COMBINING GEOLINGUISTIC SOURCES IN DIALECT … · (1) verb stem + -ndo: cantando bebendo dormindo...
Transcript of COMBINING GEOLINGUISTIC SOURCES IN DIALECT … · (1) verb stem + -ndo: cantando bebendo dormindo...
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
141
Received16May2015.
Accepted17July2015.
COMBININGGEOLINGUISTICSOURCESINDIALECTSYNTAX:
THREECASE-STUDIESTHROUGHALPIANDCORDIAL-SIN
MariaLOBO1&ErnestinaCARRILHO2
CLUNL/FCSH-UniversidadeNovadeLisboa1/CLUL/FLUL-UniversidadedeLisboa2
[email protected]/[email protected]
Abstract
Although geolinguistic projects do not usually provide adequate (if any) coverage of syntactic
variation, current studies of dialect syntax still may benefit from data that were made available in
traditionallinguisticatlases.Concomitantly,dialectcorpora,asadifferentkindoflinguisticresourcefor
the studyofnonstandard syntax,oftengrantaccess to spontaneousoraldata thatallow the studyof
phenomenathatarehardtoelucidatethroughtraditionalquestionnaires.
InthispaperwepresentanddiscussthreemorphosyntacticphenomenainEuropeanPortuguese
varieties,as theyappear throughbothkindsof sources–morespecifically, theAtlasLingüísticode la
Península Ibérica (ALPI),a traditional linguisticatlas,andCORDIAL-SIN,adialectcorpus.Throughboth
sources,wewillinvestigatethegeolinguisticsyntacticvariationconcerning:(i)thedistributionofgerund
forms; (ii) the presence of articles before prenominal possessives; and (iii) the distribution of verbal
formsendingin-ra.
The contrast and complementarity of both kinds of sources, namely, their advantages and
inconveniences, inviteustodiscussnewmethodsthat integratecomplementarymethodsincollecting
datafordialectsyntax.
Keywords
dialectsyntax,EuropeanPortuguese,dialectcorpus,linguisticatlas,methodology
©Universitat de Barcelona
M.LOBO&E.CARRILHO
142
SOBREACOMBINAÇÃODEFONTESGEOLINGUÍSTICASEMSINTAXEDIALECTAL:ESTUDODETRÊS
CASOSAPARTIRDOALPIEDOCORDIAL-SIN
Resumo
Embora os projetos geolinguísticos não forneçam habitualmente uma cobertura adequada da
variação sintática, ainda assim os estudos recentes de sintaxe dialetal podem beneficiar de dados
disponibilizadosnosatlas linguísticos tradicionais.Concomitantemente,oscorporadialetais,umoutro
tipo de recurso para o estudo da sintaxe não padrão, dão geralmente acesso a dados orais que
permitemoestudodefenómenosdifíceisdeinvestigaratravésdequestionáriostradicionais.
Neste artigo, apresentamos e discutimos três fenómenos morfossintáticos em variedades do
portuguêseuropeuatravésdestesdoistiposdefontes–oAtlasLingüísticodelaPenínsulaIbérica(ALPI),
um atlas linguístico tradicional, e o CORDIAL-SIN, um corpus dialetal. Através destas duas fontes,
investigaremos a variação sintática geolinguística relativamente a: (i) distribuição de formas de
gerúndio; (ii) presença de artigos antes de possessivos pré-nominais; e (iii) distribuição de formas
verbaisterminadasem-ra.
Ocontrasteecomplementaridadeentreestasduasfontes,nomeadamente,assuasvantagense
inconvenientes, levam-nosadestacara importânciadacombinaçãodemetodologiascomplementares
paraarecolhadedadosparaoestudodasintaxedialetal.
Palavras-chave
sintaxedialectal,portuguêseuropeu,corpusdialetal,atlaslinguístico,metodologia
1.Introduction
Althoughgeolinguisticprojectsdonotusuallyprovideadequate(ifany)coverage
ofsyntacticvariation,currentstudiesofdialectsyntaxstillmaybenefitfromdatathat
weremadeavailableintraditionallinguisticatlases.Concomitantly,asadifferentkind
oflinguisticresourceforthestudyofnon-standardsyntax,dialectcorporaoftengrant
accesstospontaneousoraldatathatallowthestudyofphenomenathatarehardto
elucidatethroughtraditionalquestionnaires.
In this paper we present and discuss three morphosyntactic phenomena in
EuropeanPortuguesevarieties,astheyappearthroughtwokindsofsources—more
specifically, theAtlas Lingüísticode laPenínsula Ibérica (ALPI), a traditional linguistic
atlas, and CORDIAL-SIN, a dialect corpus. Through both sources, we investigate the
geolinguisticsyntacticvariationconcerning:(i)thedistributionofgerundforms;(ii)the
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
143
presenceofarticlesbeforeprenominalpossessives;and(iii)thedistributionofverbal
formsendingin-ra.
