An approach to the lexical ambiguity caused by false cognates in Spanish L2. A corpus-based exploratory study

AN APPROACH TO THE LEXICAL
AMBIGUITY CAUSED BY FALSE COGNATES
IN SPANISH L2. A CORPUS-BASED
EXPLORATORY STUDY
Ma 
ıa Samped o Mella
Abs ac . This explo a o y s udy aims o offe new insigh s in o he lexical
ambigui y in he in e language a ising om he use o alse cogna es. The s udy is
based on w i en ex s om wo lea ne s’ co po a. U e ances including alse
cogna es we e e ie ed om bo h co po a and p esen ed o wo g oups o na i e
speake s who assessed hei co ec ness and s a ed whe he o no hey we e
ambiguous. While he fi s g oup was in o med be o ehand ha he sen ences had
been p oduced by non-na i e speake s (as well as he meaning o he alse
cogna es in hei L1), he second g oup had no in o ma ion ega ding he
speake s who w o e hem. The goal o his analysis is o shed ligh on he diffe en
pe cep ions ha can a ise om na i e s. non-na i e speake s’ p oduc ions,
ocusing on lexical ambigui y.
1. In oduc ion
False cogna es ha e been analyzed ex ensi ely in s udies o lexicog aphy,
language eaching and lea ning, ansla ion, and con as i e linguis ics.
Indeed, L2 s udies highligh he difficul ies ha lea ne s encoun e when
using hese kinds o wo ds (e.g. Roca Va ela 2010). Howe e , o he bes
o my knowledge no s udy has in es iga ed in de ail how alse cogna es
can c ea e ambigui y among hea e s. This may occu when a alse
cogna e is appa en ly co ec ly used in an u e ance, bu can be
in e p e ed in wo diffe en ways: wi h he meaning in he L1, and in
he L2 o he speake . This could be he case, o example, wi h he use o
he alse cogna e ac ualmen e [cu en ly] in Spanish by an English speake
in a con ex whe e he wo d can mean bo h ac ually and cu en ly,asin
El mayo p oblema que enemos ac ualmen e/en ealidad es la al a de
dine o [The bigges p oblem we cu en ly/ac ually ha e is he lack o
money]. In such case, a Spanish na i e hea e who has knowledge o
English may eel ha he sen ence is ambiguous, i.e., i can be unde s ood
in mo e han one way. This would be due o he ac ha i was said by an
English speake who may no use he wo d co ec ly in he a ge
language.
In o de o ca y ou he p esen s udy, some u e ances wi h alse
cogna es we e e ie ed om he Co pus de Ap endices de Espa~
nol
[Spanish Lea ne Co pus] (CAES) o he Ins i u o Ce an es and he
Uni e sidade de San iago de Compos ela (Spain), and he Co pus Esc i o
S udia Linguis ica ••(•) 2023, pp. 1–20. ©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on
behal o Edi o ial Boa d o S udia Linguis ica.
This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion-NonComme cial-NoDe i s License, which pe mi s use and
dis ibu ion in any medium, p o ided he o iginal wo k is p ope ly ci ed, he use is non-comme cial and no modifica ions o adap a ions a e made.
Published by John Wiley & Sons L d, 9600 Ga sing on Road, Ox o d OX4 2DQ, UK, and 350 Main S ee , Malden, MA 02148, USA
del Espa~
nol L2 [L2-Spanish W i en Co pus] (CEDEL2)o he
Uni e sidad de G anada (Spain). A ques ionnai e ha con ained eigh
i ems and wo dis ac o s was c ea ed based on hese u e ances, and
p esen ed o wo a ge g oups o na i e speake s, who assessed hei
co ec ness and indica ed when hey we e ambiguous. The con ol g oup
did no ha e any in o ma ion on he speake s who p oduced he
sen ences. Howe e , he c i ical g oup was in o med ha he sen ences
we e p oduced by non-na i e speake s and con ained some alse
cogna es, whose meanings in he speake s’ L1 we e also p o ided. The
main ocus o his s udy is o de e mine i he e is a diffe en pe cep ion in
he in e p e a ion o an u e ance when i is p oduced by a na i e o a
non-na i e speake , ocusing on he lexical ambigui y.
This pape is o ganized in 5 sec ions: sec ion 2 con ains a heo e ical
backg ound; sec ion 3 p esen s he objec i es and he esea ch ques ions
o his s udy; sec ion 4 includes he me hods, i.e., he desc ip ion o he
co pus da a, he pa icipan s, and he design o he ask; sec ion 5
p esen s he esul s and he discussion; las ly, sec ion 6 con ains he
conclusion.
2. Theo e ical backg ound
2.1. Na i e language s. in e language: he e o
The in e language is defined by Selinke (1972:214) as “a sepa a e
linguis ic sys em based on he obse able ou pu which esul s om a
lea ne ’s a emp ed p oduc ion o a TL [ a ge language] no m”. This
linguis ic sys em has a diffe en s uc u e om he lea ne ’s na i e and
a ge languages, and con ains linguis ic e o s caused by L1 ans e
(in e lingual e o s) and by difficul ies in he L2 lea ning p ocess
(in alingual e o s) (Co de 1967). In a b oade ca ego y o e o s,
Co de (1967) also makes a dis inc ion be ween mis akes, i.e., hose ha
he lea ne is able o sel -co ec , and e o s, which e eal he lea ne ’s
lack o knowledge.
