Dialectologia17(2016),51-70.
ISSN:2013-2247
Received1December2014.
Accepted10March2015.
THREETYPESOFPREPOSITIONSINSPANISHSESENTENCES.
CONSEQUENCESFORCROSS-DIALECTALSTUDIES
ÁngelJ.GALLEGO
UniversitatAutònomadeBarcelona* ∗
[email protected]
Abstract
This paper discusses agreement patterns of SE sentences in different Spanish dialects. Special
attention is paid to situations where the verb agrees with Case-marked internal arguments (cf. Torrego
1998, López 2012) bypassing the preposition (e.g., Se ayudaron a los banqueros, Eng. ‘Bankers were
helped’), and to a previously unnoticed case in which agreement occurs across a non-clitic related
preposition (e.g., Se saben de diversos factores, Eng. ‘Different factors are known’). A micro-parametric
approachisputforwardwherebytwofunctionalelementsholdthekeytoaccountingforthefacts:onthe
onehand,thefeaturespecificationofvandT(thelocusofstructuralCase)mayvary,and,ontheother,
the precise nature of what we label “P” may range over three possible manifestations: (i) a bona fide
preposition,(ii)anapplicativeelement(potentiallyassociatedtoaclitic),and(iii)thespell-outofafeature
withinagivenfunctionalcategory.
Keywords
Spanish,impersonal/passivese,syntax,agreement,prepositions
* Departament de Filologia Espanyola. Facultat de Filosofia i Lletres. Edifici B. Universitat Autònoma de
Barcelona.08193Bellaterra(Barcelona),Spain.
∗
51
Á.J.GALLEGO
TRESTIPOSDEPREPOSICIONESENORACIONESCONSEDELESPAÑOL.
CONSECUENCIASPARAESTUDIOSDIALECTALES
Resumen
Este artículo discute los patrones de concordancia de oraciones con SE en diferentes dialectos del
español.Seprestaespecialatenciónasituacionesenlasqueelverboconcuerdaconargumentosinternos
quehanrecibidocaso(cf.Torrego1998,López2012),ignorandolapreposiciónquelosintroduce(e.g.,Se
ayudaronalosbanqueros),yaunavariantenodescritapreviamenteenlaquelaconcordanciatienelugar
atravésdeunapreposiciónnorelacionadaconclíticos(e.g.,Sesabendediversosfactores).Elpresente
trabajoofreceunplanteamientomicro-paramétricoenelquedoselementosfuncionalessonclavepara
darcuentadeloshechos:porunlado,laespecificaciónmorfológicadevyT(ellocusdelcasoestructural)
puede variar, y, por el otro, la naturaleza específica de lo que llamamos “P” puede adoptar tres
manifestaciones:(i)unapreposiciónbonafide,(ii)unelementoaplicativo(potencialmenteasociadoaun
clítico),y(iii)lamanifestacióndeunrasgodeunacategoríafuncional.
Palabrasclave
español,impersonal/pasivaconSE,sintaxis,concordancia,preposiciones
1.Introduction
It is well-known that preposition stranding is a cross-linguistically restricted
phenomenon (cf. Law 2006 and references therein for discussion). Thus, Romance
languages such as Spanish prevent instances of A-bar movement stranding a
preposition,asnotedbyCampos(1991):
(1)
*Quiéncontaron
todos con? whocounted
all with
(Spanish)
Whodideverybodycounton?
[fromCampos1991:741]
Whatever the factor responsible for (1) (cf. Abels 2003, Hornstein & Weinberg
1981, Kayne 1984, and Truswell 2009 for different accounts), it plausibly holds in the
caseofpseudopassives,whichareruledouttoo:
52
Dialectologia17(2016),51-70.
ISSN:2013-2247
(2)
*José es
contado
con
portodos(Spanish)
José be
counted
with byeverybody
Joséiscountedonbyeverybody
[fromCampos1991:741]
The literature on these phenomena has emphasized the empirical observation
that pseudopassivization is more restricted than P-stranding (cf. Abels 2003 and
Truswell2009).Thegoalofthisshortpaperistodiscusspreviouslyunnoticeddatafrom
non-standard Spanish that indicate that this language can display a pseudopassive
patterninthecontextof“SEpassives.”Interestingly,pseudopassivizationisbarredwith
“BE (or periphrastic) passives,” which we take to reinforce the structural and
morphological differences of the vP of SE and BE passives (cf. Mendikoetxea 1992,
1999).
