Community Essay

IOURNAL OF EDUCATION, Volume 171 Nurnber i, i989 LITERACY DISCOURSE, AND LINGUISTICS: INTRODUCTION fames Paul Gee Department of Lingtistics, Universitl, 6f ;\outhern California, Ins Angeles What I propose in the folio'adng papers, in the main, is a way of talking aboutliteracy and linguistics. I believe that a new field of study, integating "psychd' and "socid' approaches to language irom a variety of disciplines, is emerging, a field which we might call liceracy studies. Much of this work, I think {and hope}, shares at least some of the assumptions of the {ollowing papers. These papers, though written at different timeg and for diffierent pur- poses, are, nonetheless, based on the ciaim that the focus of literary studies or applied linguistics should not be langriage, or literacy, but social practices. This claim, I believe, has a number of socially important and cognitively inter- esting consequences. "Langu.agd' is a misleading term; it too often suggests "glamrnarl' It is a truism that a person can know perfectiy the glammar of a language and not know how to use that language- it is not just r,r'&a; ;'ou say, but how you say it. If I enter my neighborhood bar ancl say to nry tattooed drinking buddy, as I sit dowrr, "May I have a match pleaselj' my grarnmar is perfect, but what I have said is wrong nonetheless lt is less otten rerrrarked that a person could be able to use a language perfectly and str-l? nqt make sense. It is not lust how you say it, but what yon are and cJo when yo say ir:. If I enter my neighbor- hood bar and say to my drinking buddy, as sit down, "Gime a nratch, wolldyaij'while placing a napkin on the bar stool to avoid getting my newly pressed designer ieans dirtli I have said the right thing, but my \ayingdoing" r:ombination is nonetheless all wrons. I am deeply indebted to Candy Mitchell ior editing this collecl ion of papers, and to fim OBiien lbr copy€diting ihe papers appearin3 here {or the first tim:. The following people are responsible {they may be aghast to hear) Ior having helped to lead me to the views I bold. First, a set of people whose writirqgs harre irupired me: ll'allace Cha{e, Michsel Cole, k'lul Gumperz, Shirley Brice Heth, Dell Hymes, William hbov, Roger and Su:rurne Scollon, Bri;ur Street, Gordon ldblls, and fim \{'ertsch. Second, a group of people not only whose writings hare inspired me, but whose discus- sion of the issues in these paners witb nre, a.lrell as whose friend-ship, has left me aln"ays in their debt: Elaine Andersen, Maria Brisk, Chip Bruce, Courtney Cazden, David Dickenson, Stert Krashen, Stele Gordon, Stere Griffl'r, Henry Girorrx, Donel lo Macedq Sarah Michaels, Bea Mikulecky, Eliiot Mishler, Candy Miti:hell. Catherine Snow, and Dennie 14blf. These papen ulti- mately all hale their origin in the kindness that Sarah Michaels and Courtney Cazden extended to me when I first arrived irr Boston try rnviting mc to ta-ke an interest in their concems. lor-rrnal o{ Education, Vo}umt l,tl, i".lrrniber l. t989. (:-, Trustees of Boston University F. Niyi Akirrnaso and Cheryl Aiirctuti1lf9!)'. nteylt "simulated iob interviewg' from two lveifare mothers in a CETA job training Pro8ram' r ne ;;-;;, asked whether she has ever shown initiative in a previous job,, r""p"rarhit"ff, y-es, therds this Walgreerls Agenlv, I *"tk"1i:,1T-:::17 opJr"tor, OK. A;d it lvas a snow stonn, OK' And it was usually srx people workir/ in a group. . :' *J* forth {p' 34}' This woman is simply using the ;;"s t*;- {ihe wrong 'tiiabcrffor this tvp".of (t*dttt:;tl":,tlT:::ltl It's a fJrfectly good grarnmar {diatect}, it iust wodt get you th-ls type ot loD m BOSTON L]NWERSIT}- *"k ."rr"t, ,o "discoursd' is part of "Disc"*d\: 'Di5orrrs9!e1e]^/:l)nlg!!g!{}g in thelrorlG; !hef'are this type of societl'. The second woman {the authord "succesd' case} respondt- 19.t .t*"]?t question by saying: ". . .I was lett alone to handle the office' ' ' ' l clrclrIt reaxy frarr. a tot of "*p.ii.tt""' But I had enough experience to ded with any situa- ;;a; ;;.'up. . . and those trrat t coutdat handle at the time, if there-was someone rn-ho had more erperience than rnyself, I asked qlSstronsto find out what procedure I would use' if something came up encl lt I drddt Know wno io ,""fiy So to, I would iot it down . . . o', a pie"e of paper, so ttrat I rvouldr/t forget itr"i if iiy""" that was more qualified ihanrnysclf, I could alf them about i .rra n *:l *ruid go about so^lving it. So I feel I'm capable of hrndling iust JU""i*y .i*ation,"whether it's on hy ovrn or rurder superviti"d,' !p , 111}^r].t woman ha^sr1t got a real problem with her grammar (remeab:r,thls ls speecn' not wziting), nor is there any real problem with the use to which she puts tlat srasrmar, but she is expressing the wrong vaiues' She views beingleft in charge f.;;;;il;-;;;; ;'a';;;sion, nairelv., supervision bv "other peopldd' knowledg"an