E o s ha e been ega ded in many diffe en ways in he lea ning p ocess
h oughou he his o y o o eign language eaching. In adi ional
me hods, e o s we e judged nega i ely, and hey we e deemed o affec
he lea ning p ocess o he a ge language. Howe e , he in e language
heo y (Selinke 1972) and he e o analysis app oach (Co de 1967)
p esen ed a diffe en iew o he e o , which was seen as a de ice used by
lea ne s and eache s when dealing wi h he L2: by he o me when
lea ning he L2, by es ing ou hei hypo hesis abou he a ge language,
and by he la e as e idence o he aspec s o he L2 ha a e unclea o
s uden s. Today, he communica i e language eaching app oach con-
side s ha “e o s a e ole a ed and a e seen as a na u al ou come o he
de elopmen o communica i e skills” (La sen-F eeman 2011:159).
2 Ma 
ıa Samped o Mella
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License
In he li e a u e on Spanish as a o eign language (SFL) eaching, he
e o has been he subjec o nume ous classifica ions in a ious
ca ego ies since he 1990s (see Penad
es Ma 
ınez 2003 o u he
in o ma ion). Fu he mo e, s udies o da e ha e adop ed a ious
pe spec i es, and ha e unde aken a ious p ocedu es o analysis (Alba
Qui~
nones 2009, Val e de Ma eos 2012), such as con as i e analysis
(e.g. Samped o Mella 2021a,2021b) and e o analysis (e.g. Na a o
Gala 1999, Belda To ijos 2015, Samped o Mella & S
anchez Gu i
e ez
2017). Al hough he SFL li e a u e has ocused on g amma ical and
lexical e o s, in ecen yea s many au ho s ha e placed mo e impo ance
on he discou se and he in e ac ional le el. In his con ex , nume ous
s udies ha e ocused on an analysis o he “p agma ic ans e ”
(Thomas 1983:91) displayed by L2 lea ne s and he p agma ic ailu e.
In addi ion o linguis ic and discou se e o s, he in e language can
likewise be subjec o communica ion ailu es, due o non-na i e speake s
no being able o communica e and being inaccu a ely unde s ood.
Indeed, e en when communica ion is p io i ized o e linguis ic s uc-
u es, communica ion may occasionally s op a some poin . In hese
cases, i is necessa y o he speake s o cla i y hei messages o o he
audience o econs uc hem: “Communica ion in ol es wo speake s
ans e ing meanings o each o he using hese p eexis ing sen ences. In
cases whe e he e is a communica ion p oblem, a p ocedu e o analy ical
econs uc ion in e enes o eco e he lea ne ’s meaning obscu ed by
his choice o he w ong sen ence” (Hamil on 2001:81). Fo ins ance,
examples (1) and (2) ex ac ed om he CAES co pus illus a e a
con adic ion be ween wha he lea ne says (o w i es) and he logical
hinking o he in e locu o which may affec he communica ion
p ocess
1
:
(1) un lad 
on es 
a en qp ie o y 
el decid
ı hace algo pa a man ene la ida.
En 
o en un piso que es 
a un ba io cen al y quiso lle a se algo. en 
o
en la habici
on desde una en ana y mi 
o los al ededo es. la habi aci
on
es 
a bien deco ada, pa ece es pe enecien e a un ico. ¡Genial! el lad 
on
pens
o: aho a odos son mios. [...] Despu
es de e mina odo, se le
ocu 
ıo una buena idea que quiz
as puede echa una sies a en una
habi aci
on an g ande, en onces se ech
o debajo de una cama y no a 
o
mucho en do mi se.[L1-Chinese, B1; CAES co pus]
‘A hie is in ouble and he decided o do some hing o sa e his li e.
He came in o an apa men which is in a cen al neighbo hood and
wan ed o ake some hing. He came in o he oom h ough he
window and looked a ound. The bed oom was well deco a ed, seems
o belong o a ich pe son. G ea ! The hie hough : now e e y hing
1
In his s udy, he examples in Spanish a e shown as hey appea in he co pus (wi h
g amma ical and o hog aphical e o s). All ansla ions a e mine.
Lexical ambigui y caused by alse cogna es 3
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License
is mine (...) A e finishing e e y hing, he had he g ea idea o
pe haps aking a nap in a big bed oom, so he h ew himsel unde he
bed and i did no ake long o him o all asleep’.
(2) Como la econom
ıa comenz
o a baja desde hace 10 a~
nos, la emp esa
decidi
o de disminui el nume o de abajado . Al p incipio mi seccion
es u o muy bien [...] Pe o la en a de los p oduc os ui bajando poco a
poco desde hace 3 a~
nos a pesa de que la econom
ıa de Jap
on ecupe aba
g adualmen e y po in yo pe d
ı el abajo hace medio a~
no [L1-Japanese,
B1; CAES co pus]
‘As he economy s a ed going down en yea s ago, he company
decided o educe he numbe o employees. In he beginning my sec o
was doing e y well (...). Bu he sales o he p oduc s we e li le by
li le going down o h ee yea s al hough he economy o Japan was
g adually eco e ing and a las I los my job hal a yea ago’.
