Text and Function

Author(s): Yu. M. Lotman, A. M. Piatigorsky and Ann Shukman

Source: New Literary History, Vol. 9, No. 2, Soviet Semiotics and Criticism: An Anthology
(Winter, 1978), pp. 233-244
Published by: The Johns Hopkins University Press
Stable URL: http://www.jstor.org/stable/468572 .
Accessed: 14/06/2014 06:08

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].

The Johns Hopkins University Press is collaborating with JSTOR to digitize, preserve and extend access to
New Literary History.

http://www.jstor.org

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
 Text and Function*
Yu. M. Lotman and A. M. Piatigorsky

HE AIM of thisarticleis to examinetwoconceptsfundamental

forthe studyof culture,"text"and "function,"in theirmutual
relationship.The concept of text is defined in accordance
withan articleby A. M. Piatigorsky.1Particularattentionis given to
such featuresof a textas itsexpression[vyrazhennost'] in a givensystem
of signs-its "fixation"-and its capacity to perform in a certain
relationship(in the systemof signalsfunctioningin a community)"as
an elementaryconcept."2The functionof a textis defined as itssocial
role, its capacity to serve certain demands of the communitywhich
creates the text.Thus, functionis the mutual relationshipamong the
system,its realization,and the addresser-addresseeof the text.
If one considers three categoriessuch as text,functionof the text,
and culture, at least two general approaches are possible. With the
firstapproach, culture is seen as a totalityof texts;in relationto the
texts functionthen appears as a kind of metatext.With the second
approach, culture is seen as a totalityof functions,and the text will
then be seen historicallyas derivingfroma functionor functions.In
this case the text and the function can be seen as objects to be
investigated on one level while the first approach necessarily
presupposes two levels of study.
Before making an investigationof this kind, however,it should be
remarked that,in principle,we have to do with differentobjects of
study.Culture is a syntheticconcept, the definitionof which,even a
working one, is fraught with difficulties.Text may, however, be
defined-if not logically,at least for workingpurposes-by pointing
to a concrete object having its own internalfeatureswhich cannot be
deduced from anythingelse apart from itself. Function, however,
seems to us to be a pure constructand here one in the sense of whicha
giventextmaybe interpreted,or in relationto whichsome featuresof
a textcan be examined as featuresof the function.

The concept of textin the sense in whichit is used in the studyof

culture is differentfromthe linguisticconcept of text. The point of

* First sistemam[Abstracts
published in Letnyayashkolapo vtorichnym
modeliruyushchim
for summer school on secondary modeling systems](Tartu, 1968), pp. 74-88.

Copyright? 1978 by New Literary

History,The Universityof Virginia

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
234 NEW LITERARY HISTORY

departure for the cultural concept of text is preciselythat moment

when the fact of linguistic expression ceases to be perceived as
sufficientforthe utteranceto become a text.As a consequence of this
the masses of linguisticmessages circulatingin the communityare
perceived as nontextsagainst the background of which stand out a
group of texts which reveal features of some supplementary
expression significantto the given systemof culture. Thus at the
momentwhen writtenculturearose, the expressionof the message in
phonological units began to be taken as nonexpression. To it was
opposed the graphic fixationof a certain group of messages which
were accepted as the sole ones, fromthe point of view of the given
culture,to exist. Not everymessage is worthyof being writtendown,
but everything written down takes on a particular cultural
significance,becomes a text.(See the identificationof writingwith
sacrednessin termssuch as "writing"[pisanie]and phrasessuch as "for
it is written,""to speak from the writing"[pisano bo est',glagolatiot
pisaniya]which were common in Russian medieval literature.)
Connected withthisis the scale of culturalvalues where the written
texttakes the highestplace afterthe supreme divinity.Thus in many
ancient and medieval cultures, religious initiationis initiationto a
writtentext(the permission to read a certain text)-as in lamaist
Buddhism; in more ancient cultures,however, initiationis the oral
communication ofthemeaningof a writing-as in the Upanishads.
The opposition"oral-written"may correspondin some culturesto
"not published typographically-printed," and so on. Expression may
also appear as the demand for a certain long-lastingmaterial.What
is engraved on stone or metal is considered a "text" as distinct
from what is writtenon perishable materials-the antithesis
"durable/eternal-transitory"; what is writtenon parchmentor silk as
distinctfrom paper-the antithesis"valuable-not valuable"; what is
printedin a book as distinctfromwhat is printedin a newspaper or
what is writtenin an album as distinctfrom what is writtenin a
letter-this is the antithesis"intendedto be preserved-intended to be
destroyed": it is indicativethat this antithesisworks only in systems
where lettersand newspapersare not intended to be preservedand is
not operative in opposing ones.
The particular"expression" of a cultural text that distinguishesit
fromgeneral linguisticexpression is not, however,found only in the
various forms of written culture. In a preliterate culture the
distinguishingfeatureof a textwas a supplementary,supralinguistic
organization on the level of expression. Thus in oral cultures,
texts-legal, ethical, and religious, and those containing scientific
knowledge about agriculture,astronomy,and so on-were endowed

