches are acceptable to dierent degrees in various
locations. is is a matter for future research.
Our study proves the inuence of gender on the assessment of linguistic vari-
eties. However, several other independent variables were included in the question-
naire and can be considered in future analyses, such as frequency of use of code-
switching, parents’ occupation and place of origin, and language use preference.
Broader data may also be obtained from applying the questionnaire to a more di-
verse sample, including dierent age groups, social classes, and educational levels.
Given the results obtained by De la Zerda Flores & Hopper (1975) regarding the
higher tolerance of better-educated participants towards non-standard varieties,
we anticipate that these independent social variables will be meaningful, too.
Another interesting way to expand this study would be to collect data on the
Mexican side of the border, specically in the corresponding Mexican towns of
Nuevo Laredo and Reynosa. To the extent that speakers are bilingual in those cit-
ies, it may be possible to assess their attitudes towards the three varieties to see
whether they mirror those of the US border towns considered. A previous study
of attitudes that explored the issue through overt preferences (Martínez 2003)
showed that speakers from the Mexican border town of Reynosa tended to exhibit
more positive attitudes towards varieties closest to their own, provided that these
were on the Mexican side of the border. Although they acknowledged the great
similarities between their variety and that of the transborder twin city, whatever
small dierences they found were reason enough to rank the US Spanish sample
last. It would be interesting to ascertain the implicit attitudes of these Mexican
Spanish speakers in order to see whether they react negatively to features of border
Spanish or they in fact merely reject its mixing with English. One may speculate
that the persistent negative attitudes towards code switching on the US side of the
border are in fact fueled by prescriptive attitudes on the part of their neighbors on
the Mexican side.
e results of the present study could be supported by a multi-methodological
approach. Even without prompting, the participants made explicit comments dur-
ing the experiment, such as laughing or frowning in disapproval or making nega-
tive remarks upon listening to code-switching. is suggests that their evaluations
towards code-switching are the same explicitly as they are implicitly, but a more
© 2015. John Benjamins Publishing Company
All rights reserved
Attitudes towards Spanish, English, and code-switching in two Texas border towns 195
complete picture may emerge from a direct open-ended questionnaire. Clearly,
given the daily presence of code-switching not only in conversation but also in
commercials, billboards, and advertisements in both Laredo and the RGV, the per-
sistent prescriptive pressure against language mixing is complex and requires a
very rened analysis.
In every speech community, individuals attempt to rationalize linguistic vari-
ation by making connections between linguistic forms and the qualities of the
speakers who use them. As Irvine and Gal (2000: 37) explain, individuals’ ideolo-
gies about language situate linguistic variation as evidence for what they believe
to be systematic behavioural, aesthetic, aective, and moral contrasts among the
social groups indexed”. In the US-Mexico border context, it is important to note
that these attitudes may vary signicantly among communities and, as our re-
search shows, even within the same community and age group. In other words,
linguistic attitudes towards Spanish, English and code-switching may uctuate
depending on whether the location being studied is on the Texas or the Mexican
border, and on the changes in recent political and social identity of the location.
Variation in language and in attitudes in the US-Mexico border indicates ongoing
dynamic sociolinguistic changes and also reects the permeability of a border that
connects rather than divides. Our study suggests that a border does not always
represent a single and common identity or linguistic attitude, but rather several in
constant change.
Adams, John A. 2008. Conict & Commerce on the Rio Grande: Laredo, 1755-1955. College
Station: Texas A&M University Press.
Alonzo, Armando C. 1998. Tejano Legacy: Rancheros and Settlers in South Texas 1734-1900.
Albuquerque: University of New Mexico Press.
Anderson-Mejías, Pamela L. 2005. “Generation and Spanish Language Use in the Lower Rio
Grande Valley of Texas.” Southwest Journal of Linguistics 24 (1-2): 1–12.
Appel, R., and P. Muysken. 1987. Language Contact and Bilingualism. London: Edward Arnold.