By contrasting both kinds of sources, and discussing their advantages and
inconveniences, we intend to evaluate the relevance of complementarymethods in
collectingdatafordialectsyntax.
Thepaperisorganizedasfollows:insection2,wecharacterizethesourcesand
themethodologyforobtainingthedataforthisstudy;insection3,wediscussdialectal
variationinthesyntacticpropertiesofthreemorphosyntacticphenomena,comparing
dataobtainedfromALPIandfromCORDIAL-SIN—variationinthesyntacticproperties
of gerunds (§3.1), variation in the distribution of determiners before possessives
(§3.2),variationinthedistributionofverbalformsendingin-ra(§3.3).
2.Methodology:twokindsofgeolinguisticsources
Asmentionedabove,inthisstudyweconsidertwosources:theAtlasLingüístico
de la Península Ibérica (ALPI), a traditional geolinguistic project that includes a
morphosyntactic questionnaire besides phonetic and lexical questionnaires, and
CORDIAL-SIN,adialectcorpus.
Forthefirstsource,weobservedtheanswersgiventoelevenquestionsrelated
to the threemorphosyntacticphenomenawewanted to investigate,as registered in
the fieldwork questionnaires.1 We considered only the Portuguese network,
comprising 93 localities distributed alongmainland Portugal. The enquiries for ALPI
wereconductedbetween1931andthe1950s(GarcíaMoutonetal.2012:2).Wehave
alsotakenintoaccountpreviousworkbyMouzo(2008)on-raformsusingALPIdata.
CORDIAL-SIN is a dialect corpus that includes transcribed samples of
spontaneousorsemi-directedspeechofruralspeakersfrom42localities,includingnot
onlymainlandPortugal,butalsotheislandsofAzoresandMadeira.Thetranscriptions
1ThedataweregentlymadeavailablebythecoordinatoroftheprojectALPI–Elaboraciónyedicióndelos materiales del Atlas Lingüístico de la Península Ibérica (CSIC), Pilar García Mouton, and by thePortuguesecoordinator,JoãoSaramago.
©Universitat de Barcelona
M.LOBO&E.CARRILHO
144
totalize approximately 600,000 words. The recordings were selected from enquiries
conducted for geolinguistic projects between 1974 and 2004. We have taken into
account data obtained for gerunds and for possessives without determiners in two
previousstudiesthatusedCORDIAL-SINasasource(Lobo2008andCarrilho&Pereira
2011). For the verbal forms in -ra, the datawere obtained through a Concordancer
programthatrunsonthePOSannotatedfiles(Concordance3.3,July2009©R.J.C.Watt
1999-2009).
Bycomparingthedataobtainedforthreedifferentmorphosyntacticphenomena
inthesetwodifferentsourcesweintendtoevaluatetherelevanceofeachmethodand
theircomplementarityforthestudyofdialectsyntax.Inthenextsection,weconsider
eachofthephenomena.
3.Threemorphosyntacticphenomena
3.1.Gerunds
3.1.1.Variationingerunds
Portuguesegerundsarenon-finiteverbalformsinwhichthesuffix-ndoattaches
totheverbstem(rootfollowedbythemevowel):
(1)verbstem+-ndo: cantando bebendo dormindo
singing drinking sleeping
There is dialectal variation in the use of gerunds, concerning their syntactic
distribution, inflectional morphology, types of connectives that introduce gerund
clausesandwordorderingerundclauses.
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
145
3.1.1.1.Syntacticdistribution
In the standard variety, gerunds may be found mainly in adverbial adjunct
clauses,witha temporal,conditional,concessive,mannerorcausalmeaning (2),and
withsomeaspectualauxiliaries(3):
(2) Estando achover, ficaremos emcasa.
being torain, stay.FUT.1Pl athome
“Ifitisraining,wewillstayhome.”
(3) Vou sobrevivendo.
go.1Sg surviving
“Igoonsurviving.”
In non-standard varieties of European Portuguese and in the Brazilian variety,
gerundscanalsooccurintheprogressive,withtheauxiliaryestar‘tobe’(4),withother
aspectual auxiliary verbs (5), as complements to perceptive verbs (6), as secondary
predicates(7),and incoordinatedsmallclauses(8). Inall thesecases,wewouldfind
thevariantwitha(to)+infinitiveinthestandardvariety:
(4) Estou escrevendo umlivro.
am.1Sg writing abook
“Iamwritingabook.”
(5) Ocão começou coxeando.
thedog started limping
“Thedogstartedtolimp.”
(6) Orapaz ouviu ocão ladrando.
theboy heard thedog barking
“Theboyheardthedogbarking.”
(7) Metia-se aágua fervendo namurta.
put-SE thewater boiling in.themyrtle
“Onepouredthewaterboilingonthemyrtle.”
©Universitat de Barcelona
M.LOBO&E.CARRILHO
146
(8) Ele cantando e eu dançando.
he singing and I dancing
3.1.1.2.Inflectionalmorphology
In the standard variety of European Portuguese, gerunds are morphologically
invariable. In somenon-standardvarieties,however,gerundsmay inflect forperson,
displayingagreementmorphologywiththesubject(Mota1997;Lobo2008,i.a.).Overt
agreementmorphemesonlyshowupforpluralandsecondsingularpersons,similarly
toinflectedinfinitives.