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
TEXT AND FUNCTION 235

necessarily with a supraorganization in the form of proverbs,

aphorisms with definitestructuralfeatures. Wisdom is not possible
except in the formof a text,and a textimpliesa certainorganization.
Hence, at thisstage of culture the distinguishingfeaturethatdivides
truthfromnontruthis a supralinguisticorganizationof the utterance.
It is indicativethat with the change to writtenand then to printed
stagesof culture,thisdemand fellinto abeyance (see the renderingof
the Bible into prose in the European cultural tradition) and was
replaced by others. Observations about preliterate texts acquire
furthersignificancewhen the concept of text in modern culture is
analyzed, since as a result of the development of radio and other
speaking mechanisms, the obligation for a text to be graphically
expressed has again been lost.
When classifyingcultures according to the feature distinguishing
textfromnontext,it should be rememberedthatthese concepts may
be reversibleas far as the limitin each given case is concerned. Thus,
withthe opposition "written-oral"one could imagine both a culture
whereonlywrittenmessages are considered textsand a culturewhere
writingis used for everyday,practical purposes while texts (sacred,
poetic,ethico-normative, etc.) are handed down as fixed norms. The
utterance"He's a real poet; he's published"isjust as possibleas "He's a
real poet; he's not published." See Pushkin'slines:

foeto slavery,
Radishchev, avoidedthecensor,
And Pushkin'sversesdid notappearin print.3

If youstartedto writestupidly
Then youwouldsurelyslip
Throughour tightcensorship
As youwouldenterthekingdomof heaven.4

Appearing in printis a criterionboth when it is said-"If this were

valuable (true, holy,poetic), it would have been printed"-and when
the opposite is affirmed.
In relationto a nontext,a texthas supplementarymeaning. If one
compares twoutterancesidenticalon the linguisticlevel,of whichone
fulfillsthe concept of a textin the systemof the givencultureand the
other does not, then it is easy to define the essence of textsemantics.
One and the same message (a written agreement, for example,
affirmedby oath, or simply by a promise, coming from a person
whose utterances,thanksto his positionin the community,are texts,
or from a simple member of the community,and so on) will be
differently evaluated from the point of view of its authoritativeness

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
236 NEWLITERARY
HISTORY

eventhoughlinguistically thereis coincidence.In thespherein which

theutteranceis receivedas a text(a poemis notreceivedas a textfor
the definitionof the scientific, religious,or legal positionof the
community, but is taken as a text in the sphereof art),it has the
meaning of truthfulness ascribed to it. An ordinarylinguistic
communication,well-formedaccording to all the lexical and
grammatical rules-thatis,"correct"in thelinguistic sense-and not
containinganythingcontradicting the possiblein its content,may
nonetheless turnout tobe a falsehood.But thiscannothappenwitha
text.A falsetextis as mucha contradiction in termsas a falseoath,
prayer,or law. It is nota textbuttheviolationof a text.
Since a texthas truthfulness ascribedto it,the existenceof texts
presupposes"a textpointofview"-thepositionfromwhichthetruth
is knownand fromwhichfalsehoodis impossible. A description ofthe
textsof a givenculturegivesa pictureof the hierarchyof these
positions.There are cultureswithone pointof viewcommonto all
texts,or witha hierarchyof pointsof view,or witha complex
paradigmof them,to whichwillcorrespondthevalue relationships
betweentypesof texts.
If one admitstheparallelism oftheoppositions "text-nontext" and
"truth-nontruth," theneveryculturecan be ascribedto one of two
typesaccordingto howitviewsitself in a historicalperspective (given,
of course, the particulartime-sectionin whichwe examine the
culture). "Culture of the closed type" sees itselfas continuing
according to tradition,from the time ("time" is here meant
conventionally) when thereexisted"fullnessof truth,"i.e., a "full
text";while"history" is thehistory of thegraduallossof thisfullness
whichlies at the sourcesof the culture."Cultureof the nonclosed
type" sees itselfas arising"fromzero," "fromnothing,"and as
graduallyaccumulating elementsof"truth"whosefullness is believed
to lie in thefuture.
Cultureof the second type,whenobservedfromoutside,seems
morefunctional, and cultureof thefirsttypemoretextual in theliteral
senseof theword.It is obvious,moreover,thatthesamevalueswill
occupydifferent placesin thevaluescalesof thesedifferent typesof
culture:
(1) In culturesof the "closed type" (for example, in Tibetan
Buddhistculture)the textis significant ("sanctified")becauseit is a
text.
(2) In culturesofthe"nonclosedtype"thetextissignificant because
it has a particularmeaningwhichdetermines itsfunctional value.
(3) Following from what has been said, in cultures of the second
type there occurs (or "takes place") the absolutizationof historical