The e is a con adic ion be ween he eade ’s expec a ions and wha he
lea ne w o e in bo h examples. In example (1), i seems s ange o a
pe son o ake a nap unde he bed, so he choice o he ad e b o place
(debajo [unde ] ins ead o encima [on]) seems o be w ong. Howe e , he
ac ha he pe son conce ned is a hie who may wan o hide om he
owne o he apa men makes his si ua ion mo e plausible. As we do
no ha e any mo e in o ma ion abou his example, we could no
gua an ee whe he his is an e o o no .
In example (2), howe e , i is mo e likely o be an e o in he choice o
discou se ma ke : he lea ne uses a connec o which exp esses elie o
s a e ha he has los his job. Al hough his could be a eason o eeling
be e (e.g. i he wo king condi ions we e no en i ely a o able), he
con en o he ex sugges s ha i is an e o . Indeed, Ramos
Sa~
nudo (2021), based on a lea ne co pus-based analysis, obse es ha
mos Spanish L2 s uden s do no no ice he meaning o his discou se
ma ke which is used as a closing discou se ma ke o indica e he final
poin o he ex : “ha quedado pa en e la dificul ad que exis e con espec o a
po in yal in, dos ma cado es de cie e con un cla o alo modal, que sin
emba go no son pe cibidos como ales po los ap endices” [i is clea ha
he e is a difficul y ega ding po in [a las ] and al in [a e all], wo closing
ma ke s wi h a clea modal alue ha a e ne e heless no pe cei ed as such
by [L2 Spanish] lea ne s] (Ramos Sa~
nudo 2021:277).
I is in e es ing o no e ha examples (1) and (2) make sense, despi e
hei g amma ical and o hog aphical e o s. I hey we e p oduced by a
na i e speake , hey would lea e no oom o doub and ambigui y.
Howe e , he ac ha hey we e p oduced by a non-na i e speake
lea es oom o doub and a double in e p e a ion: he hea e s expec he
speake no o ha e ull compe ence in he L2, and hey a e he e o e
mo e likely o commi e o s when p oducing o al o w i en ex s.
4 Ma 
ıa Samped o Mella
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License
Fu he mo e, when he con en o he message comes in o conflic wi h
he hea e s’ expec a ions, hey will p obably conside he ex o be
w ong o o con ain an ambigui y, in ha i could ha e mo e han one
in e p e a ion. The “ambigui y e o ” ha can ha e a significan effec on
he co ec in e p e a ion o a ex is hus included in some classifica ions
o lea ne s’ e o s in he Spanish li e a u e as a ype o communica i e
e o (c . V
azquez 1991). Howe e , in con as o he syn ac ic ambigui y
e o due o he o de o he wo ds in he sen ence (e.g. Fe n
andez
Sole a 2002, Belda To ijos 2015), he e is no ex ensi e bibliog aphy
abou he lexical ambigui y e o in SFL, o e en in he li e a u e on
English as a o eign language
2
.
2.2. Lexical ambigui y and in e p e a ion
Ambigui y in languages e e s o a linguis ic phenomenon in which a
ph ase o a sen ence has wo o mo e defini e meanings o eadings.
Phone ic, syn ac ic and lexical ambigui y a e cons an ly p esen in
languages (Ullmann 1976). In pa icula , lexical ambigui y is caused by
he homonymy and he polysemy o he languages, i.e. he ela ionship
be ween wo ds wi h iden ical o ms bu diffe en meanings, and he
coexis ence o many possible meanings o a wo d o ph ase espec i ely
(see Ho no & L
opez 2020 o u he in o ma ion and a didac ic
p oposal).
In addi ion o he lexical ambigui y due o he homonymous and
polysemic wo ds, a hi d ype mus be conside ed in he in e language:
he ambigui y caused by alse cogna es.Cogna es a e wo ds ha sha e a
simila meaning, spelling, and p onuncia ion in wo o mo e diffe en
languages. Fo example, he wo ds pe son and academic sha e he same
o igin and meaning, bu do no ha e exac ly he same spelling in English,
Po uguese (pe soa, acad
emico) and Spanish (pe sona, acad
emico). In
addi ion o he same o igin and meaning, cogna es some imes also ha e
he same spelling; his is he case o wo ds like pa icula and indi idual in
he h ee men ioned languages ( o u he in o ma ion, see
S ameno 2009).
As languages con ain many cogna es, lea ne s end o ans e
simila wo ds om L1 in o L2, o om an L2 in o a hi d o eign
language. This is an au oma ic and subconscious use o pas lea ne
beha io s in he a emp o p oduce new esponses. Indeed,
Wang (2021:253) highligh s ha “[i]n he p ocess o lea ning a second
2
The e is ex ensi e li e a u e conce ning he “ambiguous e o ” which is no ela ed o
he ype o ambigui y desc ibed in his pape . In he English li e a u e, “[a]mbiguous e o s
a e hose ha could be classified as ei he de elopmen al o in e lingual e o s because hey
eflec e o s ha an L1 lea ne commonly makes while a he same ime eflec he s uc u e
o he L2 lea ne ’s mo he ongue. An example o his would be He no ha e ime o a
Spanish L1 speake lea ning English” (Shaffe 2005:3).