(9) Em eles tendem aoutra, já não querem aquela(LVR05)2
in they having.3Pl theother yet not want.3Pl that
“Iftheyhavetheotherone,theydonotwantthatone.”
(10) E tendem umaárvore, não há pássaro nenhum que poise
and having.3Pl a tree, not is bird no.one that land
no chão.(STJ30)
on.the floor
“Andiftheyhaveatree,thereisnobirdthatwilllandonthefloor.”
3.1.1.3.Connectives
In the standardvariety,adverbialgerundclausesareusuallynot introducedby
anyconnective,andmaymarginallybe introducedbytheconnectiveem ‘in’. Innon-
standardvarieties,onemay findgerundclauses introducedbydifferentconnectives,
includingconnectives that typically introduce finitedomains, suchasquando ‘when’,
se‘if’(Lobo2001,2008):
2 The indicated sources for the CORDIAL-SIN examples are available at:http://www.clul.ul.pt/en/resources/412-cordial-corpus-verbatim-transcription.
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
147
(11) Mas agente, em tendo saúde, trabalho nãomata
but thepeople in having health work not kill
ninguém.(AAL35)
nobody
“Butifoneishealthy,workdoesnotkillanybody.”
(12) Tomasoutranovavidaquandoumanosepassando.(MLD35)
take.2SgothernewlifewhenayearSEpassing
“Yougainanewlifewhenayeargoesby.”
3.1.1.4.Wordorder
Finally, in the standard variety of European Portuguese, there is obligatory
subject-verbinversioningerundclauses.Innon-standardvarieties,incontrast,wemay
find either subject-verb or verb-subjectwordorders (Lobo 2003, 2008; Fiéis& Lobo
2011):
(13) elestendemascoisasemcasa,fazematodaahora(AAL36)
theyhaving.3Plthethingsathome,do.3Platallthetime
“Iftheyhavetheirstuffathome,theydoitallthetime”
(14) Agentepodendo-sedefender,nãohánadamelhor.(MLD03)
thepeoplemight.GER-SEdefendnotexistsnothingbetter
“Ifwecandefendourselves,thereisnothingbetter.”
(15) Tomasoutranovavidaquandoumanosepassando.(MLD35)
take.2SgothernewlifewhenayearSEpassing
“Yougainanewlifewhenayeargoesby.”
3.1.2.Comparingsources
3.1.2.1.DistributionofgerundsinCORDIAL-SIN
Inthedialectcorpustheoccurrencesofinflectedgerunds(16)areonlyfoundin
onelocalityoftheAzoresandinthesouthernvarieties,asshowninMap1.Thecorpus
©Universitat de Barcelona
M.LOBO&E.CARRILHO
148
allowsustoconcludethatinflectedgerundsaremainlyfoundinadjunctclausesandin
parentheticalclauses,and(almost)absentfromverbaldomainsselectedbyauxiliaries
(Lobo2008).
(16) … – estândomos agente afalar–…(CBV41)
being.1Pl thepeople totalk
“whileweweretalking”
(17) Digo assim:“ Pronto! Em tu querendos”!(CBV41)
say.1Sg so rightin you wanting.2Sg
“Isaythen:Right!Ifyouwant!”
Map1.Distributionofinflectedgerunds(adaptedfromLobo2008)
Gerundsinprogressivestructures,withtheauxiliaryverbestar‘be’arefoundin
southern varieties, in the islands (Azores and Madeira) and in one border location
closetoGalicia,asshowninMap2(adaptedfromCarrilho&Pereira2011).
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
149
Map2.Distributionofgerundsinprogressivestructures
Whileinthesevarietieswefindthegerund(18a)and(19a)inthesecontexts,in
the standard and northern varieties, we would find a ‘to’ + infinitive in the same
context,asshownin(18b)and(19b):
(18) a. Estãoardendo.(PAL36)
are burning
b. Estãoaarder.
are toburn
“Theyareburning.”
(19) a. Essa pessoa estava varrendo.(LUZ30)
that person was sweeping
b. Essa pessoa estava avarrer.
that person was tosweep
“Thatpersonwassweeping.”
3.1.2.2DistributionofconnectivesingerundclausesinCORDIAL-SIN
In the dialect corpus, gerund clausesmay be introduced by different types of
connectives.Themostfrequentisem‘in’,whichcanbefoundacrosstheterritory,but
©Universitat de Barcelona
M.LOBO&E.CARRILHO
150
we can also find gerund clauses introduced by connectives that are typical of finite
domains, suchasquando ‘when’, se ‘if’,caso ‘if’ andother connectives that arenot
foundinthestandardvariety,suchasembem‘inwell’.