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
TEXT AND FUNCTION 237
experience, while in the first type there is the absolutization of
prediction(and hence eschatology).

The separatingout of a certain quantityof texts fromamong the

mass of general linguisticmessages maybe seen as a featureof culture
that appears as a particular type of self-organization by the
community.The pretextstage is the preculturestage. And the state
where all textsrevertto theirlinguisticmeaning only corresponds to
the destructionof culture.
For the studyof culture thereexistsonly those messages whichare
texts.All the others,as itwere,do not exist,and the investigatorleaves
them out of account. In this sense it may be said that culture is the
totalityof textsor one complexlyconstructedtext. If the structural
code of the culture to which the describerbelongs is applied to the
material being studied (the study of ancient culture by our
contemporaries,or the studyof the culture of one social or national
type from the position of another), this may lead to the shiftof
nontextsinto the categoryof texts,and vice versa, according to their
distributionin the systembeing used for the description.
A conscious break witha certaintypeof culture,or ignoranceof its
code, may appear to be a denial of the systemof text meanings
inherent in that culture. Only their content as general linguistic
messages is recognized, or if there are no messages at this level,
"nonmessages." For example, the sixteenth-century heretic Feodosiy
Kosoy refused to see in the cross a symbol with text (or sacred)
significance, and ascribed to it only the meaning of the primary
message-the weapon of execution. "Kosoy says that those calling
themselves orthodox worshipped wood instead of God without
findingout what was pleasing to God. Only theydo not understand,
onlytheydo not wantto understand,althoughtheycould understand
by themselves" [typical here is the denial of the "conventional"
meaning given by the cultural code and the acceptance of the
"natural," linguisticmessage, "although they could understand by
themselves"],"for if someone beats to death the son of another man
could that man love the stickwith which his son was killed? In the
same way God hates the cross because his son was killed on it."5On
the other hand, knowledge of the culturalcode systemresultsin the
linguistic meaning of the text receding to the background and,
indeed, perhaps not even being perceived,being so overshadowed by
the secondarymeaning.The obligationto be comprehensiblemay not
be applicable to texts of this type at all, and some such texts may
successfully be replaced in cultural usage by their conventional
signals. In Chekhov's story The Peasants the "incomprehensible"

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
238 NEW LITERARY HISTORY

ChurchSlavoniclanguageis perceivedas a signalforthechangefrom

an everyday communication (a nontext)to a sacredone (a text).It is
preciselythe zero degree of the generallinguisticmessagethatgivesit
its highdegreeof semioticmeaningfulness as a text:
(semiotichnost')
"Go toEgypt... and staythereuntilI tellyou.""Attheworddondezhe
[until]Olga could not restrainherselfand burstinto tears."The
heightening of themeaningfulness of a textas a wholeis thusoften
connectedwitha loweringof its meaningfulness at the level of
ordinarylinguisticcommunication.Hence, the typicalprocess
wherebyincomprehensible textsbecome sacred: utterancesthat
circulatein thegivencommunity butwhichare incomprehensible toit
are giventextsignificance (snatchesof sentencesand of textsfrom
another culture,for example; inscriptionsleft by the departed
inhabitants of a particularregion;ruinsof buildingsof unknown
purpose; or phrases brought in from another closed social
group-for instance,the conversation of doctorsfor the patient).
Inasmuch as a high degree of text meaning is perceived as a
guaranteeof truthfulness, and text meaninggrowsinasmuchas
ordinarylinguisticmeaning is obscured,in many instancesthe
tendencycan be observedto maketextsfromwhicha highdegreeof
truthfulness isexpectedincomprehensible fortheaddressee.In order
to be takenas a textthemessagemustbe incomprehensible or barely
comprehensible and must need further or
translation interpretation.
The predictions ofthePythia,thepropheciesofa prophet,thewords
of a fortune-teller,a priest'ssermon,a doctor'sadvice,lawsand social
instructions-when theirvalue is determinednot by real language
communicationbut by text supracommunication, all mustbe in-
comprehensible and need Connected
interpretation. with thisis the urge
to partialcomprehensibility, ambiguity, polysemia. withits
and Art
essentialpolysemiacan,in principle, generateonlytexts.
Sincethedestruction ofthegenerallinguistic messageina textis an
extremeexample whichshowsup a latenttendencyand, for this
reason,is fairly rare,and sincetheaddresseeis interested notonlyin
thetruthfulness of theinformation butalso in theinformation itself,
then togetherwith the text necessarilyarises the figureof its
interpreter: thePythiaand thepriest,thescriptures and theminister,
law and its interpreter,art and the critic. The nature of the
interpreteris such that it excludes the possibilityof "anyone"
becomingit.
Connectedwiththesetextfeaturesis thetendencyto ritualizethe
moresociallysignificant textsand to makethedecodingof theritual
obligatorilydifficult.See, for example, the care with which Pestel'