Lexical ambigui y caused by alse cogna es 5
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License

language, college s uden s jus y o find he equi alen s in he na i e
language. Thus, ambigui ies appea in second language lea ning wi h
he fi s language in e e ence”. This in e e ence may become an
obs acle when ins ead o ans e ing a cogna e om one language o
ano he , L2 s uden s choose a alse cogna e. Acco ding o S ameno
(2009:219–220),
False cogna es e e o pai s o wo ds in he same o diffe en languages
ha a e simila in o m and meaning bu ha e diffe en oo s, i.e., do no
sha e a common o igin. They a e jux aposed, on he one hand, o cogna es
and, on he o he , o alse iends (o aux amis) ha a e pai s o wo ds in
wo languages ha look and/o sound simila , bu diffe in meaning.
In a b oade pe spec i e adop ed he e, alse cogna es a e alse iends,
i.e. wo ds ha a e iden ical o simila in o m in wo o mo e
languages (e en i hey do no sha e he same o igin o e ymons), bu
which diffe in meaning. Following a seman ic classifica ion, hey can
be di ided in o “ o al” and “pa ial” alse iends. “To al alse iends
imply a conspicuous seman ic diffe ence be ween he L2 and he L1:
English and Spanish in his case (e.g. English ase s. Spanish aso
[glass], English a ocado s. Spanish abogado [lawye ], English obe s.
Spanish obo [ he ])” (Roca Va ela 2010:718). Pa ial alse iends
p esen a seman ic o e lap, since wo simila wo ds ha e one sha ed
meaning and one diffe en meaning, due o he polysemy. Fo ins ance,
he Spanish noun ci culaci
on means ‘blood ci cula ion’, like he English
wo d ci cula ion, bu i also means ‘people o affic mo emen ’ in
Spanish.
In his con ex , he decoding o a message and he abili y o unde s and
and in e p e a message may be influenced i he audience knows ha i
was p oduced by a non-na i e speake , especially i he message in
ques ion seems o be unusual o unexpec ed, as in examples (1–2). False
cogna es may also lead o doub among he hea e s, and lead o hem
unde s anding a message in a comple ely diffe en way om wha he
encode was ying o con ey, e.g. (3):
(3) oy a habla e de las acaciones que he hecho es e e ano a Mad id y
oy comenza po deci que la ciudad es espan osa me ha gus ado
muchiss
ımo isi a el museo del p ado la casa eal y lo Re i o [...] pe o
lo que m
as he gus ado ue la noche de Mad id mad e mia que guay, he
hablado con un mon 
on de pe sonas in e esan es [...] ue an unas
acaciones que me ha gus ado mucho y quie o ol e a Mad id [L1-
Po uguese, A2; CAES co pus]
‘I’m going o alk o you abou he holidays I had his summe in
Mad id and I’m going o s a by saying ha he ci y is ho ible. I liked
isi ing he Museo del P ado, he Casa<sic>Real and he Re i o [pa k]
a lo (...) bu wha I liked he mos was he nigh li e o Mad id. Oh my
6 Ma 
ıa Samped o Mella
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License
God! So cool! I alked wi h a lo o in e es ing people (...) hey we e
g ea holidays ha I liked a lo and I wan o go back o Mad id’.
This is clea ly an e o due o a L1 nega i e ans e : he wo d espan oso
has exac ly he same spelling in Spanish and Po uguese, bu means
‘ e ific’ in Po uguese ( he L1 o he lea ne ) and ‘ho ible’ in Spanish
( he a ge language). In his example he e is no oom o ambigui y,
since he e is a b oade con ex which cla ifies ha he use o his
adjec i e is undoub edly an e o . Howe e , i he example only con ains
he fi s sen ence (la ciudad es espan osa), he e would be a leas wo
possible scena ios: 1) he hea e would unde s and comple ely he
opposi e meaning, i.e. Mad id is ho ible; 2) he hea e would no ice
some ambigui y, i hey knew ha he lea ne ’s L1 was Po uguese and
hey we e a Po uguese language use (e.g. a Spanish eache in a
Po uguese-speaking coun y). The ac ha he speake o he message is
non-na i e and he hea e ’s p e ious knowledge o hei L1 may influence
he hea e , in ha hey may find ambigui y in con ex s whe e i would no
exis i he ex we e p oduced by a na i e speake . I is p ecisely hese
cases o “hea e ’s ambigui y” due o alse cogna es ha a e a he hea o
his pape .
3. Objec i es and esea ch ques ions
This s udy examines diffe en ways in which na i e speake s in e p e L2
Spanish u e ances, ega dless o whe he hey know ha hey we e
p oduced by non-na i e speake s. Two ac o s a e conside ed: on he one
hand, he possible ambigui y due o he use o a alse cogna e; on he
o he , he co ec ness o he u e ances.