Arispe y Azevedo, Baltazar, Jr. 2009. e Geographic and Demographic Challenges To the
Regional Institutionalization Of the Texas Lower Rio Grande Valley.Norteamérica 4 (2):
Bilingual/ESL Department. 2014. In Laredo Independent School District. Retrieved on May 3,
2014 from
Bilingual/ESL/Dual Language/Foreign Language. 2014. In United Independent School District.
Retrieved on May 3, 2014 from
Bilingual Education. 2014. In Brownsville Independent. Retrieved on May 3, 2014 from http://
© 2015. John Benjamins Publishing Company
All rights reserved
196 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
Bilingual/ESL Oce. 2014. In ECISD. Retrieved on May 3, 2014 from
Bilingual Education/ESL. 2014. In McAllen Independent School District. Retrieved on May 3,
2014 from
Bilingual Education. 2013. In Mission CISD. Retrieved on May 3, 2014 from
Bills, Garland D., Eduardo Hernández Chávez, and Alan Hudson. 1995. e Geography of
Language Shi: Distance from the Mexican Border and Spanish Language Claiming in the
Southwestern US.International Journal of the Sociology of Language 114 (1): 9–28.
DOI: 10.1515/ijsl.1995.114.9
Bohner, G., and M. Wanke. 2002. Attitudes and Attitude Change. Hove: Psychology Press.
Calderón, Roberto Ramón. 1993. “Mexican Politics in the American era, 1846-1900: Laredo.
University of California.
Coronado, Roberto A., and Keith Phillips. 2007. “Exported Retail Sales along the Texas-Mexico
Border.Journal of Borderlands Studies 22 (1): 19–37. DOI: 10.1080/08865655.2007.9695667
Dailey, René M., Howard Giles, and Laura L. Jansma. 2005. “Language Attitudes in an Anglo-
Hispanic Context: e Role of the Linguistic Landscape.Language and Communication 25:
27–38. DOI: 10.1016/j.langcom.2004.04.004
De la Zerda Flores, N., and Robert Hopper. 1975. “Mexican American’s Evaluations of Spoken
Spanish and English.Speech Monographs 42 (2): 91–98. DOI: 10.1080/03637757509375883
Edwards, John. 1982. “Language Attitudes and their Implications among English Speakers.In
Attitudes towards Language Variation: Social and Applied Contexts, ed. by Ryan Bouchard
and Howard Giles, 20–33. London: E. Arnold.
Fasold, R. 1984. e Sociolinguistics of Society. Oxford: Basil Blackwell.
Galindo, Letticia. 1995. “Language Attitudes Toward Spanish and English Varieties: A Chicano
Perspective.Hispanic Journal of Behavioral Sciences 17 (1): 77–79.
DOI: 10.1177/07399863950171005
Garrett, Peter. 2010. Attitudes to Language. Cambridge: Cambridge University Press.
DOI: 10.1017/CBO9780511844713
Garza, Gustavo. 2002. “Evolución de las ciudades mexicanas en el siglo XX.Notas: Revista de
información y análisis 19 (7–16).
Giles, Howard, and Nikolas Coupland. 1991. “Language Attitudes: Discursive, Contextual, and
Gerontological Considerations. In Bilingualism, Multiculturalism, and Second Language
Learning: e McGill Conference in Honour of Wallace E. Lambert, ed. by Allan G. Reynolds,
21–37. New Jersey/London: Lawrence Erlbaum.
Hickey Cavazos, Concepción María. 2012. Spanish language use and linguistic attitudes
in Laredo, Texas, between 1860 and 1930. Unpublished Ph.D. dissertation, Texas A&M
Hoare, Rachel. 2001. “An Integrative Approach to Language Attitudes and Identity in Brittany.