The paper is divided as follows. Section 2 provides overview of the agreement
optionsofSEsentences.Sections3and4discussthepropertiesofwhatiscalled“hybrid
pattern” and what I call “residual pseudopassives” respectively; section 3 further
outlines an account of the facts that capitalizes on the properties of functional
categories,thusadoptingamicro-parametricapproach.Section5summarizesthemain
conclusions.
2.SEsentences:basicproperties
The literature on SE sentences has discussed the morphological and syntactic
intricacies associated to this clitic (cf. Raposo & Uriagereka 1996; D’Alessandro 2007;
Mendikoetxea1992,1999;andLópez2007,amongothers).InthecaseofSpanish,itis
knownthatSEcanparticipateinbothpassive(agreeing)andimpersonal(non-agreeing)
sentences:
(3)
a.Sevendieron loscoches
SEsold-3.pl thecars
PASSIVESE
(Spanish)
Thecarsweresold
53
Á.J.GALLEGO
b.Seayudó
SEhelped-3.sg
alosestudiantes
IMPERSONALSE
(Spanish)
tothestudents
Thestudentswerehelped
(3b)isatransivitivesentence,whichinasystemlikeChomsky’s(2001)meansthat
v is φ-complete and assigns accusative Case to the DP object los estudiantes, SE
plausibly occupying the position of the external argument (as argued by Raposo &
Uriagereka1996andLópez2007).(3a),ontheotherhand,isapassivestructure,where
visφ-defective,andtheinternalargumentreceivesnominativeCasefromT.LikeinBE
(or periphrastic) passives, the subject can remain in its base-generation position or
moveto[Spec,TP]:
(4)
a.Sevendieronloscoches(Spanish)
SEsold-3.plthecars
Thecarsweresold
b.Loscoches sevendieron(Spanish)
thecarsSEsold-3.pl
Thecarsweresold
ThesetwooptionsforSEsentenceshavebeendocumentedintraditionalatlases,
likeTomásNavarroTomás’ill-fatedALPI(AtlasLingüísticodelaPenínsulaIbérica):
54
Dialectologia17(2016),51-70.
ISSN:2013-2247
(5)
a.Secortarontreintapinos(Eng.‘Thirtypineswerecut’)
b.Secastigóalosladrones(Eng.‘Thieveswerepunished’)
[fromdeBenito2010:8,14]
One other well-known fact is that SE passives align with BE passives in many
respects.Interestinglyenough,Mendikoetxea(1999:§26.3.2.2.)notesthatSEpassives
canmanifesteitherfull(person,number)orpartial(defective)agreement,atraditional
observation that goes back to Bello (1847) (cf. Martín Zorraquino 1979 for additional
discussion):
55
Á.J.GALLEGO
(6)
a.Sevendenbotellas
SEsell-3.plbottles
PASSIVE1(fullagreement) (Spanish)
PASSIVE2(defectiveagreement)(Spanish)
Bottlesweresold
b.Sevendebotellas
SEsell-3.plbottles
Bottlesweresold
The second pattern of SE passives (non-agreeing passives, sometimes collapsed
with impersonal passives) can be found already in Old Spanish, but it is also found in
present-day non-European Spanish, as pointed out in Mendikoetxea (1999) and RAEASALE (2009). There are different factors that seem to conspire to yield the second
patternin(6)(cf.RAE-ASALE2009).Ilistthembelow:
(7)
a.Thecategoryoftheinternalargument(DPorNP)
b.Thepreverbalorpostverbalpositionoftheinternalargument
c.Thegrammaticalaspectoftheverb(perfectivevs.imperfective)
d.Thepresenceofdativearguments
e.Thespecificproximityoftheinternalargument(localityconditions)
Intheexamplesbelow,wecanseehowthejustlistedfactorshaveanimpacton
agreementprocessesinSEpassives(cf.RAE-ASALE2009:§41.12candff.):
(8)
a.Senecesitaaprendices
a’.*?Senecesitalosaprendices
SEneed-3.sglearners
SEneed-3.sgthelearners
Learnersareneeded
Learnersareneeded
b.Aquísenecesitaaprendices
b’.*?Aprendicessenecesitaaquí
hereSEneed-3.sglearners
learnersSEneed-3.sghere
Learnersareneededhere
Learnersareneededhere
c.Sevendelibros
c’.?Sevendiólibros
SEsell-3.sgbooks
SEsell-3.sgbooks
Booksaresold
Booksweresold
56
Dialectologia17(2016),51-70.
