7/30/2019 Canon in Spanish Spanish American Literatura http://slidepdf.com/reader/full/canon-in-spanish-spanish-american-literatura 1/20 Required Reading: The Canon in Spanish and Spanish American Literature Author(s): Joan L. Brown and Crista Johnson Source: Hispania, Vol. 81, No. 1 (Mar., 1998), pp. 1-19 Published by: American Association of Teachers of Spanish and Portuguese Stable URL: http://www.jstor.org/stable/345448 . Accessed: 21/04/2013 12:02 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . American Association of Teachers of Spanish and Portuguese is collaborating with JSTOR to digitize, preserve and extend access to Hispania. http://www.jstor.org This content downloaded from 157.253.50.10 on Sun, 21 Apr 2013 12:02:56 PM All use subject to JSTOR Terms and Conditions
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
7/30/2019 Canon in Spanish Spanish American Literatura
Required Reading: The Canon in Spanish and Spanish American Literature
Author(s): Joan L. Brown and Crista JohnsonSource: Hispania, Vol. 81, No. 1 (Mar., 1998), pp. 1-19Published by: American Association of Teachers of Spanish and Portuguese
Stable URL: http://www.jstor.org/stable/345448 .
Accessed: 21/04/2013 12:02
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp
.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].
.
American Association of Teachers of Spanish and Portuguese is collaborating with JSTOR to digitize, preserve
and extend access to Hispania.
http://www.jstor.org
This content downloaded from 157.253.50.10 on Sun, 21 Apr 2013 12:02:56 PMAll use subject to JSTOR Terms and Conditions
Abstract: The requiredgraduatereading ists of 56 leadingPh.D.-granting panish aculties n the United
Stateswereanalyzed o characterize he currentcanonforSpanishandSpanishAmerican iterature.The da-tabaseconsistedof14,686 tems.Little onsensuswas foundregarding uthors, ndevenless forspecificworksof literature.Onlytwo authorsand twoworks fromSpainachieved100percent representation n the lists.
Thirty-nine uthors(one female)and 22 male-authored orksfromSpain,and 24 authors(two female)andtenmale-authoredworksfromSpanishAmericawerepresenton 75percentor more of the lists.Atthe otherend of the spectrum,nearly1,000differentitlesappearedustonce inthe database,demonstrating resenceon onlya single reading ist. The findingsshow littleagreementaboutwhatconstitutes iterary alue n thisfield.This has importantmplicationsorgraduate ducation.
The issue of literarycanons has beencentralto both academic andpopulardis-course in this country nthe lastdecadeofthe twentieth entury Gates).AlthoughhefieldofHispanic tudieshas not been insu-latedfromthis debate,there has been noformalattemptto describe the Hispaniccanon. We previously looked at a single
genre in ourmost recenthistoricalperiod,the contemporary novel (Brown and
Johnson). Ourfindings indicatedthat forthe post-1936novelcanon there is no una-
nimityandvery little agreement.This ar-ticle reports the findings of a study de-
signed to describe our current canon for
Hispanicliteratures in the United Statesacross alleras andgenres.Theoverarchingquestion that we sought to answer was:
does consensus indeed exist formost erasandgenres, and, if so, what is ourshared
literarycanonat this time?
Methods
Therequired raduate eadingistsof56
Ph.D.-grantingSpanish faculties in this
country supplied the data on which this
descriptionof ourliterarycanon is based.Our choice of programswas foundedon
published rankings of U.S. graduateschools,includinghe 40highest-rated ro-grams listed by Gourmanand 63 Spanishgraduate programs listed in Peterson'sGuide.Everyregion of the United Stateswas represented. Twenty-five ists came
from nstitutionsnthenortheast,nine fromthe midwest, five from the Pacific coast,four rom heRockyMountainegion,eightfrom the south Atlanticarea andfive fromsouthcentral tates. Listswereacquiredbymeans of letters to departmenthairs,with
follow-up elephonecalls when necessary.Wheneverpossible, graduatereading istsat thePh.D. level wereselectedfromthose
received,therationalebeingthat areading
list for the Ph.D. representsthe broadestpossible compendiumof requiredworks.Combined M.A./Ph.D. lists were usedwhen available. f an institutionhad inde-
pendentM.A.andPh.D.readingists,thesetwo lists were merged; duplicateentrieswere counted only once for that school.