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
TEXT AND FUNCTION 239

workedout the ritualforinitiationinto secretsocietiesand the role of

ritual in early Decembristorganizations.

The division of all messages circulatingwithina communityinto

textsand nontextsand the concentrationon the formeras objects of
studyby the historianof culture do not exhaust the problem. If one
excludes nontextsfrom consideration (for example, when studying
writtenculture,to make the reservationthatoral sources will not be
considered), then we are faced with the need to define the
complementaryfeatures of expression. Thus withinthe sphere of
writing,the graphic fixationof a textmeans nothing.At thislevel it is
equivalent to nonexpression. In its functionas a fixator,however,
transforming an utteranceintoa text,thereis Church Slavonic,which
separates secular writing(in whichcase and at thislevel is a nontext)
from canonical. But within canonical writing another division is
possible (for example, onlyold books maybe texts).Thus a hierarchy
of texts is created with a successive growth of text meaning. An
analogous example is the hierarchyof genres in the systemof classical
literature,where the feature "to be a work of art" increases as one
moves up the scale of genres.
Cultures witha paradigmaticconstructionhave a single hierarchy
of textswithsuccessivegrowthof textsemioticsso thatat the top is the
Text of that culture with the most coefficientsof value and truth.
Cultureswitha syntagmaticconstructionhave a set of various typesof
textswhich embrace various aspects of realityand have equal value
positions. In most actual human cultures these two principles are
complexlyinterwoven.
The tendency to an increase in text meanings properly speaking
corresponds to types of cultures with a high degree of semiotic
meaningfulness.Because, however,in each textthereinevitablyarises
a conflictbetween its linguisticand its text meaning, the opposite
tendencyalso exists.When a certainsystemof truthsand values ceases
to be perceivedas true and valuable, faithis lostin those means which
made the givenmessage be perceivedas a textbybearingwitnessto its
truthworthiness and cultural significance.Features of the text that
were pledges of itstruthfulness become signsof itsfalsehood. In such
circumstancesa secondary,invertedrelationshiparises: in order that
the message should be perceivedas valuable and true (i.e., as a text),it
must not have expressed textfeatures.In these circumstancesonly a
nontextcan fulfillthe role of a text.Thus the teachingsof Socrates in
the dialogues of Plato are the highestteaching inasmuch as it is not
teaching,not a system;the teaching of Christ,which appeared in a