Journal of Sociolinguistics 5 (1, Feb): 73–84. DOI: 10.1111/1467-9481.00138
Ihemere, Kelechukwu U. 2006. “An Integrated Approach to the Study of Language Attitudes and
Change in Nigeria: e Case of the Ikwerre of Port Harcourt City.” 36th Annual Conference
on African Linguistics, 194–207.
International Trade. 2012. In Laredo Development Foundation. Retrieved on April 3, 2013 from
Irvine, Judith T., and Susan Gal. 2000. “Language ideology and Linguistic Dierentiation.” In
Regimes of Language, ed. by P. Kroskrity, 35–83. Santa Fe: SAR Press.
© 2015. John Benjamins Publishing Company
All rights reserved
Attitudes towards Spanish, English, and code-switching in two Texas border towns 197
Lambert, W. E., R. C. Hodgson, R. C. Gardner, and S. Fillenbaum. 1960. “Evaluational Reactions
to Spoken Languages.e Journal of Abnormal and Social Psychology 60 (1): 44–51.
DOI: 10.1037/h0044430
López, José Joaquín. 2006. “Dynamic growth in the Rio Grande Valley.Southwest Economy
March/April: 11–14.
Loureiro-Rodríguez, Verónica, May Boggess, and Anne Goldsmith. 2013. “Language Attitudes in
Galicia: Using the Matched-guise Test among High School Students.Journal of Multilingual
and Multicultural Development 34 (2): 136–153. DOI: 10.1080/01434632.2012.729591
Martínez, Glenn A. 2003. “Perceptions of Dialect in a Changing Society: Folk Linguistics along
the Texas-Mexico Border.Journal of Sociolinguistics 7: 38–49.
DOI: 10.1111/1467-9481.00210
McKenzie, Phyllis. 2004. e Mexican Texans. College Station: Texas A&M University Press.
Mejías, Hugo A., and Pamela L. Anderson. 1988. “Attitude toward Use of Spanish on the South
Texas Border.Hispania 71 (2): 401–407. DOI: 10.2307/343088
Mejías, Hugo A., Pamela L. Anderson-Mejías, and Ralp Carlson. 2003. “Attitude Update: Spanish
on the South Texas Border.Hispania 86 (1): 138–150. DOI: 10.2307/20062822
Monday, Jane Clements, and Frances Brannen Vick. 2007. Petra’s Legacy: e SouthTexas
Ranching Empire of Petra Vela and Miin Kenedy. College Station: Texas A&M University
Pieras-Guasp, Felipe. 2002. “Direct vs. Indirect Attitude Measurement and the Planning of
Catalan in Mallorca.Language Problems & Language Planning 26 (1): 51–68.
DOI: 10.1075/lplp.26.1.04pie
Rodriguez, Rodolfo. 2010. “Bilingual Education.Handbook of Texas Online.
Romaine, Suzanne. 1995. Bilingualism. Cambridge: Blackwell Publishers Inc.
Solís Obiols, Marina. 2002. “e Matched Guise Technique: A Critical Approximation to
a Classic Test for Formal Measurement of Language Attitudes.Noves SL. Revista de
Sociolingüística 1: 1–6.
Toribio, Almeida Jacqueline. 2001. “On the Emergence of Bilingual Code-Switching Competence.
Bilingualism: Language and Cognition 4 (3): 203–231. DOI: 10.1017/S1366728901000414
Trudgill, Peter. 1972. “Sex, Covert Prestige, and Linguistic Change in the Urban British English
of Norwich.Language in Society 1: 179–195. DOI: 10.1017/S0047404500000488
U.S. Census Bureau. 2010. American Community Survey. Retrieved on June 23, 2013 from
U.S. Department of State. 2010. “Factsheet: US-Mexico Trade and Investment at a Glance.