ISSN:2013-2247
d.Selesda
caramelosalosniños
SEcl.datgive-3.sg candiestothechildren
Childrenaregivencandies
e.Seveíaaunladoyaotrodelcaminolasmansiones...
SEsee-3.sgatonesideandtootherof-thetrackthemansions
Mansionswereseenatonesideandtheotherofthetrack
As for non-European varieties, RAE-ASALE (2009: 3094) notes that “The
distributionisnotperfect[…]ithasbeenobservedthatAndean,Chilean,andRiverPlate
Spanishfeatureoverlappingmoreclearly”(mytranslation).Someexamplesaregivenin
(9),takenfromRAE-ASALE(2009):
(9)
a.Ensupartidoserespetalaslibertades... inhispartySErespectthefreedoms
Freedomsarerespectedinhisparty
(MexicanSpanish)
b.Seatendióoncesolicitudes...
(MexicanSpanish)
SEattendelevenapplications
Elevenapplicationswereattended
Tosumupsofar,SEpassivesentencesdisplayvariousagreementpatternsinthe
different varieties of Spanish. For the most part, such patterns concern either the φcomplete/φ-defectivestatusofT(thelocusofnominativeCase)orthepossibilitythat
the internal argument (the would-be subject) is within the search domain of T (cf.
Chomsky2001,Legate2014).Inanyevent,thisvariationconcernsSEpassives,whichdo
notfeatureDOM.WewouldliketoconcentrateonSEsentenceswithDOM(so-calledSE
impersonals),forthesamedichotomyisfoundthere.
57
Á.J.GALLEGO
3.SEpassives(1):thehybridpattern
As noted at the outset of this paper, the clitic SE can participate in passive and
impersonalstructures.Therelevantminimalpairwasgivenin(3),andisrepeatedhere
as(10)forconvenience:
(10)
a.Sevendieronloscoches
SEsold-3.plthecars
PASSIVESE
(Spanish)
(Spanish)
Thecarsweresold
b.Seayudóalosestudiantes IMPERSONALSE
SEhelped-3.sgtothestudents
Thestudentswerehelped
Although the verb typically fails to agree with the internal argument in (10b),
agreementdoesoccurinsomeinstancesofCase-markedinternalarguments.Abstractly,
thispattern,whichisdubbed“hybrid”byRAE-ASALE(2009),canbedepictedasin(11):
(11)
[SET[VPV...[aXP]]]
⏐____________↑
Again, we see that agreement may or may not occur already in previous stages
andinnon-EuropeanvarietiesofSpanish:
(12) a.Aestosnosepuedenpremiar tothesenotSEcan-3.plaward
Thesecannotbeawarded
(Quijote)
(SalvadorHoy)
b.Sepremiaronalosmejoresjinetes
SEaward-3.pltothebestriders
Thebestriderswereawarded
[fromRAE-ASALE2009]
58
Dialectologia17(2016),51-70.
ISSN:2013-2247
If we consider impersonal SE more closely, notice that the v of this structure
shouldbev*,thuscapableofassigningaccusative.However,itseemsthatthisCaseis
restrictedtoanimateinternalarguments:
(13)
a.*Elarroz,selocomecadadomingo
thericeSEiteat-3.sgeverySunday
Therice,itiseateneverySunday
(Spanish)
[fromOrdóñez2004:6]
b.Aunhombre,noselojuzgasinpruebas (Spanish)
toamannotSEhimjudgewithoutproof
Amanisnotjudgedwithoutevidence
Thispatternseemsprettyrobust.Soonecouldassumethegeneralizationin(14):
(14)
IftheinternalargumentisCase-marked(a-XP),thenSEvisv*(φ-complete)
Thissaid,therearesomeexceptions.Theexamplein(15)indicatesthat,incertain
circumstances,vcanassignaccusativeevenwithinanimate(nonCase-marked)internal
arguments(thesentenceisadaptedfromMarías2008):
(15) Cuandosereproduceloacontecido,sinquererselodeforma(Spanish)
whenSEreproduceithappenedwithoutwantSEitdistort-3.sg
Whenonereproduceswhathashappened,onedistortsitinvoluntarily
ItseemsthatthispatternishighlyrestrictedinthecaseofEuropeanSpanish.Itis
moreactiveinnon-Europeanvarieties.Inparticular,RAE-ASALE(2009:§41.12m)argues
that accusative assigning v* with inanimate internal arguments is licensed in the
Andean,Chilean,andRiverPlateareas.