This content downloaded from 157.253.50.10 on Sun, 21 Apr 2013 12:02:56 PMAll use subject to JSTOR Terms and Conditions
Reading ists at the M.A. level were usedwhen no otheroptionwas offered.The 56lists that comprisethis study include ten
Ph.D.orequivalentists,six combined ists,ten mergedM.A./Ph.D.lists, and 30 M.A.lists.
A databaseof the literatureportionsofthese reading istswasdevelopedandcom-
piled on a Universityof Delaware main-framecomputer,with the assistanceof Dr.LawrenceHotchkiss,LeadConsultant/Ana-
lystoftheComputingndNetworkServices
Department tthe Universityof Delaware.
For each institution,we enteredthe infor-mationon thereading ist.Listings eceived
commonly ncluded he authorandtitle ofeach work.We provided he date of publi-cationof each work as well as the national-
ity andgender of each author.Nationalitywas determined according to countryofbirthexceptwhenplacementon a readinglist conflicted with that criterion (e.g.,Cort6s);wemaintainedheintegrity f eachschool's classification
regardlessof the
author'sbirthplace.Forpoets,dateofbirthalso wassupplied.Thirtyyearswere addedto the dateofbirth norderto situatepoetsin their appropriatecenturies, since thiswould be the age by whichthey could rea-
sonably be expected to publish. For allother authors,dates of publicationdeter-mined hecenturyplacement.Weclassifiedworks into six genre categories,introduc-
ing greaterspecificity o the commonhead-
ing of "Prose" as a single entity. Thesegenredivisionswere:novel,poetry, heatre,shortfiction,andessay, with the sixth cat-
egory of "other" eservedfor those itemsthat couldnot be labeled underthese divi-sions (e.g., films).
In order to ascertain commonalities
amongvarious ndividualistingsofessaysand short fiction,some individualentrieswerecombinedunder he title of the collec-
tion in which they appeared. This enableddifferent selections from a single book to
register as repeated references to the same
volume. For example, the presence of
Larra's "Vuelva Ud. mafiana" on one
institution's list was tantamount to a listingof "En este pais" on another; both are con-
only orauthorswhoalreadyhadsignificantrepresentationon the reading ists, whichwe determinedo bepresenceonone-quar-ter of the lists. Its purposewas to revealcanonicalworks whose status could be ob-scuredbyreferences o individualhaptersor stories,which the computercounted asdistinctworks.
Listingsof a singleworkundermultipletitles (aproclivityof some authorssuch as
Sender,but also a reflectionofalternateorabbreviateditles)were consolidatedwhen
theywerenoted,to countasone entrywiththe initial date of publication.Erroneous
titles, inaccurate enre classifications,and
misspellingswere corrected when found.Some works and authors could not beidentifiedor classified,as they did not ap-pearin standard eferenceworks or inanybibliographical isting available o us. We
were, therefore,forced to exclude these
items, which totaled 82 entries. Since no
single unidentifiablework appeared onmore than one reading ist,the eliminationof these mystery tems hadno effecton ourcalculationsof canonicity.
The analysisofliterarygenresbycentu-ries covered the years from 1100 to 1991.Ourclosingdaterepresents he last dateofrevision for the reading ists received andtherefore s the last possibledate ofpubli-cation for an included work. This purelyhistorical division allows a dispassionatelook at literary evolution, free from theinfluenceof external frames of referencecontained ndivisionssuch as "Generationof '98"or even "GoldenAge."Withineach
century,literature rom Spainwas subdi-vided into novel, poetry, theatre, short
fiction,essay, and "other."Literaturerom
SpanishAmericawas subdivided he same
way for the 55 institutions that had SpanishAmerican literature on their reading lists;one institution had none. The five Portu-
guese-language reading lists that were re-
ceived were not included in the present
study.We sought the answer to a fundamental
This content downloaded from 157.253.50.10 on Sun, 21 Apr 2013 12:02:56 PMAll use subject to JSTOR Terms and Conditions
THE CANON IN SPANISH AND SPANISH AMERICAN LITERATURE3
question:which works and authors were
represented,andhow oftendidthey appearon the reading ists?For allgenres except
poetry,we analyzed he reading ists to seehow many titles and how many authorswerepresenton each list.Weincludedanycombination of works but counted eachauthoronly once per institution.We thendeterminedheproportionalepresentationof everywork and authorcited.Forpoetrywe recorded he author'spresenceoneach
list, again allowing any combination ofworks.This enabledus to achievean accu-
rate count on which to base calculationsofproportionalepresentation.We could notdetermineproportionalepresentationromtitles of poeticworksbecause an accuratecount could notbe obtained: ndividual o-ems, though comprisinga much smallershareof the poet'swork,wouldbe countedthe same as a singlevolumeofpoetry.Wedidnotrelate eachpoemto the first collec-tioninwhich it appeared,because to do sowould introduce
inaccuracies,giventhe
wide use of anthologies.We also looked atthe distributionof writersandtheir works
by gender. In analyzing his data,percent-ageswerecalculatedo fourdecimalplacesbutexpressed o the nearestwholenumber.