The pa icipan s we e asked o a e en sen ences, o which eigh
con ained a alse cogna e and wo we e dis ac o s wi h a polysemic and
a homonymous noun. One g oup o pa icipan s ( he c i ical g oup)
knew ha he u e ances we e p oduced by L2-Spanish lea ne s and
con ained some possible alse cogna es while he second g oup ( he
con ol g oup) did no ha e his in o ma ion. The aim o his design is o
espond o he ollowing esea ch ques ions:
(1) A e Spanish speake s mo e inclined o find lexical ambigui y i
hey know ha sen ences ha e been p oduced by non-na i e
speake s?
(2) Do Spanish speake s a e sen ences diffe en ly when hey know
ha hey ha e been p oduced by non-na i e speake s?
The main hypo hesis is ha hea e s a e mo e c i ical and mo e inclined
o de ec ambigui y when hey know ha speake s a e non-na i e. I is
he e o e impo an o in es iga e whe he pa icipan s ha e diffe en
Lexical ambigui y caused by alse cogna es 7
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License
pe cep ions owa ds he speake s who made he u e ances depending on
whe he hey ha e his in o ma ion o no .
4. Me hods
4.1. Co pus da a: sea ch and esul s
This s udy is based on he Co pus de Ap endices de Espa~
nol (CAES)
(Spanish Lea ne Co pus) and he Co pus Esc i o del Espa~
nol-L2
(CEDEL2) (L2-Spanish W i en Co pus). Bo h co po a con ain w i en
ex s p oduced by lea ne s o Spanish as a o eign language a a ious
le els, om A1 o C1 (CAES) and om A1 o C1+(CEDEL2), acco ding
o he Common Eu opean F amewo k o Re e ence o Languages
(Council o Eu ope 2001). In hei cu en o ms, he CAES and
CEDEL2 co po a con ain w i en p oduc ions om lea ne s o en
common na i e languages: A abic, Manda in Chinese, Japanese, F ench,
Po uguese, I alian, G eek, English, Ge man, and Russian. The CAES
also includes da a om L1-Polish lea ne s and CEDEL2 om L1-Du ch
lea ne s. The CAES co pus cu en ly con ains o e 970,000 wo ds, and
he CEDEL2 o e 1,100,000 wo ds.
The sea ch ool o bo h co po a allows e ie al o s a is ical
in o ma ion and ex ual examples o elemen s, lemmas, wo d classes
and g amma ical ca ego ies, wi h fil e s on he pa ame e s ha make up
he co pus (basically he lea ne ’s L1 and le el o p oficiency in
Spanish, bu also age, sex, coun y o o igin, e c.). I is also possible o
dis inguish be ween lowe and highe case wo ds, and accen ed and
non-accen ed wo ds. Howe e , nei he CAES no CEDEL2 include an
e o ag sys em which would acili a e e ie al o he diffe en ypes o
e o s made by lea ne s (see Rojo & Palacios 2016, o u he
in o ma ion abou he CAES co pus and Lozano 2022 o an o e iew
o he CEDEL2 co pus).
When ca ying ou his s udy, he fi s s ep was o sea ch in bo h
co po a o a se o diffe en wo ds ha a e alse cogna es in Spanish and
in ano he language (e.g. success in English s. suceso [inciden ] in
Spanish). A e ca e ully eading all he ou pu (i.e. he esul s), i was
obse ed ha mos o he cases we e clea ly co ec o inco ec
u e ances. Fo example: Es e mismo d
ıa, o o suceso lleg
o a susci a de
nue o emociones 
agicas (L1-G eek) [The same day, ano he inciden
again p o oked agic emo ions] s. Y in en a *p ocu a po la mad e, pe o
sin *suceso (L1-Po uguese) [and [he] ies o * y [look o ] he mo he ,
bu wi hou *inciden [success]). Only a ew ins ances could hus be
conside ed ambiguous. The e a e a o al o wo English-Spanish and wo
Po uguese-Spanish alse cogna es; wo o hem a e o al alse iends,
while he o he wo a e pa ial alse iends (see Table 1):
8 Ma 
ıa Samped o Mella
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License
A o al o eigh ambiguous u e ances in se en ex s which con ain hese
alse cogna es we e e ie ed om he co po a, and used o design he
expe imen al ask (3.2). I is impo an o no e ha hese alse cogna es a e
also polysemic wo ds in Spanish. Fo example, among o he meanings, he
Spanish e b a ende means ‘ o pay a en ion’, ‘ o help’ and ‘ o wa ch o e ’;
aco da (se) means ‘ o ag ee o’ o ‘ o se a da e’ as well as ‘ o emembe ’,
e c. Un o una ely, ambiguous u e ances con aining alse cogna es ha
we e nei he polysemic no homonymic wo ds in Spanish we e no ound in
any co pus a all. Howe e , in spi e o hese addi ional meanings, he
u e ances only ha e one in e p e a ion, in addi ion o he possible
ambigui y de i ed om he use o he alse cogna e.