(16)
a.Seplanificanlosescapes,selostecnologiza
SEplan-3.pltheescapesSECLtechnologize
Escapesareplanned,theyaretechnologized
(Spanish)
59
Á.J.GALLEGO
b.Fracasansolocuandoselasusamal
fail-3.pljustwhenSECLuse-3.sgbad
Theyfailonlywhentheyareusedinawrongway
(Spanish)
c.Selosentiendesinquehayansidoexplicados
(Spanish)
SECLunderstand-3.sgwithoutthathavebeenexplained
Theyareunderstoodwithouthavingbeenexplained
[fromRAE-ASALE2009:3098]
Onemoreexamplesofthisexoticpatternis(17),thistimefromEuropeanSpanish
(cf.MartínZorraquino1979,Fernández-Ordóñez1999):
(17)
a.Esteúltimo[avión]yaestálistoydebeserretirado,puesporcadadíaquepasa
ynoseloutilizasepierdedineroyademáshayquepagarmulta
(LaNación,7-IX-1975,pág.20,c-7,apudMartínZorraquino1979
Thislastplaneisreadyandmustbetakenaway,sinceeverydaythatgoesonanditis
notusedwelosemoneyandwehavetopay
b.Ellomoselodaunavueltaenlasartén,selometealaolla,selocubrecon
aceitedeoliva
Themeathastobeturnedupsidedowninthepan,youputitintothepot,youcoverit
witholiveoil
(CampodeSanPedro,Segovia,COSER3702,apudFernández-Ordóñez1999)
[fromdeBenito2013:147]
Insum,pronominalizationofCase-markedinternalarguments,likeaPedro(Eng.
‘toPedro’)in(18),asin(19):
(18)
SecriticaaPedro SEcriticizetoPedro
Pedroiscriticized
(Spanish)
60
Dialectologia17(2016),51-70.
ISSN:2013-2247
(19)
Pronominalizationof(18)(#indicatesthattheformisnotpreferred)
a.Selocritica (non-leísta/AmericanSpanish)
b.Se{#lo/le}critica
(leísta/EuropeanSpanish)
This raises the question whether Case-marked internal arguments receive true
accusative. If they do not, then that would explain the restricted availability of lo/la
(onlywithanimates),andthepreferenceforleinEuropeanSpanish.Thisprocessoflo>
le shift with SE can be seen even by speakers that are not leístas with masculine in
regulartransitivesentences,asnotedbyOrdóñez(2004).
(20)
Sihayquefusilar-lo,SElefusila ifthere-be-3.sgthatshoot-CLSECLshoot-3.sg
Ifhemustbeshot,heisshot
(EuropeanSpanish)
[fromP.Preston,Franco,citedbyOrdóñez2004]
Unlike European Spanish, Mexican Spanish shows no le clitic with standard
transitivesentences—itisanon-leístadialect.Alldirectobjects,masculineorfeminine,
deploythestandardmasculinevs.femininedistinction:lo/la.Thiscanbeseenin(21):
(21)
a.AJuanlovieron contento
toJuanCLsee-3.pl happy
Juan,hewasseenhappy
(MexicanSpanish)
(MexicanSpanish)
b.AMaríalavieroncontenta
toMaríaCLsee-3.plhappy
María,shewasseenhappy
However,inthepresenceofSE,MexicanSpanishobligatorilyshiftstole.