(20) Embemlhepartindemacasa,elasmorrem.(CBV72)
inwellhimbraking.3Plthehouse,theydie.3Pl
“Whenyoubreaktheirhouse,theydie.”
(21) Borregos é pequenos e quando sendem aí mais
lambs is small and when being.3Pl theremore
dum ano(...) já é carneiros.(MLD52)
than.a year already is sheep
“(Wecall them) lambswhen theyare small, andwhen theyareolder thana
yearit’salreadysheep.”
(22) Isto osarrendamentos, já se sabe, quenãohá
this therentings, alreadySEknows, thatnothas
comoascoisas estandem nasmãos dosdonos
as thethings being.3Pl inthehands oftheowners
“Therentings,youknow,thereisnosuchthingastheybeinginthehandsof
theirowners.”
Thefrequencyofeachconnectiveinthelocalitieswithinflectedgerundsisgiven
inTable1:
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
151
ConnectiveLocalitieswiththe
inflectedgerund
Localitieswithoutthe
inflectedgerund
em‘in’66/235
28%101/274
36%
quando‘when’47/235
20%1/2740,4%
embem‘inwell’7/235
3%0
bem‘well’4/2351,7%
1/2740,4%
caso‘if’1/2350,4%
0
se‘if’ 02/2740,7%
onde/donde‘where’1/2350,4%
1/2740,4%
como‘how’1/2350,4%
0
Table1.Frequencyofconnectivesintroducingadverbialgerundclausesinthe10localitieswithinflected
gerundsandinthe31localitieswithoutinflectedgerunds
Map 3 illustrates the distribution of each connective of gerund clauses across
Portugal.
Map3.Distributionofconnectivesofgerundclauses
©Universitat de Barcelona
M.LOBO&E.CARRILHO
152
Ascanbeseen,whilegerundclauseswithemarecommonacrossthecountry,
gerund clauses with quando and em bem are found in the geographic area where
inflectedgerundscanbefound.Overalladverbialgerundclausesaremorefrequentin
the southern and insular varieties,where inflected gerunds can be found (there are
274gerundadverbialclausesinthe31localitieswithoutinflectedgerunds,whilethere
are235adverbialgerundclausesinthe10localitieswiththeinflectedgerund).
Whenweconsiderwordorderinadverbialgerundclauses,theCORDIAL-SINdata
allowustoseethatthepresenceoftheinflectiondoesnothaveacleareffectonword
order.InPortuguesedialects,bothSubject-VerbandVerb-Subjectwordorderscanbe
found,unlikewhathappensinthestandardvariety(Lobo&Fiéis2015).
Localitieswith
inflectedgerunds
Localitieswithout
inflectedgerunds
Su-V4264%
3560%
V-Su2436%
2340%
Total 66 58
Table2.Proportionofsubject-verbandverb-subjectwordordersinadverbialgerundclauses
3.1.2.3.GerundsintheALPI
The morphosyntactic questionnaire of ALPI does not include questions that
favourtheoccurrenceof inflectedgerunds.Thethreequestionsthat includegerunds
(Table 3) are all third singular person contexts and they do not include peripheral
adjunct clauses, the oneswhere inflected gerunds aremore common. Furthermore,
only Q. 352 presents a context subject to variation between the gerund and a +
infinitive. Therefore, the relevance of these questions for dialect variation in
Portuguese is rather limited, since the crucial linguistic contexts have not been
includedinthequestionnaire.
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
153
Q.Nb Page Question
326 22 Todosealcanzateniendopaciencia
Tudosealcançatendopaciência
“Everythingisreachedhavingpatience”
327 22 Sepuedevivirqueriendotrabajar
Pode-seviverquerendotrabalhar
“Onecanlivewantingto.work”
352 25 Alpadrelevieronllorando
Opai,viram-noachorar
“Thefather,(they)saw-himcrying”
Table3.QuestionsincludinggerundsintheALPI
Nevertheless, question 352 allows us to conclude that the gerund variant in
complements to perceptive verbs seems to be geographically more limited than in
progressivecontexts(Map4,vs.Map2above).
Questions 326 and 327 allow us to map gerund clauses introduced by the
connectiveem ‘in’ (Map5). Comparing these resultswithMap3 above,we can see
thatprobablythekindofmannergerundclauseconsideredintheALPIquestionnaireis
notthepreferentialcontextforthiskindofconnective.Thegeographicdistributionof
em+gerundclauses ismuchmore limited in theALPIdata.Gerundclauseswithem
occur only in five locations in these data: Montalegre (Vila Real); Reguengo
(Portalegre);Povos(Lisboa);FerreiradoAlentejo(Beja);MontedosSapos(Beja).
©Universitat de Barcelona
M.LOBO&E.CARRILHO
154
Map 4. Distribution of the variants gerundand a + infinitive in complements toperceptiveverbs,basedonQ352fromALPI(fromCarrilho&Lobo2012)
Map 5. Gerund clauses withconnectiveem‘in’,basedonQ326and327fromALPI
3.1.2.4.Summary
MorphosyntacticvariationingerundsinPortugueseisattestedinbothsources.