4.2. Pa icipan s and ask
Fo y-six na i e speake s o Spanish pa icipa ed in his s udy. Thi y-six
we e emales and en we e males, and hei mean age was 21 yea s old, wi h
a minimum o 20 and a maximum o 28. All o hem we e en olled on
linguis ic cou ses a he Uni e si dade de San iago de Compos ela in Spain.
They we e di ided in o wo diffe en independen g oups, and he
esea che ca ied ou all he da a collec ion, in o de o ensu e ha he
da a we e all collec ed in exac ly he same way. The pa icipan s comple ed
a ask and a demog aphic ques ionnai e du ing no mal class ime and ook
on a e age 5–10 minu es o comple e i . They we e no in o med o he aim
o he esea ch un il he su ey had been comple ed by all he s uden s.
The demog aphic ques ionnai e included a se ies o ques ions ha aimed
o ga he in o ma ion abou he pa icipan s’ age, gende , academic
backg ound and na ionali y. The ask con ained en u e ances ha we e
p esen ed o hem. As men ioned abo e, eigh we e eal sen ences ex ac ed
om CAES and CEDEL2 (see 4.1.), whe eas he o he wo we e dis ac o s
ha we e c ea ed ad hoc wi h se e al wo ds con aining lexical ambigui y:
solo, which means ‘only’ and ‘alone’, and banco, which can be bo h ‘bench’
and ‘bank’. The sen ences we e Es a 
e abajando solo po la ma~
nana [I will
be wo king only/alone in he mo ning] and Luis y Ma a se ci a on en el
banco de la esquina [Luis and Ma a me a he bench/bank on he co ne ].
Table 1. False cogna es e ie ed om he co po a.
English o
Po uguese Spanish
To al/
pa ial Meaning
Suppo (EN) Sopo a To al ‘Bea ’
A end (EN) A ende Pa ial ‘Wa ch o e ’
Toda ia (PT) Toda 
ıa To al ‘Howe e ’ (PT) ‘s ill/ye ’ (SP)
Aco da (se) (PT) Aco da (se) Pa ial ‘Wake up’ (PT), ‘ emembe ’
(PT, SP)
Lexical ambigui y caused by alse cogna es 9
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License
p oduced by non-na i e speake s. Howe e , in some specific cases hese
diffe ences be ween he wo g oups a e no s a is ically significan (e.g.
sopo a , oda 
ıa and he dis ac o i ems).
In e es ingly, dis ac o s ob ain he bes sco es in he c i ical g oup,
and a e highe han he a e age (2.32/3 s. 1.87/3), while in he con ol
g oup hey a e sligh ly abo e he a e age (2.28/3, 2.36/3 s. 2.25/3).
These esul s show ha he pa icipan s in he c i ical g oup a e s ongly
influenced by knowing ha he speake s a e non-na i e, in ha hey
ound mo e ins ances o lexical ambigui y and awa ded a lowe sco e o
co ec ness o he sen ences p oduced by non-na i e speake s han hey
did o he dis ac o s. In his con ex , hey ound he dis ac o s mo e
co ec , confi ming he p e ious hypo hesis.
Taking in o accoun he diffe ences be ween u e ances a. and b. se
ou in 4.1., Figu es 1and 2show he specific dis ibu ion o sco es o
each sen ence p o ided by bo h he c i ical and he con ol g oups.
These figu es show ha mos o he sco es awa ded by he con ol
g oup a e be ween 2 and 3, whe eas he esul s om he c i ical g oup a e
be ween 1 and 2. The e does no seem o be a majo diffe ence in he
con ol g oup when assessing sen ences a. and b. Indeed, he only majo
diffe ence is in oda 
ıa a. s. b. The o he i ems do no show any
endency, which means ha al hough he u e ances o b. seemed o be
unusual and a con adic ion o logical hinking, hese do no affec he
co ec ness and he sco e awa ded. Howe e , his is no he case o he
c i ical g oup: all he unusual sen ences a e less co ec o hese
pa icipan s, as can be obse ed in Figu e 1. The mos e iden case is
a ende , o which he fi s u e ance was sco ed be ween 2 and 3 by 64%
o pa icipan s, while he second one only ecei ed 24%. Meanwhile,
aco da has mo e simila esul s, wi h 56% o u e ance a. s. 52% o
u e ance b.
0
20
40
60
80
100
A ende a A ende b Sopo a a So po a b Toda ía a Toda ía b Aco da a Aco da b
0 1 2 3
Figu e 1. Sco es awa ded by he c i ical g oup.
16 Ma 
ıa Samped o Mella
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License

In o e all e ms, he mos co ec sen ences o he con ol g oup a e
hose ha con ain he wo ds aco da and sopo a . The sen ences wi h
sopo a we e also sco ed highes by he c i ical g oup. Fo hese la e
pa icipan s, he leas co ec u e ances we e su p isingly hose wi h he
e b aco da . While hese da a do no seem e y imp essi e, hey do show
ha on a e age, speake s do no use he same c i e ia when assessing he
same u e ances: he pa icipan s seem o be nega i ely influenced when
hey know ha sen ences a e p oduced by non-na i e speake s. In hose
cases, hey a e mo e c i ical owa ds hese speake s, especially i he
u e ances ha hey assess do no seem o ollow a logic.