Results
One hundredpercentof the 74 Spanishfacultiesthatwere askedto participate e-
plied.Eighteenofthem,however,couldnotbe included nthis study.Ten did not actu-
allyoffer he Ph.D.Three did not use read-
ing lists, and three used readinglists thatwere not generated by facultybutby indi-vidualgraduate tudents.We includedoneformulation f requiredworks of literaturethatrepudiated he title of "ReadingList."Two model lists ofrecommendedreadingsforPh.D. candidatesouse indevising heir
own lists were not included in the presentstudy of readings that are explicitly re-
quired to fulfill degree requirements.The number of all entries for the reading
lists ranged widely. The shortest list (atthe
M.A. level) contained 44 items, and the
longest (a Ph.D. list) had 988. Our database
containing llthe lists totalled14,686tems.Atotalof780authorsappearedon the lists,including ne"Anonymous"romSpainand
one "Anonymous"romSpanishAmerica.Countriesrepresented, n alphabeticalor-
uador,El Salvador,Guatemala,Honduras,Mexico, Nicaragua, Panama,Paraguay,Peru,PuertoRico,Spain, he UnitedStates,Uruguay,andVenezuela.Includingdupli-cate titlesatdifferentnstitutions,he read-
ing lists featured3,480entries in the cat-
egory of novel, 2,309 dramaentries, 716short fictionitems, 1,685nonfiction temsand six entries in the categoryof "other."Thecount orpoetry,with ndividual oemsweightedequallywith ullvolumesofpoetryand ncludingduplicateitles, otalled6,490.
For the novel,the minimumnumberonareading istwas14,andthe maximumwas
171;the mean with standarddeviation orall 56 schools was 62? 34 (1S.D.).Theatreentries
rangedrom a low of five
playsto a
highof 107; hemeanwas41 ?26. For short
fiction, he minimum equiredwasone,andthe maximumwas35;themean was 13? 8.Nonfictionrangedfrom wo entries to 152.The mean was 30 ? 24. In the category"other,"he numberofentriesranged romone tofive,with a mean of 3 ?3. Forpoetry,the lowest numberof allentries(againwith-out distinguishing ndividualpoems fromvolumesofpoetry)was 14,and thehighestnumberof entrieswas 656;the meanwas116? 110.
Onlytwo works and two authors(allow-ingforacombination fworks)wereon 100
percent of the graduate reading lists.LazarillodeTormes1554)andDonQuijotede la Mancha 1605)were the books (Table3). The two authors were Miguel deCervantesandBenito PerezGald6s(Table1). Although the author "Anonymous"
achieved 100percent representationn ev-ery genre except theatre, we excluded this
author from our report for the obvious rea-
son that he or she was not a single entity.
The findings are presented in Tables 1-4.
Absence of a century or genre from these
tables indicates that no work in that period
This content downloaded from 157.253.50.10 on Sun, 21 Apr 2013 12:02:56 PMAll use subject to JSTOR Terms and Conditions
or of thattypereached50percentpenetra-tion of the reading ists;forpoetry,worksneeded to be cited in an identicalmanner
on this percentageof lists.Expanding he requirement or canoni-
cal statusslightly, oencompassauthorson
95 percentor more of the lists, yieldedan
enlargedcanon.Fernandode Rojas,Pedro
Calder6nde la Barca,Lopede Vega, and
CamiloJos6 Celafrom Spain,along with
RubenDarioandPabloNerudaofHispanicAmerica,all figuredon 98 percent of the
reading ists. CloselyfollowingwereTirso
de Molina from Spain and SpanishAmerica's GabrielGarciaMarquezon 96
percentof the lists.Featuredon 95percentwereJuanRuiz,FedericoGarciaLorcaand
Miguelde UnamunoromSpain,andC6sar
VallejoandJorge Luis Borges from His-
panicAmerica.Works with 95 percentor
greaterrepresentationwere, in addition othe twoalreadynoted: he PoemademioCid
(1100) on 98 percent, Rojas'La Celestina
(1502), Tirso's El Burlador de Sevilla(1630), and Calder6n'sLa vida es sueffo
(1635),on 96 percent;andRuiz'sLibrode
buenamor(1283)on 95 percent.No Span-ish Americanworkwas presenton 95 per-cent ormore of the reading ists.