I belierc that efiy socially use ful definition of "litereqy'' must be couched in terms of the notion of Discourse. Thus, tr define "literacy'' as the mastery of or fluent control over a secondary Drscourse. Therefore, literacy is always plwal: fiteracr'es {there are many of them, since there arc many secondary Dis- courset and we all haw some and hil to haw others). I{ vre wanted to be rather pedantic and literalistig then r+e could define "literacy'' as "mastery of or fluent control o\Er secondary Discoursas lrra./v:ngpnnl' {which is alrnost all of them in a modem society|. But I see no gain foom the addition of the phrase "involv- urg printi other than to assuage the feelings of people committed {as I am notf to reading and writing as decontextualized and isolable skills. Wb can talk about domina nt litetacies and nondominailt literacies intemrs of whether they involle mastery of domin'nt or nondominant secondary Discourses. We can also talk atrout a literacy b engliberating {"powerful"f il it can be used as a "meta-languagd' ia set of rneta-words, meta-values, meta-beliefsf for the critique ol other literacies and the way they constitute us as persons and sit- uate us in society. Liberating literacies can recon-stitute and resituate us. IvIy definition of "literad' rnay seem imrocuous, at least to someone al- ready corwinced that decontextualized views of print are meaningless. Nonetheless, several 'theorems'' follow from it, theorems that have rather direct and unsettling consequeirces. First theorem: Discourses {and therefore literaciesl are not like languages in one very important regard. :.-rmeone can speak English, but not fluently. However, someone carinot cngage in a Discourse in a less than fully fluent manner. You are either in it or vodre not. Discourses are connected with dis- 10 BCSTON I.INIVERSFY plays of an identit7j failhg to fully display an identity is tantarnount to an- nor-rncing 1cu donlt harc that identity, that at best yor.rlre a pretender or a beginner. Very often, learners of second languages "fossilizd' at a stage of de- velopment significantly short of fluenry. This cadt happen with Discourses- If yor/w fosilized in the acquisition of a Discourse prior to full "fluel{' {rtta are no longer in the prcrcess of apprenticeship!, then your verv lack of fluenry marks you as a nonlmembe.r of *re group that controls this Discourse. That is, you dodt harrc the identity or social role which is the basis for trhe existence of ttre Discourse in the fust place. In fact, tle lack of flr'rency may very well mark you as a pretender to the social role instantiated in the Discourse {an outsider with pretensicns to being atinsiderl. There iq thus, no t'orkable "affirmative actiorl' for Discourses: you carlt be let into the game after missing the apprenticeship and be expected to have a tair shot at playing it. Social goups will not, usually, give their vrcial goods- whether these are status or solidarity or both-to those who are not "natiw:d' or "{luent lrsery' {though "mushIake," discussed below., may sometimes pro- vide a way for non-irritiates to gain access|. While this is an empirical clatrn, I belierie it is one rastly supported fu the sociolinguistic literature {Milroy, 1980, 1987r Milroy &, Milroy, 19851. This theorem {that there are no people who are partially literate or semi- literate, or, in any other r'ay, literate but not fluently so) has one practical consequence: notions like "functional literad' and 'tompetenry-based literad' are simply incoherent. As far as literacy goes, there are only "fluent speakers' ard "apprenticed imetaphorically speaking, because reme:nber, Dis- courses are not iust rvays o{ talking, but ways of talking, acting, thinking, valu- ing, etc.).