Accessed on May 16, 2014 from:
Velázquez, Isabel. 2009. “Intergenerational Spanish Transmission in El Paso, Texas. Parental
Perceptions of Cost/Benet.Spanish in Contact 6 (1): 69–84. DOI: 10.1075/sic.6.1.05vel
Zentella, Ana Celia. 1985. “e Fate of Spanish in the United States: e Puerto Rican
Experience.” In Language of Inequality, ed. by Nessa Wolfson and Joan Manes, 41–95.
Berlin/New York/Amsterdam: Mouton.
© 2015. John Benjamins Publishing Company
All rights reserved
198 Natalie Rangel, Verónica Loureiro-Rodríguez and María Irene Moyna
María Irene Moyna
Department of Hispanic Studies
Texas A & M University
College Station, TX 77843-4238
Authors’ address
Natalie Rangel
Department of Spanish and Portuguese
University of Texas at Austin
150 W 21st Street, Stop B3700, Austin,
TX 78712
Verónica Loureiro-Rodríguez
Department of Linguistics
University of Manitoba
Winnipeg, MB R3T 2N2
View publication statsView publication stats
... De la Zerda Flores and Hopper (1975) found that only those Texan Mexican-Americans who labelled themselves Chicanos (term used to refer to a North-American of Mexican descent) gave CS high ratings, and that less educated participants with lower incomes were the least tolerant of accented English and CS. More recently, Rangel, Loureiro-Rodríguez, and Moyna (2015) found that in the Texas border towns of Edinburg and Laredo, Spanish and English were matched for status, while Spanish was evaluated more favourably for solidarity than English, revealing participants' close ties with Mexico or a change over time in attitudes towards Spanish. Anderson and Toribio's (2007) study is the only one to date to have explored attitudes towards different types of Spanish-English CS, although they used written texts (versions of Little Red Riding Hood) in lieu of recorded voices. ...
This study examines Puerto Rican bilinguals’ attitudes towards five speech varieties (Spanish, English, Spanish with English lexical insertions, inter-sentential code-switching, and intra-sentential code-switching). While previous research on language attitudes in Puerto Rico has exclusively employed direct methods (i.e. interviews, surveys, questionnaires), ours is the first to use the matched-guise test. The participants are students at the University of Puerto Rico, Río Piedras campus, who report code-switching to different degrees. They judged the speech varieties on personality, socio-economic, and ethnicity/identity attributes; their judgements on each of the fifteen attributes were analysed with cumulative link mixed-effects models accounting for the ordinal nature of the ratings. Some findings, such as certain negative attitudes towards English and code-switching and an association between these speech varieties and attributes pertaining to higher socio-economic classes, replicate those from earlier studies in Puerto Rico. However, other results, coinciding with recent studies conducted in Puerto Rico, reflect an attitudinal change in progress, by which participants seem more accepting of lexical insertions and intra-sentential code-switching. New results emerging from this study evidence participants’ sensitivity towards different types of code-switching, hence, their awareness of the types of switching that predominate in different speech communities.
... In her obituary for Doug Sahm, music critic Margaret Moser would say that, "He wrote the music of a lifetime, a unique soundtrack for the Lone Star state of mind" (Moser 1999). It is thus hardly surprising that part of that soundtrack would be sung in a mixture of Spanish and English, a reflection of the linguistic reality of San Antonio and South Texas more generally (Anderson- Mejías, 2005;Mejías and Anderson, 1988;Mejías et al., 2002Mejías et al., , 2003Rangel et al., 2015;Sawyer, 1970), albeit one that is unusual, almost subversive, in musical lyrics. The implications of this artistic choice were not lost on the band. ...
This study analyzes Spanish-English code-switching in the music of the Texas Tornados, a bilingual-bicultural San Antonio band. Their entire repertoire was transcribed and analyzed quantitatively and qualitatively to ascertain the form and functions of code-switching. We found that 39% of songs included language mixing, with English being the most frequent matrix language and Spanish lexical insertions and inter-sentential switches prevailing. Lexical insertions are used to exoticize songs and for humorous effect, while inter-sentential code-switching presents similar ideas in sequence demonstrating high poetic virtuosity. Such artistic use of language represents the subaltern status of Spanish, reflecting the sociolinguistic reality of Texas.