(22)
a.AJuanSEleviocontento toJuanSECLseehappy
Juan,hewasseenhappy
(MexicanSpanish)
61
Á.J.GALLEGO
b.AMaríaSEleviocontenta
toMaríaSECLseehappy
María,shewasseenhappy
(MexicanSpanish)
This shift to le does not occur in Río de la Plata Spanish. This south-American
dialect, contrary to Mexican Spanish or European Spanish, has doubling with Casemarkedinternalargumentsbeyondstrongpronouns:
(23)
a.(lo) vi
aJuan
CLsaw-1.sgtoJuan
IsawJuan
(RiverPlateSpanish)
(RiverPlateSpanish)
b.*(la)vi
alalibreta
CLsaw-2.sgtothenotebook
Isawthenotebook
Inthisdialectnoleshiftoccurswithdirectobjects:
(24)
a.Se(lo)escuchó[alniño]
SECLheard-3.sgto-theboy
Theboywasheard
(RiverPlateSpanish)
(RiverPlateSpanish)
b.Se(la)escuchó[alaniña]
SECLheard-3.sgto-theboy
Thegirlwasheard
Descriptively, Spanish dialects that allow clitic doubling with Case-marked direct
objectsdonotshifttoleinimpersonalSEconstructions(cf.Ordóñez&Treviño2007for
anaccount).
From all the discussion above, one can plausibly conclude that impersonal
sentenceswithSEaredividedintotwodialectsinSpanish:
62
Dialectologia17(2016),51-70.
ISSN:2013-2247
(25)
a.DialectA:visφ-defective(accusativeunavailable)
b.DialectB:visφ-complete(accusativeavailable)
Technically,thisamountsto(26):
(26)
a.DialectA:
[vPv[VPV[PPa[DPOBLIQUE]]
⏐___↑
b.DialectB:
i.[vPvφ[VPV[KPaDPACC]]]
⏐___________↑
ii.[...Tφ...[vPv[VPV[KPaDPNOM]]]
(leístaSpanish)
(non-leístaSpanish)
(hybridpattern)
⏐___________________↑
As reported by Ordóñez & Treviño (2007), Mexican and Argentinian varieties of
Spanish, which belong to dialect B, may show agreement with an internal argument
precededbyaccusativea(theaforementioned“hybridpattern”).
(27)
a.Finalmente,secastigaronalosculpables(MexicanSpanish)
finallySEpunished-3.pltotheculprits
Finally,theculpritswerepunished
b.Seevacuaronamásde120.000damnificados
SEevacuated-3.pltomoreof12.000damaged
(ArgentinianSpanish)
Morethan120.000damagedpeoplewereevacuated
[fromOrdóñez&Treviño2007:12]
The data in (27) pose a puzzle. They clearly indicate that the φ-Probe in T can
agree with the internal argument, but this is unexpected, given that the latter has
alreadybeenCasemarked(byv),andisthus“inactive”inChomsky’s(2001)terms.Itis
nonethelesspossible⎯anditiswhatwewouldliketoproposehere⎯thatdialectB
dividesfurtherintoasubdialectthatfailstoCasemarktheinternalargument.
Awaytogoaboutthissubtlermicro-parametricdistinctionistotakeSpanishato
vary within the relevant varieties. Building on much literature on this topic (cf. López
63
Á.J.GALLEGO
2012, Torrego 1998, and references therein), we assume that the vocabulary item a
correspondstothreedifferentelementsinSpanish:
(28)
Athree-wayanalysisforainSpanish
a.Aspell-outofatruepreposition
b.Thespell-outofaCase/clitic-relatedprojection(cf.López2012,Torrego1998)
c.Thespell-outofafeatureofaCase/clitic-relatedprojection
Clearly,inthevarietiesofSpanishthatlicense(28b),aisnotapreposition,andit
isnotthestandardCase-markingmorphemeofDOM—forotherwiseagreementwould
fail —, so we are left with option (28c): a is the spell-out of a feature, not even a
projecting category. Given that the v of dialect Bii is φ-defective and that a is not a
preposition,itfollowsthattheinternalargumentcanlong-distanceagreewithT.
HavingconsideredthebasicCase-agreementconfigurationswhereSEisinvolved,
we would like to briefly consider a pattern that seems to be intimately related to the
onein(28c),andwhichquicklyevokestheprofileofpseudopassivestructures.