By stretchingthe definitionof "canoni-cal" o 75percent, he canon ncreased ur-ther. NowaddedfromSpainwere nine ad-
ditionalauthorsof novels, 14 more poets,six addeddramatists,woauthorsof short
fiction, and two authors of nonfiction.Anumberof authorsachieved 75 percentor
greaterrepresentation n multiplegenresindependently:Cervantesas a novelistand
authorof shortfiction,Quevedoas a novel-
ist andpoet,GarciaLorcaas apoetanddra-
matist,andValle-Inclinas a novelist anddramatist.Withduplicates emoved, he 75
percentcanon totalled39 (Figure 1). Ca-nonicalSpanishworks alsowereexpandedby this measure, although by less than halfthe number of additional authors. Five nov-
ten canonicalHispanicAmericanworksbythis measure.As with Spanishliterature,this wasabouthalf he numberof additional
authorsincluded (countingreferences tomultipleworks).A totalof six novels,one
workof short fiction,one work of poetry,and two of nonfictionwere canonicalac-
cording o this standard Figure4).If atrulybroaddefinitionof canonicals
adopted-one by which a work or authorneed onlyappearon 50percentor moreof
the readinglists in this study-then our
canonencompassesmuchlargernumbers
of worksand authors. For
Spain,when
these new entrieswere addedto existingtotals,the yieldwas 21 novelists,25 poets,16dramatists,hree authorsof shortfiction
andsevenauthorsofnonfiction,or agrandtotal of 72 authorsin all categories.Thisnumbershrunk o 63whenmultiplegenrelistings of the same writerwere removed.Inadditiono thoseappearingarlier, hese
nowincludedanotherentryforLope(as a
poet as well as a dramatist),and also for
Unamunoand Azorin(bothas authorsofnonfictionas well as novelists). Quevedoearneda thirdentry(as an authorof short
fiction)andCervantes fourth(asa drama-
tist). Three of the authorswere women:Emilia PardoBazain n 77 percent,SantaTeresaon73percent,andRosaliadeCastro
on 52percentof the reading ists. Intermsofworks, herewere55,one of themfemale-authored.PardoBazain's 886Lospazosde
Ulloa was on 71 percent of the lists.Our canon of Spanish American authors
was enriched by opening the canon's gates
at 50 percent representation. A combined
total of 17 novelists, 18 poets, four drama-
tists, five authors of short fiction, and 12
authors of nonfiction were canonical by this
This content downloaded from 157.253.50.10 on Sun, 21 Apr 2013 12:02:56 PMAll use subject to JSTOR Terms and Conditions
THE CANON IN SPANISH AND SPANISH AMERICAN LITERATURE5
measure, adding up to 56, reduced to 49whenduplicate ntries orthe same authorswere removed. These included Borges,
Cortizar,Rulfo,Marti,andPaz ntwogenrecategories (Borges as an author of shortfiction and a poet, Cortaizar nd Rulfo asnovelistsandauthors of shortfiction,andPaz and Marti as poets and authors of
nonfiction),as well as SorJuanain three
categories(asapoet,dramatist, ndauthorof nonfiction). She and Gabriela Mistralweretheonly wowomen ncluded.Twenty-nineworksof literaturewere canonicalby
thismeasure,one writtenby a woman:SorJuana's1691Respuesta SorFilotea,on 64
Spainthe authorcanon was largestin thetwentiethcentury,with22 authors(12with75percentorgreaterrepresentation) Fig-ure 1). Followingwere the sixteenth cen-
turywith12(halfat 75percentormore)andthe nineteenthcenturywith 11 (ten at 75
percentor
above).For canonical
Spanishworks, a different distribution occurred.