The purpose of this article is to examine the institutionalization of a region of the United States of America: the Texas Lower Rio Grande Valley. It will use and build on theories of regional institutionalization, geography, and demography by Paasi, Harvey, Gilbert, and other theorists. This research asks how history, geography, and demography challenge the present and future regional institutionalization of the Texas Lower Rio Grande Valley. To answer this question, recent case studies on institutionalization and regional development in Canadian and U.S.-Mexican regions will be used to explain phenomena.
The matriarch of one of the most important families in Texas history, Petra Vela Kenedy has remained a shadowy presence in the annals of South Texas. In this biography of Petra Vela Kenedy, the authors not only tell her story but also relate the history of South Texas through a woman's perspective. Utilizing previously unpublished letters, journals, photographs, and other primary materials, the authors reveal the intimate stories of the families who for years dominated governments, land acquisition, commerce, and border politics along the Rio Grande and across the Wild Horse Desert. From Petra's early life in the landed ranchero society of northern Mexico, through her alliance with Luis Vidal-an officer in the Mexican army to whom she bore eight children-until her move to Brownsville after Vidal's death, Petra lived in Mexico. When she moved to Texas, having taken Vidal's name, she represented a link to the landed families of the region. Mifflin Kenedy, a steamboat captain who had first come to Texas during the Mexican War, married into her world, acquiring local respectability and stature when he took Petra as his wife. The story of their life together encompasses war, the taming of a frontier, the blending of cultures, the origin of a ranching empire, and the establishment of a foundation and trust that still endure today, giving millions to Texas through charitable gifts. An attractive woman of business acumen, strong religious convictions, and intense family loyalty, Petra Vela Kenedy's influence through her husband and her children left a legacy whose exploration is long overdue. Copyright © 2007 by Jane Clements Monday and Frances Brannen Vick. All rights reserved.
Just about everyone seems to have views about language. Language attitudes and language ideologies permeate our daily lives. Our competence, intelligence, friendliness, trustworthiness, social status, group memberships, and so on, are often judged from the way we communicate. Even the speed at which we speak can evoke reactions. And we often try to anticipate such judgements as we communicate. In this lively introduction, Peter Garrett draws upon research carried out over recent decades in order to discuss such attitudes and the implications they have for our use of language, for social advantage or discrimination, and for social identity. Using a range of examples that includes punctuation, words, grammar, pronunciation, accents, dialects and languages, this book explores the intricate and fascinating ways in which language influences our everyday thoughts, feelings and behaviour.
This article revisits attitudes toward Spanish among students attending the University of Texas-Pan American in the Lower Rio Grande Valley of Texas. Eighty-five percent of the undergraduate students at this institution consider themselves Hispanic. In part due to the presence of the university, this four-county area has continued to experience numerous social and economic changes over the past two decades. In 1982, data were collected among UT-Pan American students in an attempt to gauge their attitudes toward the Spanish language along four sociolinguistic attitudinal dimensions-communication, instrumental, sentimental, and value. The present article analyzes data similarly collected, eighteen years later, in 2000. Comparisons are made between the attitudes expressed and the demographic variables of gender, age, and generation. Conclusions regarding retention of Spanish are then drawn.
Attitudes - cognitive representations of our evaluation of ourselves, other people, things, actions, events, ideas - and attitude change have been a central concern in social psychology since the discipline began. People can and do have attitudes on an infinite range of things. But what are attitudes, how do we form them, and how can they be modified? This book provides the student with a comprehensive and accessible introduction to the basic issues in the psychological study of attitudes. It presents up-to-date coverage of the key issues that will be encountered in this area, including attitude formation and change, functions of attitudes, attitude measurement, attitudes as temporary constructs, persuasion processes, and prediction of behaviour from attitudes.