4.SEpassives(2):residualpseudopassives
Asjustnoted,theexamplesin(27)showthattheφ-ProbeonTcanlong-distance
agree with the internal argument, ignoring the would-be preposition — actually a
feature,underthepresentaccount—a.Thisissomewhatsurprising,asitresemblesa
pseudopassive.
Yetmuchmoresurprisingly,othervariants(mainlyAmerican)ofSpanishdialectA
manifest agreement with DPs contained in lexical PPs. The following data are from
differenton-linesources:
(29)
a.Dijoquesehablaronconlas
autoridades (AmericanSpanish)
saythatSEtalked-3.plwiththe authorities
Hesaidthattheauthoritiesweretalkedto
[http://www.santiagodigital.net/index.php?option=com_content&task=view&id=13837&Itemid
=17]
64
Dialectologia17(2016),51-70.
ISSN:2013-2247
b.
EnSantiagoanocheseinformarondecuatrohomicidios(AmericanSpanish)
inSantiagolastnightSEinformed-3.ploffourhomicides
FourhomicideswerereportedlastnightinSantiago
[http://www.periodismoglobal.cl/2006/08/la-democracia-de-la-udi.html]
c.
Elcomercioonlinesumó[...]100millonesdetransacciones(AmericanSpanish)
thetradeonlineadded-3.sg100millionsoftransactions
[...]cuandosellegaronalos74,3millonesdeoperaciones
whenSEarrived-3.pltothe74,3millionsofoperations
The online trading added 100 million transactions when 74,3 million operations were
reached
[http://www.elpais.com/articulo/economia/comercio/electronico/volvio/batir/record/2010/elp
epueco/20110506elpepueco_7/Tes]
d.
Enrealidadsedependendetantosfactores
(AmericanSpanish)
inrealitySEdepend-3.plofso-manyfactors
queestoprovocaunaextremadificultad
thatthisprovokesaextremedifficulty
Actually,onedependsonsomanyfactorsthatitmakesthingsextremelydifficult
[http://diegotenis9.wordpress.com/]
MoredatacanbeobtainedfromtheCREAdatabase,andfromGoogle:
(30)
a.Sólosedisponendedatosdematrículas... justSEdispose-3.plofdataofregistration
Wejusthavedataonregistration
(ElSalvador)
b.Aunquenosedisponendecifrasexactas...
althoughnotSEdispose-3.plofnumbersexact
Althoughwedon’thaveexactnumbers
(CostaRica)
65
Á.J.GALLEGO
c.Sísesabendediversosfactoresqueinfluyen...
yesSEknow-3.plofdiversefactorsthatinfluence
Wedoknowfactorsthatinfluence
(Spain)
[fromCREA:http://corpus.rae.es/creanet.html]
(31)
a.Todavíaseconfíanenlosmilagros
yetSEtrust-3.plinthemiracles
Theystillbelieveinmiracles
(México)
(Chile)
[http://www.sinembargo.mx/30-03-2014/947521]
b.Cuandosehablandelassupuestasdesigualdades
whenSEtalk-3.ploftheallegedasymmetries
Whentheytalkabouttheallegedasymmetries
[http://blog.lanacion.cl/2014/03/11/desigualdades-de-genero-en-el-emprendimiento/]
These data are rather restricted due to normative pressures, but they are not
isolatedon-linehits.Themainconclusiontobedrawnfrom(29)isthatcertaindialects
ofSpanishdisplay,contrarytowhatistypicallyassumed,pseudopassives.
This raises at least two questions. The first one is whether, apart from “SE
pseudopassives”,Spanishcanalsodisplay“BEpseudopassives”.Theanswerisnegative,
assentenceslikethosein(32)areruledoutbyAmericanSpanishspeakers,whofinda
sharpasymmetrywithrespecttotheexamplesin(30-31):
(32)
a.*Fueronhabladasconlasautoridades
(AmericanSpanish)
be-3.pltalked-3.fem.plwiththeauthorities
Authoriteswerespokento
b.*Fueroninformadosdecuatrohomicidios
(AmericanSpanish)
be-3.plinformed-3.masc.ploffourhomicides
Fourhomicideswerereported
66
Dialectologia17(2016),51-70.