Leadingwas the seventeenthcenturywith16 canonical itles, seven of which had 75
percentorgreaterpresenceon thereadinglists. The twentieth century was secondwith 14 canonical itles, onlytwoof whichachieved 75percentrepresentation n the
readingists. Next camethe nineteenth en-
turywith11titles,sixof them on75percentormore of the lists (Figure2).
TheSpanishAmerican uthor anonwasevenmoreheavilyweightedinfavorofthetwentieth entury,witha totalof29 authors.Thiswasmore than twice thatoftheprevi-ous century,the nineteenth,with 14 (Fig-ure 3). Few authorswere included from
preceding centuries: three from the six-teenthcentury, hreefrom heseventeenth,and none from the eighteenth centuryachieved 50 percentor morepresence on
the lists. Canonical works reinforced this
pattern,with 18twentieth-centurycanonical
titles (eight on 75 percent or more reading
lists), followed by seven from the nine-
teenth century (two on 75 percent or
above), three from the seventeenth centuryand one from the sixteenth, with none on 75
percentofthe reading ists (Figure4).At the oppositeend of the canonspec-
trum,our analysisturnedup many single
entries for both works and authors.Thecountoftitles hatappeared nlyonceinourdatabaseof 56 readinglists was 966.Thenumberofauthorsappearing nceonlywas229.
Discussion
Ourgoal in the present study was toachieveadetaileddescription four shared
literary anonatthis time.Recognizingheimportanceof required graduatereadinglistsas a measureof whatwe deemvaluable,weunderstood hat ananalysisoftheir con-tents wouldreveal he endproductofcanonformationn this country. naggregatewelooked to graduatereading ists from ead-
ing Ph.D.-grantingaculties ocodify helit-
erarychoices thatpredominatet aspecifictime.
Ourresults indicate that a substantialcanon does not exist inour field.ForSpan-ish literature,only two works and two au-thors are taughtto all graduatestudents.For SpanishAmerican iterature,no workor author earns unanimousapproval.Thecount of worksthat all graduatestudentscanexpectto have readin common, hosewith95percentorgreaterrepresentationnthe reading ists, consistedof sevenworksfromthe seventeenth centuryand earlier
panicAmerica,GabrielaMistralof Chileand SorJuanaInes de la Cruz of Mexico
were presenton 75percentormoreof thereadingists,butneitherreached80percentrepresentation. Again, they stood alone.The rest of our canon s exclusivelymale.
Our search for women parallels oursearch for commonground.Inboth caseswe areforced o lookata75percentor evena 50percentcanon o test ourassumptions,sincethe95percentcanonwasso small.An
expanded measure reveals that another
expectation is incorrect: the belief thatgreatestcanonpresencewouldcome fromclassicalperiodsofour iterary istory.Onlyat the skeletallevel of our seven-Spanish-itemuniversalworkscanon,on 95percentor more ists,did classicalperiodspredomi-nate. Contraryto common wisdom, thetwentieth enturyactually epresentsone ofthe areas of strongest agreementamongscholarswithregardto authors.ForSpan-ishliteraturehe numberof authorswith50
percentorgreater representationromthenineteenth and twentieth centuries wasmore than double the number of authorsfromthe sixteenth and seventeenth centu-ries. Even forworks,at the level of 50per-centandabove,the most recenttwo centu-ries hadslightlymore titles on the readinglists than did the two classical ones. For
SpanishAmerica the canon is even moreskewed toward the twentiethcentury:29
canonicalauthors came fromthis period,with 17 on 75 percentor more of the read-
ing lists; 14 authors came from the nine-teenth centuryandsix fromall other cen-turiescombined.SpanishAmericanworksshow the same pattern,with 25 titles fromthe most recenttwocenturiesandonlyfourotherworksinthe canon.Even nthis well-
representedand progressive period onlyonewoman,a Nobel-prizewinner, abriela
Mistral,appearedon half or more of thelists.
The limited canon that does exist is more
of an author canon than a works canon. For
both Spanish and Spanish American litera-
ture,it is twice as likelythat an authorwillbe required than a specific work by that
author.Spanish eadingists had39 authorson 75percentor more of the lists,countinganywork;ncontrast,he lists revealedonly
22 works of literaturewith this degree ofpenetration. SpanishAmerican lists fea-tured24authorswith75percentorgreaterdistribution,whileonlyten works reachedthis level.Byinference,agreements easierto achieve foran author hanfora particu-larworkthat best representshim orher.