However, the corpus and the linguistic atlas provide different and, to some extent,
complementarypicturesconcerningthegeographicalandlinguisticdistributionofthe
considered variants. While CORDIAL-SIN data lack certain contextual manifestations
(namely, gerunds in complements to perceptive verbs, which by contrast are
considered in the ALPI questionnaire), the geographical distribution of the relevant
phenomenaseemstobelessrestrainedinthecorpus(eventhoughtheCORDIAL-SIN
networkislessextendedthantheALPInetwork).Ontheotherhand,theALPIdatado
not include important contexts for studying variation in gerunds, namely the
combinationwithauxiliariesandgerundsinadjunctclauses.
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
155
3.2.Determinerbeforepossessives
3.2.1.Variationintheoccurrenceofarticlesbeforeprenominalpossessives
One of the phenomena that is subject to intralinguistic and crosslinguistic
variation is thepresenceofadeterminerbeforeapossessive (23avs.23b-e) (cf. i.a.
MattoseSilva1989;Miguel2002;Rinke2010).InthehistoryofPortuguesetherewas
agradualchangetowardsahigherfrequencyofdeterminersbeforepossessives.Inthe
Brazilian Portuguese variety, the absence of the determiner is nowadays the most
frequent variant (23b), although, there is still variation, mostly dependent on
geographicandstylisticvariables(Callou2008).3InModernEuropeanPortuguese,the
standardvariantcorrespondstothearticlebeforethepossessive(23a).
(23) a. os meus livros EP
themy books
b. meus livros BP
my books
c. mislibros Spanish
d.meslivres French
e.mybooks English
We can still find variation in Modern EP, however, in contexts with kinship
nouns. There is thus lexical conditioning in non-standard varieties of EP for the
absence of the determiner before possessives. Variation between the presence or
absence of the determiner before possessives is displayed only with this semantic
subclassofnouns:
3AccordingtoCallou(2008),thearticleisstillpreservedinthesouthernregionsofSãoPauloandPortoAlegre(withahigherfrequencyinwrittenregisters),ismostlyabsentinthenorth-eastregionsofRecifeandSãoSalvador,andRiodeJaneiroisinanintermediateposition.
©Universitat de Barcelona
M.LOBO&E.CARRILHO
156
(24) a. o meu pai
themy father
b. meu pai
my father
3.2.2.Comparingsources
3.2.2.1.AbsenceofthedeterminerbeforepossessivesinCORDIAL-SIN
A search in the dialectal corpus shows that the articlemay be absent in non-
standard varieties before possessives that precede kinship nouns, as shown in (25)-
(28):
(25) E minha tia lia aqueles livros.(GRJ06)
and my aunt read those books
“Andmyauntreadthosebooks.”
(26) Olha, fala com teu avô[…](CTL18)
look talk with your grandfather
“Look,talktoyourgrandfather”
(27) Olhe, só tive minhasirmãs à noite que me ajudaram
look,only had my sisters atnightthat me helped
a lavar as tripase não tive mais ninguém.(PIC29)
towashthe guts andnothad morenoone
“Look,IonlyhadmysistersthathelpedmeatnightwashingthegutsandIhadnoone
else.”
(28) Masmeu pai tinha era gado,fazendas[…].(PST25)
but my father had wascattle,farms
“Butwhatmyfatherhadwascattleandfarms”
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
157
Map 6. Distribution of the absence ofdeterminer before possessive with kinshipnouns
Map 7. Frequency of the absence ofdeterminer before possessive with kinshipnouns
Although the absence of determiner can be found across Portugal (both
mainlandandislands),whenwelookatthefrequencyofthenon-standardvariant,we
observethatithasmuchhigherratesintheislands,particularlyinMadeira(asshown
inMaps6and7,fromCarrilho&Pereira2011).
3.2.2.2.AbsenceofthedeterminerbeforepossessivesinALPI
The morphosyntactic questionnaire of ALPI includes four questions that have
preverbalpossessives(Table4).However,onlythreeofthemhavekinshipnouns.The
datafromquestionnumber261,theonlyonethatdoesnothaveakinshipnoun,have
beenstudiedandmappedbyFernández-Ordóñez(2011).
0 < 20 % 20 < 45 % 45 < 65 % > 65 %
©Universitat de Barcelona
M.LOBO&E.CARRILHO
158
Q.Nb PageQuestion
260 17 Miscuñadosymisprimos
Osmeuscunhadoseosmeusprimos
“Mybrothers-in-lawandmycousins”
261 18 Suscorderosestánennuestroprado
Osseuscordeirosestãononossocampo
“Hislambsareinourfield”
394 29 Siembraparaquecomasufamilia
Semeiaparaasuafamíliacomer
“Hesowsforhisfamilytoeat”
403 29 Abandonaranasutío
Abandonaramoseutio
“Theyabandonedtheiruncle”
Table4.QuestionsincludingpreverbalpossessivesintheALPI
As can be seen inMap 8, the area corresponding to possessives preceded by
determinerscoversthewholePortugueseterritory.