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
240 NEW LITERARY HISTORY

societywhere the creationof religioustextswas restrictedto a narrow

categoryof people of a certaincaste and high degree of literateness,is
a textpreciselybecause it emanates fromone who has not the rightto
create texts.The idea thatonly prose is truthfulin Russian literature
at the moment of the crisisof the "Pushkin period" and the startof
the "Gogol period," Dziga Vertov's slogan about documentary
cinema, and attemptsby Rosselliniand De Sica to do withoutstudio
shotsand professionalactors-all theseinstances,when the authorityof
the text is defined by its "sincerity,""simplicity,""uncontrivedness,"
are examples of nontextsfulfillingthe functionof texts.
Since a text is manifestedin these cases by its nonexpression,the
value of the message is determinedby its truthfulness on the level of
general linguistic semantic well-formedness and ordinary"common
sense." Since, however,the more truthfultextsare takenas the more
authoritative, it is obvious that here, too, alongside the general
linguisticmeaning,we have to do withsome additional textmeaning.
Since as a resultof the conflictof twoconstantlywarringtendencies
in culture-to semiotizationand to desemiotization-the text and
nontextmay change places in relationto theirculturalfunction,the
possibilityarises of isolatingthe featuresof one kind of textfromits
linguistic message. Text meaning may be polemically refuted
by subtextmeaning. Thus the letterof Ivan the Terrible to Simeon
Bekbulatovichhas all the typicalfeaturesof thatkind of textthatis a
petition. It begins with a ritual address and the obligatory
self-belittlingformula: "To my Lord the Grand Prince Simeon
Bekbulatovichof all Russia Little Ivan Vasil'ev and his littlechildren
Little Ivan and LittleFedor make supplication.9"6
All the text elements carry informationabout a humble request,
while all the subtextelements bear informationabout a categorical
command. The noncorrespondence of text and subtext information
creates supplementary meanings. Moreover, the authorityof the
given text principle is undermined. Literaryparody is constructed
along analogous lines.

The systemof text meanings determines the social functionsof

texts in a given culture. Hence, one can distinguishthree types of
relationship: subtext (general linguistic)meanings, text meanings,
and the functionsof texts in the given systemof culture. Thus, a
culture may be described at three differentlevels: the level of the
general linguisticcontent of its constituenttexts,the level of text
content,and the level of textfunctions.The distinctionamong these
three levels may seem superfluousin those verynumerous instances
when subtextmeaningshave a singleand invariantcorrespondenceto

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
TEXT AND FUNCTION 241

certain definitetextsand the textshave a single correspondence to

certain definitepragmaticfunctions.The practice of studyingcases
such as these explains why researchersdo not separate these levels.
One has but to turn to instancesof noncoincidence (of subtextand
text meanings,or of text and functionalones) for it to become clear
that there must be three quite independent approaches.
Let us consider the most elementary case of divergence-the
nonexpression of one of the links.

Subtext Text Functionin the system

message semantics of culture
1. + + +
2. + +
3. + +
4. +
5. - + +
6. - + -
7. - - +
8. -

Cases 1 and 8 are trivialexamples: in the formerall three typesof

meaning are presentand coincide. An example could be any one of a
number of texts, for example, a fairy tale performed before an
audience forwhom the directperceptionof folkloreis stillalive. Here
there is a language communicationwhich,in order to become a text,
needs a particular kind of expression, and inherentin the text is a
certainculturalfunctionwhichcan be servedby the textalone. Case 8
is introducedonly for completenessof the description;it is complete
silence when this has no cultural function.
Case 2 is the instance which was discussed above: a general
linguisticmessage can fulfilla particulartext functiononly if it does
not have features which, in the system till then in force, were
considered essentialfora text.In order to carryout the textfunction
the message must be deritualized from its previous obligatorytext
features.Thus, at certainmoments(forexample, in Russian literature
afterGogol) the literarytext,in order to be perceivedas art,had to be
not poetry (which is a text with expressed features that distinguish
it from nonliterary speech), but prose, where this distinction is
expressed by a zero index. In this case it is the high value of the
subtextcontentthat gives the textits high authoritativeness("where
thereis truth,thereis poetry,"in Belinsky'swords). A textof thistype

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
242 NEW LITERARY HISTORY

does not in principleneed an interpreter(the rejectionof the church

as mediator between text and man-"confess yourselves one to
another"; the demand for laws comprehensiblewithoutthe help of
lawyers;a negativeattitudetowardsliterarycriticism-see Chekhov's
remarkthatone needs only to read his works: "Everythingis written
there"). The removal of the text from the usual norms of semiotic
meaning and its outward desemiotizationare conditionsfor the high
semioticmeaning of the textin thisinstance.
Case 3 is connected with the preceding one and complementsit.
Where the function of a text can be fulfilledonly by a message
withouttextexpression,ritualizedtextslose the capacityto fulfillthe
functionforwhichtheywere intended. The man who holds thatGod
mustbe addressed in simplicityand sinceritycannot prayin the words
of a prayer learned by rote; for Tolstoy, Shakespeare was not art
because he was too artistic, and so on. Texts with emphasized
expression are perceived as "insincere"and consequentlynot "true,"
that is, theyare nontexts.Case 3 is complementaryto case 7.
Case 4 is the most common instance: it is a message withoutthe
supralinguisticfeaturesof a text.From the point of view of cultureit
does not exist,and thus has no culturalfunction.
Case 5 containsno general linguisticmessage: at thislevel it can be
nonsense, or a textin another language thatis incomprehensiblefor
its audience, or, as in case 7, it may be silence. (See the Romanticidea
thatonlysilencecan adequately express the poet: "Only silencespeaks
comprehensibly"-Zhukovsky; Tyutchev's "Silentium"; or
Tsvetaeva's "Prokrast'sya").The supporters of Nil Sorsky believed
that the best way of union withGod was silent("mental") prayer.
Case 6 is the opposite instance, when an incomprehensibleand
insignificantsubtext message cannot be a text or acquire cultural
value.
Case 7: these are instances where nonsigns functionas signs (for
example, the burningof Rome as a spectacle,thunderas a sign,etc.).7
Anotherinstanceof noncoincidenceis when the linksare displaced
and interchanged. For example, only by being anothertext can the
culturalfunctionof a certaintextbe fulfilled.In thisdisplaced system
only low texts (ironic ones, for instance) can serve "high" cultural
functions,only secular ones can fulfillsacred functions,etc.