An unexpected findingof this studyisthatforHispanic iteratures,ananalysisof
graduatereading ists also illuminates he
enigmaticprocessof canonformation.ForSpanishandSpanishAmerican iterature,canonformation ppears o takeplace onlyinmicrocosm;hecanon oreach nstitution
evidently is shaped independentlyat the
departmentalevel. The large numbers ofauthors and works thatappearonce onlyamong56readingists indicate hat nmanycases, individual convictions about thecanon arejustthat-the opinionof one lan-
guage facultyor
perhapseven one
special-ist at a singleuniversity.Unanimitys lack-
ing, and strong agreement is not wide-
spread.Thepresenceof authorsandworksthat are so obscureas to be unidentifiablefurthersuggests thatfacultymembers ex-ercise greatlibertyin compilinggraduatereading ists.Apossibletrendtoward ndi-
vidually-fashionedeadingistsprepared yone student, ypicallywithguidance roma
faculty ommittee,points oward ven more
variabilityn readingselections. These in-dividually-tailoredeadingists,usedbyfiveof 64Ph.D.-granting panish acultiescon-
tacted,are focused on one student nsteadof an entiregraduatecohort.They do nottransmita commoncanon.
THE CANON IN SPANISH AND SPANISH AMERICAN LITERATURE7
dantsrecognizedesignatedworksoflitera-ture (Herrnstein Smith 32). This rolereflectsthe originalmeaningof the canon,
whichwas "thechoice of books nour each-ing institutions."Bloom15).
Underlying his selectionprocessis thesharedassumption hat choices reflect lit-
eraryworth.Although "literaryvalue" is
contingent ndproblematic,nddespite hefact that other factorsalso affectcanon or-
mation,this perception s key. Selections,it is assumed, are based on evaluations.Assessments are of two kinds:formaland
informal(HerrnsteinSmith).Informalap-praisals ncludeorallyexpressed opinionsand coursesyllabi; ormalones encompassliteraryprizes,scholarlyattention hroughpublications,and inclusion in literaryan-
thologies.Contributingo thecomplexity fthe process is the fact that these assess-ments both reflect and establish literaryworth:"what recommonly akento be the
signsof literaryvalueare,in effect,also its
springs"34).Theconceptof"literaryalue"s elusive,andnouniversally-acceptedriteria xistbywhichto definea "masterpiece"eservingofimmortality.mportantttributes f suchaworkcan be identifiedbut notquantified.They includeawork'saesthetic attainment
(Hume), its ability to provide modelsorideals(Cook),and tsinnovationntermsofliterary istory(Bloom).Other mportant
aspectsare historical and
politicalsignifi-cance,communication f tradition,nsightinto he humancondition, elationotheory,and culturalcontent. Scholars'valuejudg-ments are alsoaffected,consciouslyornot,bycertainnonliteraryactorswhose impactis difficultto measure. Among the mostnotablearepolitics (Guillory), raditionora reactionagainst t (Gates),historicalandculturalcontexts (Lauter,Canons), he de-
mographicsofevaluators Lauter,"Race"),and the desire for inclusion of minorityvoices (Palumbo-Liu). Still another
unquantifiableactor s inertia.In Hispanic studies, research on the
canonhas focusednot on literaryvalueperse but on assessments that establishvalue.
Studies havebeen conductedprimarily y
thosewho work n the newer iterary ealmsof the nineteenthand twentiethcenturiesand literatureby women. Interest in this
areahas soaredsince the late 1980s:a 1996literature earchyielded52journal rticles,six bookchapters,andthree books on the
subjectof the SpanishandSpanishAmeri-can literary canon (MLABibliography).Sources of information hat have been uti-lizedbyscholars ncludepollsofprofessors'curricularelections (Holt),analysisof thecontents of literaryanthologiesand manu-als (Brown, Mullen "Emergence,"
Mancing,Perez),reviewsof trends n schol-arlypublications Brown,Debicki, Perez),and nvestigation f critical eception nthe
past (Gies, Gold). A recent study of thecanonwith relation o the changingcharac-teristics as well as the contents of antholo-
sessments of value with canonformationhave been advanced. One argues thatchoices ofworksare influencedbythe cur-rentsofscholarlydialoguenafield,andtheotherpositsthat he canon s shapedbythe
availabilityf texts
(Harris).Our
findingsindicatethat neither of these is the sole oreventheprimary eterminantf thepresentHispanic iterarycanon.The presumptionthatscholarlyactivitys amajormpetus orreevaluation of the canon has not beenconfirmedbyapositivecorrelation etweencurrentscholarshipandreading ist selec-tions (BrownandJohnson).Similarly,he
hypothesis that for Hispanic poetry thecanon s determinedargelybypresencein
study.Although hemajority four canoni-cal reading ist poets areon 40 percentormore of the 100 anthologiessurveyed byMancing,anumberofmajordiscrepanciesbetweenthis canonandthe 1986anthology
This content downloaded from 157.253.50.10 on Sun, 21 Apr 2013 12:02:56 PMAll use subject to JSTOR Terms and Conditions
canon also exist, indicating hatgraduatereading lists are influenced by othersourcesalso.