Map8.Determinerfollowedbypossessive,basedonQ261fromALPI(fromFernández-Ordóñez2011:52,map28)
Determiner followed by possessive
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
159
For the other three questions, we can find variation in the presence of the
determiner,ascanbeseeninMap9.
Map9.Absenceofdeterminerbeforepossessive,basedonQs260,394,403fromALPI
Although ALPI does not include data for the islands, the locations where the
absenceofthedeterminerwasfoundinthecontinentalterritoryshowafairlysimilar
distributionasthelocalitieswherethesamephenomenonwasfoundintheCORDIAL-
SINdata–westernborder,Minhoandcentralcoast(comparewithMaps6-7).
3.3.Verbalformsendingin-ra
3.3.1.Variationinthedistributionofverbalformsendingin-ra
Verbalformsendingin-ramayhavedifferentvalues.Theymayrefertoanevent
thatoccurredinthepastpriortoapastevent,correspondingtothepluperfect(mais-
que-perfeito).Thisusageisavailableinthestandardvarietyandcanalternatewiththe
verbal complex with the auxiliary ter ‘have’ in the imperfect followed by the past
participleofthemainverb:
©Universitat de Barcelona
M.LOBO&E.CARRILHO
160
(29) a.Quando acampainha tocou, ela já terminara
when thebell rang, shealreadyfinish.PPF
“Whenthebellrang,shehadalreadyfinished.”
b. Quandoacampainha tocou, ela já tinhaterminado
when thebell rang, shealreadyhadfinished
“Whenthebellrang,shehadalreadyfinished.”
They may also occur in conditional clauses ((30a), (31a)), as a variant to the
imperfect subjunctive (imperfeito do conjuntivo) ((30b), (31b)), similarly to what
happens in Spanish where both forms are traditionally classified as imperfect
subjunctive(Rojo&Veiga1999):
(30) a. Tivera eumenosvinteanos,tudo seria diferente.
have.ra I less twenty years, all would.bedifferent
b. Tivesse eumenosvinte anos,tudo seriadiferente
have.IMPSBJI less twenty yearsallwould.bedifferent
“IfIwastwentyyearsyounger,everythingwouldbedifferent.”
(31) a. Seforacharrueco,era omesmo.(SRP09)
ifbe.raplough was thesame
b. Sefosse charrueco,era omesmo.
ifbe.IMPSBJ plough was thesame
“Ifitwasaplough,itwouldbethesame.”
In non-standard varieties, some modal verbs (haver de ‘have to’ and dever
‘must/should’), as well as the volitional verb querer ‘to want’can also bear the -ra
ending ((32a), (33a), (34a)), and in this case they are equivalent to the imperfect
indicativeortheconditionalinthestandardvariety((32b),(33b),(34b)):
(32) a. Eunãodeveradeteridoapanhartomate.(STJ02)
Inot shouldofhavegonepicktomato
b. Eunãodevia/deveriateridoapanhartomate.
Inot shouldhavegonepicktomato
“Ishouldn’thavegonetopicktomatoes.”
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
161
(33) a. queaquiloquehaveradecairporsi.(CLH17)
thatthat.thingthathadtofallbyitself
b. queaquiloquehavia/haveriadecairporsi.
thatthat.thingthathad/would.havetofallbyitself
“thatthatthingwouldhavefallenbyitself
(34) a. Euquiseraserrico.
Iwantedto.berich
b. Euqueriaserrico.
Iwantedto.berich
“Iwantedtoberich.”
Verbal formsending in -ra canalsooccur in someexclamative contextswitha
counterfactualvalue(35a).Inthesamecontext,wecanfindthepluperfectsubjunctive
(auxiliaryter‘have’intheimperfectsubjunctivefollowedbythepastparticipleofthe
mainverb)(35b).
(35) a. Senãodápassada,comera!(STA19)
ifnot give step ate
b. Senãodápassada,tivessecomido!
ifnot give step had eaten
“Ifyoudonotwalk,youshouldhaveeaten!”
Finally,verbalformsin-ramayoccurinidiomaticexpressionswithadesiderative
value(tomaraandquemmedera):
(36) Tomaraagentesaúde!(AJT32)
caughtthepeoplehealth
“Wewishwewerehealthy.”
(37) Quemmederaterunsolhosassim.(MTV43)
Whomegavehavesomeeyesso
“IwishIhadeyeslikethat.
©Universitat de Barcelona
M.LOBO&E.CARRILHO
162
3.3.2.Comparingsources
3.3.2.1.Verbalformsin-rainCORDIAL-SINandinALPI
The morphosyntactic questionnaire of ALPI includes four questions that
potentially induce verbal forms in -ra (Table 5). Three of them correspond to
conditional clauses, in which there is variation in Spanish between the imperfect
subjunctivein-raorin-se.