6. Conclusion
This explo a o y s udy aimed o shed new ligh on he specific aspec s o
alse cogna es. F om a b oade pe spec i e, i has also analyzed whe he
he e a e any diffe ences be ween he way ha na i e speake s in e p e a
ype o Spanish u e ance when hey know o do no know ha hey ha e
been p oduced by non-na i e speake s. The esul s o a compa a i e
s udy based on da a p o ided by CAES and CEDEL2 co po a show ha
his in o ma ion significan ly affec s he in e locu o ’s pe cep ion o he
message. In his ega d, he pa icipan s p o ed o be mo e c i ical when
hey knew ha he u e ances we e p oduced by a non-na i e speake ,
and especially i hese u e ances we e somewha unusual o illogical.
Fu he mo e, hey we e mo e inclined o find lexical ambigui y in hose
sen ences han pa icipan s who did no ha e his in o ma ion. These
diffe en in e p e a ions could ha e been due o he o hog aphical,
g amma ical, communica i e o p agma ic e o s o he in e language,
bu hese sen ences we e co ec ed be o ehand in o de o a oid esponse
biases o influencing he pa icipan s. The esul s he e o e show ha in
0
20
40
60
80
100
A ende a A ende b Sopo a a So po a b Toda ía a Toda ía b Aco da a Aco da b
0 1 2 3
Figu e 2. Sco es awa ded by he con ol g oup.
Lexical ambigui y caused by alse cogna es 17
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License
o de o assess u e ances, pa icipan s pay mo e a en ion o he
speake s’ p ofiles han o he u e ances a he discou se le el, e en i he
sen ences do no con ain any e o s. The in e language consequen ly
seems o be subjec o a mo e nega i e pe cep ion han he L1.
Despi e he con ibu ions o he p esen s udy, some limi a ions need
o be acknowledged. Fi s , his s udy is based on semi-na u al da a
p o ided by lea ne s’ co po a, which did no include many u e ances
con aining a alse cogna e ha had mo e han one meaning. Fu he -
mo e, examples highligh ing lexical ambigui y we e limi ed and no e y
a ied. I is he e o e likely ha diffe en in e p e a ions could be
p o ided i he s udy had been based on mo e si ua ions c ea ed ad hoc
wi h mo e di e se u e ances. Fo hese easons, i migh be use ul o
eplica e he findings o his s udy wi h con olled da a and a la ge
quan i y o ealis ic si ua ions.
In sho , his explo a o y s udy is a fi s s ep owa ds going beyond he
ocus on he alse cogna es om a heo e ical pe spec i e o based on he
difficul ies ha non-na i e lea ne s encoun e when lea ning he
dis inc ion be ween he meaning in he L1 and in he Spanish L2.
Ne e heless, mo e sou ces o da a will be necessa y in o de o p o ide a
ull pic u e o he lexical ambigui y in L2 and na i e speake s’ pe cep ion
o he in e language.
Acknowledgemen
This s udy was unded h ough a g an om Xun a de Galicia (Spain)
i led “Use o gene al and lea ne co po a o he esea ch and eaching
o discou se in Spanish as a o eign language” [g an code: ED481D–
2022–016].
Re e ences
ALBA QUI ~
NONES, V. 2009. El an
alisis de e o es en el campo del espa~
nol como
lengua ex anje a. Algunas cues iones me odol
ogicas. Re is a Neb ija de
Ling€
u
ıs ica aplicada a la ense~
nanza de las lenguas 5:1–16.
BELDA TORRIJOS, M. 2015. E o es del uso del espa~
nol en es udian es ex anje os:
c eaci
on de una axonom
ıa y su e ique ado. PhD diss.
CAES. Co pus de Ap endices de Espa~
nol. Uni e sidade de San iago de
Compos ela & Ins i u o Ce an es [on-line]: h ps://gal an.usc.es/caes/
CEDEL2. Co pus Esc i o del Espa~
nol-L2. Uni e sidad de G anada [on-line]:
h p://cedel2.lea ne co po a.com/
CORDER, S. P. 1967. The Significance o Lea ne s’ E o s. In e na ional Re iew o
Applied Linguis ics in Language Teaching 5:161–170.
COUNCIL OF EUROPE, 2001. Common Eu opean F amewo k o Re e ence o
Languages: Lea ning, Teaching, Assessmen . Camb idge: Camb idge Uni e si y
P ess.
FERN
ANDEZ SOLERA, S. 2002. Fac o es sin 
ac icos y sem
an icos en el p ocesamien o
del lenguaje. PhD diss.
18 Ma 
ıa Samped o Mella
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License
HAMILTON, R. P. 2001. The insignificance o lea ne s’ e o s: a philosophical
in es iga ion o he in e language hypo hesis. Language and communica ion
21:73–88.
HORNO,M.&L

OPEZ, N. 2020. Polisemia y homonimia en la ense~
nanza no
uni e si a ia: e isi
on y p opues a did
ac ica. REGROC: Re is a de g am
a ica
o ien ada a las compe encias 3:37–54.
LARSEN-FREEMAN, D. 2011. Techniques and P inciples in Language Teaching.