Less well understood thanthe factorsinfluencing literary selections is the dy-namicby which individualappraisalsde-
velopintowidespreadagreement hatcer-tain works fulfill agreed-upon"valuablefunctions" at a given time (HerrnsteinSmith). This evolutionaryprocess is asmuch a sociologicalas a criticalenterprise.Presumably t involvesnegotiatingagree-ment amongholders of differentviews, in
microcosm and in macrocosm: at a locallevel (amongmembers of a departmentalfaculty),andat acosmopolitanevel (amongmembersofacommondisciplineat arangeof institutions). t is probable hat the onlywaytoapprehendhe mechanismsofcanonformation s to studya specificmarkerofvalueovertime, to see whatchanges andwhat remains he same. This typeof inves-
tigation s in development or the graduate
readingists
analyzedhere.
Thedebateoverthe canonand ts forma-tion must be "asmuchpedagogicalas it istheoretical"Alberti ii).Ourenunciation fthe present canon raises issues of both
types.Intheoreticalerms, he most evidentand serious implicationof our findings isthatwemaynot have commonconceptionsof literaryvalue.Pedagogical mplicationsinvolve epercussionshatresult roma lackof consensus. By abdicatingall but a re-
duced curatorial ole, andby carryingoutan idiosyncraticnormativerole, our smallcanonmaynot serve thegraduate tudentswhose reading ists we havecompared.
These discoveriesmay challengeus toreexaminethe foundationof commondis-course in Spanishand SpanishAmerican
literary tudies.One leaderwith a half cen-
turyofexperience nourprofession,he lateRobert G. Mead, Jr., urged us to do pre-
ciselythis. Ina letter to the authors ollow-ingthepublication f ourpilotstudyon thetwentiethcenturynovel,1ProfessorMead
serted,"there s too much 'freewheeling,'personal bias in choice of texts ... in short,
unnecessary onfusion stoclearstandardsandgoals in the teachingof literature. t isno wonder, then, that a recognizedcanon
has notyet emergedin regard o literaturein Spanish..."(Mead).
Whetherstemming romunchecked n-dividualautonomy,differences of critical
opinionor some other cause, the fact re-mains that at this time there is very little
agreementon the subjectofrequired ead-
ing inour field.Knowing his,we are facedwith choices. The theoreticaldecision wemustconfront s the questionof whetheror
not we wanta fixedcanon,andwhatvaluesshouldshapeits formationfthe answer saffirmative.Pedagogicalconcerns involvethe issue of suggestingstandards orpost-graduateeducationnSpanishandSpanishAmerican iterature. f we choose to adopta shared canon,what authorities shouldoversee its construction?By what criteriashould selections be made? And what
type-monolithic, "core" with choices,worksand/or authors-should it be? Forourselvesandoursuccessors,discussionofourcommoncanonbelongsatthetopofour
THE CANON IN SPANISH AND SPANISH AMERICAN LITERATURE19
40
ii I
i::ill
MY
..........
Figure3.
RepresentationfSpanish
Americanauthorsonthe 56graduatereading
ists
by century.Authorswho havepublished nmultiplegenresareenteredonly ntheir most
prolificgenre.Twenty-fouramesappearon75percentor moreofthelists,and26appearon 50 to 75 percent.
' i
IM. .. .
Figure 4. Representation f SpanishAmericanworks of literatureon the 56 graduatereading istsbycentury.Tentitlesappearon75percentormoreofthelists,and19appearon 50to 75percent.