Second theorem: Prirnary Discourses, no Inatter whose they are/ can never really be liberadng iiteracies fur a literary to be liberating it must contain both the Discotuse it is going to E'ritique and a set of meta-elements {language, words, attitudeq values| in terms of which an analysis and criticism can be carried out. Prirnary Discourses are initial and contain only themselves. They can be embedded in later Discourses and critiqued, but they can nelrer serve as a meta-language in terms of which a critique of secondary Discourses can be carried out. Our second theorem is not likely to be wry popular. Theorem 2 says that ali primary Discourses are limited. 'Liberatiod' {"powe1'}, in the sense I am using the term here, resides in acquiring at least one more Discourse in tenns of which ow own primary Discourse can be analyzed and critiqued. This is aot to say that primary Discourses do not contain critical attitudes and critical language {indeed, nuny of thern contain implicit and explicit racism and classisml. It is to say that they cannot carry oro;t ert authen;tr'c criti- cisrn, because *rey cannot r,erbal ize the words, actt values, and attitudes they use, and they cannot nobi-lize enplicit meta-knowledgc. Theorem 2 is quite traditional and conser tive-it is the analogrre of SocrateCs theorem that the 1\'{r d\*Jsr I f,\i' I Nr1 JOURNAL OF EDUCAfiON, -',blume 17i Nurrrber l, 1989 uncxamined hfe is not *'orth living. tnterestingly enough, \'ygotsky (1987, chapter 6| comes very closely to statlqg this theorem explicitly. Other theorems can be deduced from the theory of literacy here de- veloped, but these two shor.ild make clear what sorts of consequences the theory has. It should also make it quite clear that tJre theory is not a neuttal meta-language in terms of which one can argue for iust any conclusions about literacy. Not all Discourses involve w::ithg or readiag, dtough many do- However, all vrnting and readir.g is embedded in some Discc,urse, and that Discourse always involves more than writing arrd reading {e.g., ways of tdking, actinS, valuing, and so fonh). You cannot teach anyone to write or read outside any Discourse (there is no such thing, unless it is cdled "moving a per/' or "typ- ing" in the case of writing, or "movilg onds lipd' or "mouthing wordd' in the case of reading|. Within a Discourse )'ou are always teaching more than writing or reading. When I say "teacH' here, I mean "apprentice someone in a master- apprentice relationship in a sccial practice {Discourse} wherein you scaffold their growing ability to say', dc, value, beliele, and r;o forth, within thet Dis- course, through demonstratirg your mastery and supporting theirs e'ven when it barely exists [i.e., you make it look as if they can do what they really carlt do)1' That is, you do much the same thing middleclass, "super bab/' produc- rng parents do when they €o books' with their child1sn. Now, there are many Discourses connected to schools {different ones for different rypes of school activities and dilferent pans of the curriculum) and other public institutions. These "middle-class mainstreant'' sorts of Discourses often carry with them power and prestige. It is often ftlt that good listeners and good readers ought to pay attention to maaning and not Socus on the petty details of mechanicg "correctness/' the superficial fuatures of language. Un- fortunately, many middle-class mainstream status-giving Discourses often do stress superficial teatures o{ lalrguage. \,!try? Preciselybecause suchsuperficial f':atures are the best test as to whether one was apprcnticed in the "righf place, at the "righ/' tirne, with the "righf' people. Such supe.rficial batures are exactly the pans of Discourses nrost irnpervious to owrt instruction and are only fully mastered when everytJrilg else in the Discourse is rnastered. Since these Dis- courses are used as "gated' to ensure that the "righC' people get to the "righf piaces in our society, such superficial features are ided. A person who writes in a petition or office me'mo: "If you cancel the show, all the performers would have did all that hard work for nothingl'has signaled tbat he or she isdt the "right sort of persorf' (was not frrlly acculturated to the Discourse that sup- ports this identity). That signai stays meaningfui long after the content of the memo is iJrgotten, or even when the content r^/as of no interest in the first place.