Ox o d: Ox o d Uni e si y P ess.
LOZANO, C. 2022. CEDEL2: Design, compila ion and web in e ace o an online
co pus o L2 Spanish acquisi ion esea ch. Second Language Resea ch
38.4:965–983.
NAVARRO GALA, R. 1999. An
alisis de e o es en el ma co de la ling€
u
ıs ica
con as i a: la exp esi
on esc i a. X Cong eso In e nacional de la ASELE: Nue as
Pe spec i as en la Ense~
nanza del Espa~
nol como Lengua Ex anje a. eds. M.
F anco, C. Sole , Ja ie de Cos e al., 481–490. C
adiz: Uni e sidad de C
adiz.
PENAD
ES MART
INEZ, I. 2003. Las clasificaciones de e o es ling€
u
ıs icos en el ma co
del an
alisis de e o es. Lin ed: Ling€
u
ıs ica en la Red 1: h ps://ebuah.uah.es/
dspace/handle/10017/24703.
RAMOS SA~
NUDO, A. M. 2021. Los ma cado es discu si os de cie e en el espa~
nol
como lengua ex anje a. Apo es del CAES a la ense~
nanza del espa~
nol como
lengua ex anje a: es udios aplicados de g am
a ica y p agm
a ica. eds. C.
Balles e o de Celis & M. Samped o Mella, 257-285. Ve ba: Anua io galego de
ilolox
ıa 81. Uni e sidade de San iago de Compos ela.
ROCA VARELA, M. L. 2010. The p oblem o alse iends in lea ne language:
E idence om wo lea ne co po a. Language Windowing h ough Co po a. eds.
I. Moskowich, B. C espo, I. La eo Ma 
ın & P. Lojo Sandino. Co u~
na:
Uni e sidade da Co u~
na.
ROJO,G.&PALACIOS, I. 2016. Lea ne Spanish on Compu e . The CAES ‘Co pus
de Ap endices de Espa~
nol’ P ojec . In M. Alonso-Ramos (ed.), Spanish Lea ne
Co pus Resea ch. Cu en T ends and Fu u e Pe spec i es. Ams e dam: John
Benjamins, 55–87.
SAMPEDRO MELLA,M.&S

ANCHEZ GUTI
ERREZ, C. H. 2017. E ec o de la ense~
nanza
ex ensi a en la memo izaci
on del l
exico: un es udio de in es igaci
on-acci
on.
Re is a Sin agma 29:61–78.
SAMPEDRO MELLA, M. 2021a. La exp esi
on del o ecimien o y la pe ici
on en
ap endices de espa~
nol de L1 po ugu
es. C
ı culo de ling€
u
ıs ica aplicada a la
comunicaci
on 85:113–125.
SAMPEDRO MELLA, M. 2021b. Es imado S . s. Hola ho el:elan

alisis con as i o
de la in e lengua pa a la ense~
nanza de la a iaci
on p agm
a ico-discu si a. La
a iaci
on en espa~
nol y su ense~
nanza.Re lexiones y p opues as did
ac icas. eds. V.
Ma ce , V. 
Al a ez Rosa & M. Ne o , 134–150. Salamanca: Uni e sidad de
Salamanca.
SHAFFER, D.E. 2005. Classi ying Language Lea ning E o s. The In e ne TEFL
Jou nal 58:1–7.
SELINKER, L. 1972. In e language. In e na ional Re iew o Applied Linguis ics
10:209–241.
STAMENOV, M. 2009. Cogna es in language, in he mind and in a p omp ing
dic iona y o ansla ion. Cogna es in language, in he mind and in a p omp ing
dic iona y o ansla ion. eds. S. G€
op e ich, A. Lykke Jakobsen, I. M. Mees,
219–252. Denma k: Sam undsli e a u .
THOMAS, J. 1983. C oss-cul u al p agma ic ailu e. Applied Linguis ics 4:91–112.
ULLMANN S. 1976. Sem
an ica. In oducci
on a la ciencia del signi icado. Mad id:
Aguila .
Lexical ambigui y caused by alse cogna es 19
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License
VALVERDE MATEOS, A. 2012. An
alisis de e o es de ap endien es de anc
es lengua
ex anje a (FLE) basado en co pus o ales. PhD diss.
V
AZQUEZ, G. E. 1991. An
alisis de e o es y ap endizaje de espa~
nol / lengua
ex anje a. F ank u am Main: Pe e Lang.
WANG, D. 2021. An Analysis o Tole ance o Ambigui y in Second Language
Lea ning. Sino-US English Teaching 18.9:250–256.
Recei ed Sep embe 13, 2023
Accep ed Sep embe 28, 2023
Ma 
ıa Samped o Mella
Facul ad de Filolog
ıa
A da. Cas elao, s/n,
15705, San iago de Compos ela
Spain
[email p o ec ed]
20 Ma 
ıa Samped o Mella
©2023 The Au ho s. S udia Linguis ica published by John Wiley & Sons L d on behal o Edi o ial Boa d o S udia
Linguis ica.
14679582, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1111/s ul.12225 by Uni e sidade de San iago de Compos ela, Wiley Online Lib a y on [22/02/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License