Q.Nb. Page Question
386 28 Situvieradinerolocompraría
Setivessedinheiro,comprava-o/comprá-lo-ia
“IfIhadmoney,Iwouldbuyit”
387 28 Siestudiaseaprendería
Seestudasse,aprendia/aprenderia
“IfIstudied,Iwouldlearn”
388 28 Sipudieralamataría
Sepudessematava-a/matá-la-ia
“IfIcould,Iwouldkillher”
389 28 Yoquisieraserrico
Euqueriaserrico
“Iwouldliketoberich”
Table5.Questionsincludingpotentialcontextsforverbalformsin-raintheALPI(threeconditional
clausesandacounterfactualmodalcontext)
Ifwemap thedifferentverbal formsending in -rabasedon thedialect corpus
andontheALPIdata,accordingtotheircontexts,weobtainthefollowingmaps:
As illustrated in Map 10, we find verbal forms in -ra with different values in
CORDIAL-SIN, although they arenot very productiveoverall, even for thepluperfect
value. The uses corresponding to the pluperfect do not have a clear geographic
distribution in the corpus data and can be found in four locations (two northern
locations, one in the Azores and one in the south-east coast). As for modal verbs
ending in -ra, we may observe that they are only found in the Azores and in two
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
163
localitiesofsouthernPortugal.Inthecorpusdata,verbalformsin-racorrespondingto
the imperfect subjunctive are only found in two locations near the western border
withSpain.
In theALPI data, the relevantquestions rarely induceda verbal form in -ra: in
conditionalclausesthereareonlytwooccurrencesoftheseformsinthecentralregion
–Almeirim(Santarém),Montemor-o-Velho(Coimbra)–forthequestionsanalyzedby
MouzoVillar(2008),andtwooccurrencesof-ra formsforQ.389thatweanalyzed–
AlmeirimandRioMaior(Santarém).
Maps10and11.Occurrencesofverbalformsendingin-raaccordingtocontextinCORDIAL-SINandinALPI(mapsfromCarrilho&Lobo2012)
Inconditionalstructures,thepredominantformsareimperfectsubjunctivesin
-sse,asinthestandardvariety;forQ.389,wheretheverbalformhasamodalvalue,
mostresponsescontainedanindicativeimperfect(queria ‘wanted’).Theareaswhere
the forms were registered in CORDIAL-SIN and in the ALPI data are not the same,
althoughtheyaresimilarforthemodalvalue.
©Universitat de Barcelona
M.LOBO&E.CARRILHO
164
4.Conclusions
The three case studieswe analyzed comparing data obtained from the dialect
corpusanddataobtained inresponsestotheALPIquestionnaireallowustoconfirm
the classical advantages and drawbacks of these two kinds of sources and to some
extenttheircomplementarity.
Directorindirectquestionnairesofatlasesinducelessspontaneousresponsesto
morphosyntacticphenomena.Thiscouldexplaintosomeextenttheabsenceofnon-
standardorregionalvariants.Asforgerunds, for instance,theareawherethey(asa
varianttoa+ infinitive)werefound intheALPIdata ismuchmorerestrictedthan in
theCORDIAL-SINdata.Verbalformsin-raalsohaveawiderdistributioninthecorpus
whencomparedtotheALPIdata.Aquestionnaireallows,however,directcomparison
betweenthedifferentnetworkpoints.Thedataobtained inthecorpusshowusthat
thequestionnairehastobecarefullybuiltsoastoincludetherelevantcontexts.This
did not happen for all phenomena in the case of ALPI: crucial contexts of
morphosyntactic variation in the Portuguese domain (gerunds with auxiliaries,
peripheralgerundadverbialclauses)werenotconsidered.
The corpus also has some limitations in what concerns the study of
morphosyntacticphenomena.Somecontextsmaybeabsentjustbecausetheyarenot
veryfrequentinspontaneousspeechanddidnotshowupintheparticulardiscourse
situation. We did not find gerunds following perceptive verbs in the corpus, for
instance.Thecorpusallowsus,however,tomeasuretheproductivityofthedifferent
structuresandoflinguisticvariantsandtocarryamoredetailedstudyofthesyntactic
distribution of linguistic forms. The study of the frequency of possessives without
determinersinthecorpusgaveusimportantinformationonasymmetriesintheuseof
thenon-standardvariantacrossthePortugueseterritory.
Wewould liketoconcludethatacarefullybuiltmorphosyntacticquestionnaire
taking into account the relevant contexts is crucial for a better understanding of
syntactic variation. As we have seen, the ALPI questionnaire does not include
fundamental morphosyntactic variables for the Portuguese domain. In that respect,
thecorpusdatagiveusageneralpictureofrelevantmorhosyntacticphenomenaand
©Universitat de Barcelona
Dialectologia.Specialissue,V(2015),141-166.ISSN:2013-2247
165
are thus extremely important to serve as the basis for the preparation of a more
accuratemorphosyntacticquestionnaire.