Now, one can certainiy encourage students to simply "resist" such "superficiai featurc.s of languagel' And, indeed, they will get to do so from the ll BOSTON UNTVER.SITY bottom o{ society, where their lack of mastery of such superficialities was 1t:T: p pl"o them anyway. But, of course, the problem is that such ,,super_ hcralitieg' cannot be taught qr a regular classroorn in any case; they car{i be "picked u1' later, outside the f'll context of an early apprenticeship(at home and at schoollin "middleclass-likd' school-based *lo "f a"ing "rra tii o. tu", is preciselywhy they work sowelles "gatesf'This is alsoprecisety tire tiateay of E' D. Hi:sch, |rls much-talked-aboui book culrarai tit"-or i;ssi1.*rni"i points out that without having mastered an extensive list of Lir.i"riii", p.o- gte. ;an be Janaoftenaref oaluded from "goodC, contron"d by ;rr-.r:l-rrt;;"p, m the society. Hirsch is a'rong-in thinking that this can be taught {in a=ciass- room.of all pla""ul) ."p"tt.ft9T the sociariyiituated practices thai rhese groups have rncorporated into their homes and daily lives. There is a real coritradic- Ionh9f,-engwe ignore it at theperil of ourstudents andour** "ga f"liV (no middle-class 'tuper bab'y', producing parelrs ignore it!. !*y""4 slenging thesocial structurg is there much hope? No, there is 39t. so -we berter get on about the process of changing the social structnre. Now, whose iob is t}'at? I would say, people who have bien allotted the job of teaching Discourses, 6or example, English teacherq language teachers,'corn- position teacherq TESOL teachers, studies-skills teachers. we can pause, also. to remark-on the paradox that e en though Discourses "rt ,,ofb" overtly taughg and carrnot readily be mastered late in the game, the university **,. teachers to olertly teach and wants students to demonstrate mastery. riachers of Discourses take on an impossible iob, allow themsehrs to be eviuated on how wellthey do it, and accep {airly low status all the whire for doing ir. So what can teachers of Discourses do? WeIl, tlere happens to i-. * advantageto failing to master mainstream Discourses, that is] ihere is an atl- vanuge to being socially "maladayrtedl' \{hen we have really mastered arrythrng {e.g., a Discoursef, r*e have little or no conscious awareness of it lindeed, tikJ d3*"*s, Discourses wouldrlt work if people were consciously aware of what they were doing while doing itl. uowevei when we come across a situation where we are unable t* accommodate or adapt (as many mfurority students do on being faced, late in the game, with having to acquire mainstrearn Dis- courses), we becorne consciously aware of wh"t w" at tryorg to do or ,r. b"org called upon to do. Let rne girre an example that works similarly, that is, the caG ot classroorn second language leaming. Almost no one really acquires a second laqguage in a classroom- Howener, it can happen that exposrire to another languap, hevins to uanslate it into and otGrwise rehL it t. 1=o, u*r, fanguage, c:ut cause you to b9c9me consciously ar,r.are of frow yo* fioi language ramrks {horv it means}. This "meta-lorowledgd, can actualry make you better able to manipulate your first language. . Vygotsky {1987} says tJrat iearning a foreign lauguage %llows the chiid to *9:ry*ld Fgnative language as a single instantiatiorrif "hrrgrriJi; t;;""1, {p' 2221. Anti here we have a crue- crassroom instruction 1in 6nguage, co;- |OURNAL OF EDUC.{tr*ION. Voiume lZi Number l, 1989 -tJ position, study skillg l{Titirrg, critical thinking, content-based literacy, or *t "t"*.t "rtr iead to metak.to-l.dg., to seeing horr the Discourses ),rou harE already got relate to those you are atteiiipting to acquire, and how F o* yo{ are tryirig to acquire relareto self and society. Metaknorvledge is libe.rationand po'*a UJ"".6.it leade to the ability to manipulate, to anallze, no resist while advanclng. Such metaknowledge caa make "maladapted' students smarter than %daited , ones. Thus, the liberal classroonr that arpids oven talk of form and supei{icialities, of how thiags work, as well as of their socio-cultutal- politlcal baslq is no help. Such talk can be por*erful so long as one never thinks that in talking about gr-amr raq fcrm, or superficialities one is gettlng Peo_Ple to actually aJquire Dit"onts"r {or laaguageq ior that matter}. Such talk is always political talk. il.ri, ttt bg qn"stion: If ore cannot aequire Discogrses save ttroug5 actirie social piactic., "''a lt is di{ficult ta compete with the mastery o( those ad- mitted early to