The possibilityof separating text from functionleads us to the

conclusion thatthe descriptionof cultureas a set of textsis not always
enough forfulldescription.Thus, forexample, if,in any culture,one
did not find any sacred textsbut did discover certain scientificones
(astronomicalcalendars, for example), one mightconclude that the

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
TEXTANDFUNCTION 243

societybeing studied did not have, in its set of culturalfunctions,any

religious function but did have a scientificone. A more detailed
examination of the question would, however, reveal the need for
greater care; the scientifictexts may have been used by the com-
munity,or part of it, for religiousfunctions.For example, a single
text, scientificin nature-say, a new and powerful medicine-may
be scientificforone part of the community,religiousforanother sec-
tion, and magical for a third,thus serving three differentcultural
functions.
There are many instancesin the historyof science when scientific
ideas, precisely because of their powerful effect,became in fact a
brake on scientificdevelopment,inasmuch as they came to serve a
nonscientificfunction and became, for part of the community,a
religion. On the other hand, such textsas a doctor's advice, whose
effectivenessis determinedby the degree of unconditionaltrust,may
lose effectivenessif the patientadopts a "scientific"approach (based
on criticalverification).It is widelyknown thatthe spread of medical
knowledgeamong the population does harm,in certainconditions,to
medicineby endowing a nonscientifictext(the patient'sown opinion)
withthe functionof a scientificone.

It followsthatthe descriptionof a particularculturalsystemshould

be made along threelevels: (1) thedescriptionof subtextmessages,(2)
the descriptionof cultureas a systemof texts,and (3) the description
of culture as a set of functionsserved by the texts. After such a
description the interrelationshipsof all these structuresshould be
defined. It willthenbe obvious thatthe absence of a textwhen thereis
absence of the correspondingfunctionis in no way equivalent to the
absence of a textwhen the correspondingfunctionis maintained.
Two typesof culturemay thenbe postulated: one willtend towards
a specialization of its texts so that to each cultural functionthere
corresponds an adequate type of text; the other type of culture will
tend to obliteratethe boundaries betweentextsin order thatidentical
textsshould serve the whole set of culturalfunctions.In the firsttype
the textis more important,and in the second, the function.
UNIVERSITY OF TARTU

(Translated by Ann Shukman)

NOTES

1 "Nekotoryeobshchiyezamechaniyaotnositel'norassmotreniyatekstovkak raznovid-
nostisignala" [Some general remarksconcerningthe treatmentof textsas variantsof a

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions
244 NEW LITERARY HISTORY

signal], in Strukturno-tipologiceskie
issledovaniya
[Structural-typologicalstudies] (Moscow,
1962).
2 Ibid., p. 145.
3 Polnoesobraniesochinenii[Collected works],Vol. II, bk. 1 (Moscow, 1937-1959), 269.
4 Ibid., p. 152.
5 "Istiny pokazanie k voproshashim o novom uchenii" [A demonstration of
the truthto those inquiringabout the new doctrine],Appendix to Pravoslavnyy sobesednik
[An orthodox companion] (Kazan', 1863), p. 509.
6 PoslaniyaIvana Groznogo[The lettersof Ivan the Terrible] (Moscow-Leningrad,
1951), p. 195.
7 [This paragraph, which is absent in the Russian text, has been taken from the
French version in Semiotica,2 (1969). Tr.]

This content downloaded from 185.44.77.62 on Sat, 14 Jun 2014 06:08:32 AM

All use subject to JSTOR Terms and Conditions