ISSN:2013-2247
The asymmetry between (30-31) and (32) provides support for the idea that SE
and BE passives are morphologically and syntactically different, as has been argued in
theliterature(cf.Mendikoetxea1999).
Thesecondquestionisaparametricone:Howdoesagreementtakeplaceinsuch
varietiesofSpanish?Atfirstglance,thedialectsallowing(30-31)mustbeabletolicense
a‘reanalysis’process(howeveritmustbeimplemented,anissuewecannotinvestigate
here;cf.Hornstein&Weinberg1981,Kayne1975,2004,amongmanyothers)whereby
Tcanlong-distanceagreewiththecomplementsofP.
It is important to point out, to conclude, that even though pseudopassivization
seemstobeanoptioninSpanish,prepositionstrandingisstillimpossible.Thatistosay,
sentenceslikethosein(30-31)withtheagreeingDPin[Spec,TP](afterA-movement)or
[Spec,CP](afterA-barmovement)areimpossible.Whatistrulysurprising,andhasgone
unnoticedintheliterature,istheveryexistenceoftheexamplesin(30-31).Thisnotonly
suggests that Spanish does have a residual type of pseudopassives, it also seems to
threaten the empirical generalization that pseudopassives are cross-linguistically more
restrictedthanprepositionstranding.
5.Conclusions
Thispaperhasmadetwointerestingpoints.Ontheempiricalside,wehaveshown
that, along with the hybrid pattern of SE sentences, some dialects of Spanish feature
whatappeartobesomeformofpseudopassiveconstruction(seedatain30and31).Of
course,amorecarefulstudyisneeded,andthefactorstocontrolforare(atleast)the
following:(i)thetypeofverb(non-pronominal,agentive)thatallowspseudopassives,(ii)
thetypeofprepositionthatcanbecomeinertforagreementprocesses,(iii)thecategory
of the agreeing element (DP or NP), and (iv) the relevant source of data (journal,
newspaper,forum,CREA,Google,etc.).Quitepossibly,thesecouldjustbetyposorthe
resultoforalspeech,butthefactthatthis‘extended’hybrid(pseudopassive,ifweare
correct)patternisnotfoundwithadjuncts.Inotherwords,exampleslikethosein(33)
areunattested.
67
Á.J.GALLEGO
(33)
a.*Sehablaronenlasaulas
SEtalk-3.plintheclass
Peopletalkintheclass
(Spanish)
b.*Seaspiraronalpuestopormuchosmotivos(Spanish)
SEaspire-3.plto-thepositionformanyreasons
Peopleaspiretothepositionformanyreasons
Onthetheoreticalside,thispaperhasarguedthatthenatureofprepositionsmust
be divided into three types. The distinction between lexical and functional (or fake)
prepositions is not new in the field (cf. Abels 2003, Cuervo 2003, Demonte 1987, 1991,
1995,Pesetsky&Torrego2004,Romero2011),butwehavetriedtosharpenitinorderto
accountforthe(28b)/(28c)distinction.Muchworkisrequiredinthestudyoffunctional
categories,especiallyinthecontextofdialectalvariation,andthispaperisnothingbuta
smallcontributiontothisgoal.
References
ABELS, K. (2003) Successive cyclicity, anti-locality, and adposition stranding, PhD Dissertation,
UniversityofConnecticut.
BELLO,A.(1847)Gramáticadelalenguacastellana,destinadaalusodelosamericanos,edición
connotasdeRufinoJoséCuervo,2vols.,RamónTrujillo(ed.),Madrid:Arco/Libros.
BENITO, C. de (2010) “Las oraciones pasivas e impersonales con SE: Estudio sobre el ALPI”,
Dialectologia,5,1-25.<http://www.publicacions.ub.edu/revistes/dialectologia5/>
BENITO, C. de (2013) “(Esa tela) se la descose: La pronominalización del paciente en las
impersonalesreflejasdelespañolpeninsular”,Borealis,2.2,129-157.
CAMPOS,H.(1991)“PrepositionStrandinginSpanish?”,LinguisticInquiry,22,741-750.
CHOMSKY,N.(2001)“DerivationbyPhase”,inM.Kenstowicz(ed.),KenHale:ALifeinLanguage,
Cambridge,MA:MITPress,1-52.