the game when one has entered it as late as higlr school or col- lege, what can be done to see ro it rhat meta-knor,vle@ *d resistance are colupled with Discourse developmentS The problem is deepenedtry the -fa9t thai true acquisition oi many mainstrerm Discoulses involws, at ieast_while betng in them, active complicity with values that conflict witt onds home- and Jommunisy-based Discourses, especially for masy r pmen and minorities. The question is too big for me, but I harre two views topgsh-nonetheless' First, true acquisition (wbich is always full fluency) will rarely if oier happen. Even for "qtitti"s close to acquisition to occur, classrooms must be active apprenticeships in'hcademiC social practices, and, in most cases, mu* co-n' n".t *ith thise social practices as they are also carried on outside the 'tompositior/' cr "languagd' clasg eisewhere rn the Unirarsity' *con{ though mre acquisition is probably not possible, "mushfrkd' Dis- course is possible. Mack {in press) detines "mushfakei' a term hom prison culture, ai making "do wiih somethrng less when the red thing is not avail- able. So when prison inmates make hats {rom underwear to protect their bfil from lice . the hats ale mushfake. Elaborate craft items made ftrom used vooden rnatch sticks arc another exanrple of mushfakej' "Mushgke Discoursd' means partial acquisition coupled with meta-knowledgp and strategies to "make dd' fstrateges-ranging from alwal's having a memo edited to ensure-no plwal, pos- !"sLrJ, atrd third-person ,d, agreement errors to active us€ of black culture skils at "psyching out'' interviewe$, or to skategies of "rising to the meta'lerel in an intiwiew so the interviewer is thrown off stride by having tlre rules of the game implicitly refurred to in the act of carrying them outf' - , ;'M*hI"k ;' toistance, and meta-lcnon'ledge: this seems to me like a good combination for successful students and successful social clrange. So I propose that we ought to produce "mush{akingi resistitg students, fulI of meta- knorvledge. but isrlt that to politicize teachingl A Discourse rs an integration of sayingfdoing, and va.iuing, and ail socially based valuing is political. All suc- BOSTON LINTVERSTT "' I+ cessful teachin'g, that is, teaching that inculcates Discourse and not iust con- tent, is political. That too is a trursrn' As a lirguist L* pt"";; ;;sted in.the functioning of language m Discourses and u,",""i"tlnia?r[i:n"iti"" + this sort of linguistics is how language-within-urr**'"lii ;t'q:,li,id 6 ;iarv situated apprenticeships) arrd horr the languages r'orn aiffetittt Diicoursgs transfur into' intertete wt -rr' and otherwise un "o""'i"#;,h;;;-f"t* the linguistic. ,"-:*: "1^**l: L"i.ai* "tJio i*"t'"u*l-u"t"* gtottpt rl.sgci;w' To see what is at stake here, l will briefiv ai'"t'"I ;;;;J which clearlv brings out a host of important issues in *ris d"J*"i"' ittlt tto' with an explanation oi its context' i#;; b"il. n " ,"*.i, J"msrcated in terms of "lines" and "stanzas;' umts *riJ"u i Lri"* are the basis oi speech: CONTEXT OF TEXT: A yourg middiell-ass-mottter rerularlv reads storybooks to both her s- t ,a t v""tJia*al"friJ"' tieiJ-v,e"t--otd h"d had a birthdav partv' whichhadhad so*t p'oUit* i'ithe next fw {ays the s-year 9ld'has told seve:al ;il;t#;bJi*; ui*fr*"v p*iy' "pon-s'tttt^e'ents in the language ol her primary Discourse "v"tt*' A i* tiiq" -l'ater' when the mother rvas reading a story- book to her T-ve""'o, ti'? il>i"'-Ji Gaittt t '*ted to "read" {she could not cie' code), and pretena,,i "'L-'iffi*" *C^yP:::]hng what had.haipened a; il.iili"f,iiv p"tty Ht' oifoitt"l'ti"rnpt at this was not very good' but eventual- lv after a fev*'ui"', **ti#iJJ*-th'the.mo'ht' te"dins to the other girl'-the 'd#Ji_i,iJii"a"*a ,i,. ^r.irir*Grlory-*hi.h ir ttot {iust} ul the languase ot ner Primary Discourse sYstem: STAI ZA ONE (Inttoductionl l. This is a story ;. G;G;kid'*how' once friends 3. But got into a big f$ht 4- And wlere not STANZA TWO {Frame: Signal hng of Genre} 5. You can read along i'n your storyDooK 6. I'm gorura read aLoud {story-readi4g prosody irom norv onl STANZA THREE ll itle) 7. "F{orr'the FrienJs Cot Untriend" SI\NZA FOUR {Settir^g,: tntroduction of CharactersJ 8. Once upon a tlme ihE" *"t *uee boys.'n three girls - -L ^ ^:,r^ ;. iG.;; "amed Betty Lnu, Paliis' and Parslun' were the Srrrs I0. And Michael, Jason' and Aaron wcre tne Doys I l. TheY were kiends STANZA Fl\€ {?roblem: Sex Differences} 12. The boys would play lianstormers i5. A;d;hA girls would play Cabbage Patches r;ilil