Finally, we would like to highlight that, in spite of the traditional problems
encountered when analyzing both sources, in general the data obtained from the
corpus and from the ALPI questionnaire are quite consistent: both sources show a
lexical conditioning in the absence of determiner before possessive and a southern
distribution of gerund forms in verbal periphrases and complements to perceptive
verbs (although, since the atlas does not include the islands, some important non-
standardmorphosyntacticphenomenaarenotsufficientlyrepresented).
To conclude, we would like to stress the importance of taking into account
different sourceswhen studyingmorphosyntactic variation. The data obtained from
the corpus and from the responses to the ALPI questionnaire complemented each
other.
References
CALLOU, D. (2008) “Gramática, variação e normas”, in S. Rodrigues Vieira & S. Figueiredo
Brandão(orgs.),Ensinodegramática:descriçãoeuso,SãoPaulo:Contexto,13-29.
CARRILHO, E. & M. LOBO (2012) “Contribution à l'étude de la variation syntaxique dans le
domaineibéroroman”,inM.Oliviéri,G.Brun-Trigaud&Ph.DelGiudice(eds.),LaLeçon
des dialectes.Hommagesà Jean-PhilippeDalbera, Alessandria: Edizioni dell'Orso, 323-
336.
CARRILHO,E.&S.PEREIRA(2011)“Sobreadistribuiçãogeográficadeconstruçõessintácticasnão-
padrãoemPortuguêseuropeu”, inTextosSeleccionadosdoXXVIEncontroNacionalda
Associação Portuguesa de Lingüística, CD-ROM, Lisboa: APL, 125-139.
[http://www.apl.org.pt/docs/26-textos-seleccionados/Carrilho_Pereira.pdf].
FERNÁNDEZ-ORDOÑEZ, I. (2011) La lengua de Castilla y la formación del español,Madrid: Real
AcademiaEspañola.
FIÉIS,A.&M.LOBO(2011)“AspectosdasintaxedasoraçõesgerundivasnoPortuguêsMedieval
e no Português Europeu Contemporâneo”, in Textos Seleccionados. XXV Encontro
NacionaldaAssociaçãoPortuguesadeLingüística,Porto:APL,419-434.
©Universitat de Barcelona
M.LOBO&E.CARRILHO
166
GARCÍAMOUTON, P., D. HEAP &M.-P. PEREA (2012) “The Present and the Future of theAtlas
Lingüístico de la Península Ibérica (ALPI)”, Dialectologia. Special Issue, III, 1-6.
[http://www.publicacions.ub.edu/revistes/dialectologiaSP2012].
LOBO,M. (2001)“OnGerundClausesofPortugueseDialects”, inA.Veiga,V.M.Longa&J.D.
Anderson(eds.),ElVerbo.EntreelLéxicoylaGramática,Lugo:Ed.TrisTram,107-118.
LOBO,M.(2003)AspectosdaSintaxedasOraçõesSubordinadasAdverbiaisdoPortuguês,PhD
Dissertation.UniversidadeNovadeLisboa.
LOBO,M. (2008) “Variaçãomorfo-sintáctica em dialectos do português europeu: o gerúndio
flexionado”,Diacrítica,22.1,25-55.
LOBO,M.&A. FIÉIS (2015) “Ordemdepalavrasemoraçõesgerundivasdoportuguês”.Paper
presented at the ‘Simpósio Gramática, Variação e Mudança’.Gallaecia. III Congresso
InternacionaldeLinguísticaHistórica.SantiagodeCompostela.
MATTOS E SILVA, R. V. (1989) Estruturas trecentistas: Elementos para uma gramática do
PortuguêsArcaico,Lisboa:ImprensaNacional–CasadaMoeda.
MIGUEL,M.(2002)“Possessivepronouns inEuropeanPortugueseandOldFrench”, Journalof
PortugueseLinguistics,2,214-240.
MOTA, M. A. (1997) “Les traits nombre et personne/nombre en portugais - l'oral dans ses
variétés”, in M. Bilger, K. van den Eynde & F. Gadet (eds.), Analyse Linguistique et
approches de l'oral. Recueil d'études offert en hommage à Claire Blanche-Benveniste,
Orbis/Supplementa,Tome10,Leuven-Paris:Peeters,339-345.
MOUZOVILLAR,R.(2008)“Análisecontrastivaentregalegoeportugués:asoraciónscondicionais
no ALPI”, in Textos Seleccionados. Encontro Nacional da Associação Portuguesa de
Lingüística, Lisboa: APL, 589-596. [http://www.apl.org.pt/docs/24-textos-
seleccionados/40-Villar.pdf].
RINKE, E. (2010) “A combinação do artigo definido e pronome possessivo na história do
português”,EstudosLingüísticaGalega,2,121-139.
ROJO,G.&A.VEIGA(1999)“Eltiempoverbal.Lostiempossimples”,inI.Bosque&V.Demonte
(orgs.),GramáticaDescriptivadelaLenguaEspañola,vol.II,Madrid:Espasa,2867-2934.
©Universitat de Barcelona