CUERVO,M.C.(2003)Dativesatlarge,PhDdissertation,MIT.
68
Dialectologia17(2016),51-70.
ISSN:2013-2247
D’ALESSANDRO, R. (2007) Impersonal Si constructions: agreement and interpretation, Berlin:
MoutondeGruyter.
DEMONTE,V.(1987)“C-command,prepositionsandpredication”,LinguisticInquiry,18,147-157.
DEMONTE,V.(1991)Detrásdelapalabra.Estudiosdegramáticadelespañol,Madrid:Alianza.
DEMONTE,V.(1995)“DativealternationinSpanish”,Probus,7,5-30.
FERNÁNDEZ-ORDÓÑEZ, I. (1999) “Leísmo, laísmo y loísmo”, in I. Bosque and V. Demonte (dirs.),
GramáticaDescriptivadelaLenguaEspañola,Madrid,Espasa-Calpe,1317-1397.
HORNSTEIN,N.&A.WEINBERG(1981)“Casetheoryandprepositionstranding”,LinguisticInquiry,
12,55-91.
KAYNE,R.(1975)Frenchsyntax,Cambridge,MA:MITPress.
KAYNE,R.(1984)Connectednessandbinarybranching,Dordrecht:Foris.
KAYNE, R. (2004) “Prepositions as probes”, in A. Belletti (ed.), Structures and Beyond. The
CartographyofSyntacticStructures,vol.3,Oxford,NY:OxfordUniversityPress,192-212.
LAW,P.(2006)“Prepositionstranding”,inM.Everaert&H.vanRiemsdijk(eds.), TheBlackwell
companiontosyntax,Oxford:Blackwell,631-684.
LEGATE,J.(2014)Voiceandv,Cambridge,MA:MITPress.
LÓPEZ,L.(2007)LocalityandtheArchitectureofSyntacticDependencies,NewYork:Palgrave.
LÓPEZ,L.(2012)Indefiniteobjects.DifferentialObjectMarking,ScramblingandChoiceFunctions,
Cambridge,MA:MITPress.
MARÍAS, J. (2008) “Sobre la dificultad de contar”, Discurso de ingreso en la Real Academia
Española,27deabrilde2008,Madrid:RealAcademiaEspañola.
MARTÍN ZORRAQUINO, M. A. (1979) Las construcciones pronominales en español. Paradigma y
desviaciones,Madrid:Gredos.
MENDIKOETXEA, A. (1992) On the nature of Agreement: The Syntax of ARB SE in Spanish, PhD
Dissertation,UniversityofYork.
MENDIKOETXEA,A.(1999)“Construccionesconse:medias,pasivaseimpersonales”,inI.Bosque
and V. Demonte (eds.), Gramática descriptiva de la lengua española, Madrid: Espasa
Calpe,1575-1630.
ORDÓÑEZ, F. (2004) “Se Across Romance”, Talk given at GURT conference, Georgetown
University.
ORDÓÑEZ,F.&E.TREVIÑO(2007)“UnambiguousSE”,TalkgivenatXVIIColloquiumonGenerative
Grammar,UniversitatdeGirona(Girona),13-15June2007.
69
Á.J.GALLEGO
PESETSKY, D. & E. TORREGO (2004) “Tense, Case, and the nature of syntactic categories”, in J.
Guéron&J.Lecarme(eds.),Thesyntaxoftime,Cambridge,MA:MITPress,495-537.
REALACADEMIAESPAÑOLABancodedatos(CREA)[online].Corpusdereferenciadelespañolactual.
<http://www.rae.es>[Retrieved20April2014].
RAE-ASALE(2009)NuevaGramáticadelaLenguaEspañola,Madrid:Espasa.
RAPOSO, E. & J. URIAGEREKA (1996) “Indefinite SE”, Natural Language and Linguistic Theory, 14,
749-810.
ROMERO,J.(2011)Losdativosenespañol,Madrid:ArcoLibros.
TORREGO,E.(1998)TheDependenciesofObjects,Cambridge,MA:MITPress.
TRUSWELL, R. (2009) “Preposition-stranding, Passivisation, and Extraction from Adjuncts in
Germanic”,LinguisticVariationYearbook,8,131-177.
70