Allegory as Interpretation

Author(s): Morton W. Bloomfield

Source: New Literary History, Vol. 3, No. 2, On Interpretation: I (Winter, 1972), pp. 301-317
Published by: The Johns Hopkins University Press
Stable URL: http://www.jstor.org/stable/468317
Accessed: 08/11/2010 09:09

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=jhup.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

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 support@jstor.org.

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

http://www.jstor.org
 Allegory as Interpretation

MortonW. Bloomfield

I
OFTHEbasic functionsof allegoryis to make literarydocu-
ONE ments relevant. Historically, the allegorical method as the
West knows it was developed in Alexandria to interpret
"properly" Homer, and somewhat later there and in Palestine to in-
terpret "properly" the Old Testament, so that it could be seen as the
foreshadowing and prediction of Christ or the future kingdom of God.
Allegory in this sense is the seeing of the significance of a literary work
beyond its meaning. The only stable element in a literary work is its
words, which, if we know the language in which it is written, have a
meaning. The significance of that meaning is what may be called
allegory. The problem of interpretation is the problem of allegory-
whether historical or ahistorical. Historical allegorical interpretation
attempts to read the significance of a literary work in terms of its
original or assumed original significance. This sophisticated approach
is essentially modern and was first developed in the Renaissance and
flowered in the last two centuries. Historical interpretation may also
be used for the purpose of modern interpretations. One can argue
that the work must be established in its own mode first, before we
can interpret it properly for our own time. Ahistorical interpretation,
on the other hand, is interested in the "universal" (i.e., contemporary)
significance of a work which may be psychological, ethical, structural,
mythic, religious, or several of these. It is the oldest type of allegorical
interpretation.
In this view of the matter, allegory is that which is established by
interpretation, or the interpretative process itself.1 Allegory, in this
i I'm rather close to Northrop Frye in some ways. He writes, for instance, "It is
not often realized that all commentary is allegorical interpretation, an attaching
of ideas to the structure of poetic imagery." Anatomy of Criticism: Four Essays
(Princeton, 1957), p. 89. L. C. Knights prefers to use the term "metaphoric
process" for allegory as interpretation. (See "King Lear as Metaphor," reprinted
in Further Explorations [Palo Alto, I965], p. 169.) I think this is confusing,
for the allegorical process is only metaphoric by a vague analogy.
302 NEW LITERARY HISTORY

sense, is the method of modernization, that which has made, makes or

keeps modern those literary documents of the past which can bear such
a load of continual reinterpretation.
Allegory is, in this sense, that which conquers time, that which
perpetually renews the written word. The age that does not need, or
thinks it does not need, the past does not need this kind of allegory.
Allegory insists on a continuing and relevant meaning. When the past
is not listened to, allegory declines. It is closely connected with texts.
The humanistic method is exegetical and allegorical in so far as it
depends on texts. In this, it fundamentally differs from science, which
only exegeticizes metaphorically: reads the book of nature.
The allegorical or historical interpretative mind is continually telling
us that the past is relevant and the quotations of the past apply today.
It is text-oriented. Jesus again and again says, "as it is written"-and
breaks our lonely isolation by linking us through quotations with past
Law taking texts and applying them to the present case-casuistry in
the best sense of the word. These insist on the perpetual relevance-
at least in theory-of the past.
Paul de Man has written, "All representational poetry is always also
allegorical, whether it be aware of it or not, and the allegorical power
of the language undermines and obscures the specific literal meaning of
a representation open to understanding. But allegorical poetry must
contain a representational element that invites and allows for under-
standing, only to discover that the understanding it reaches is necessarily
in error."'2 The poetics of representational and allegorical literature
are close to each other.
In this view of allegory, which does not set it against a representa-
tional or mimetic reading, and which sees it as the mode of under-
standing the significance of texts, most scholars and literary critics are
allegorists. If we interpret a work of art, we are seeking its significance
-either historically or ahistorically. Except for textual scholars who
attempt to preserve and correct the verbal surface of a work (and who
may occasionally use significance "allegory" to do it), we may put
all interpreters of literature into the general category of allegorists-
many of whom are modernizers. True literary scholarship aims at
making literature of the past continuously relevant either by establish-
ing its original significance or its modern significance.3
2 See "Lyric and Modernity," Forms of Lyric, Selected Papers from the English
Institute, ed. Reuben A. Brower (New York and London, 1970), p. 175.
3 Occasionally its intermediate significance, as when, for instance, one writes
on Shakespeare in the eighteenth century. Such scholarship, however, seeks to
make the eighteenth century, if not Shakespeare, relevant. It is possible that some
ALLEGORY AS INTERPRETATION 303

The notion of interpretationhas been under attack in recent years.

Some contemporaryFrenchstructuralistshave denied that there is any-
thing to interpret, and refuse to distinguishin any fundamental way
literatureand literarycriticism.4Susan Sontag has written a collection
of essayssignificantlytitled AgainstInterpretation;and FredericJame-
son speaks quite bluntly in saying, "Exegesis,interpretation,commen-
tary have fallen into disrepute."5In spite of this understandableweari-
ness with the mass of exegesisavailableand continuallypiling up, and
with historywhich continuallydemands exegesis, I do not think that
we need take this attitude too seriouslyif intellectual endeavor and
society are to continue at all. A society cannot exist without criticism
and its use of allegorization. I think that, given the persistenceof our
society, interpretationneed not fear for its existence, because I) the
very criticsof interpretationthemselvesinterpret,and 2) many of the
critics of interpretationare actually not reacting against interpretation
but against a particularinterpretation,and 3) intelligence forces in-
terpretationon us. Furthermore,the social sciences are beginning to
realize that the paradigm of social interpretationis to be found in
literaryand linguisticinterpretation.This last point arguesthat cultural
manifestationsmust be read as texts are read. Everythingthat happens
in society has its own grammar which can explain it. Literary, phi-
losophical, journalistic, ritualistic and conventional forms, play and
work all have their own languagewhich can be read by the interpreter.
Interpretationwill disappear only when thinking disappears, for it
can exist even without writing. One can always think-thoughts are
beyondthe reachof anyone.
Turning to the term allegory,we may see that it has been used, as we
have implied, to cover many different meanings. Furthermore,when
the subjectis discussed,words like symbol,icon, myth, emblem, image,

kinds of structural analysis are not hermeneutic as such, but subsidiary or com-
plementary to it. See Gerard Genetto, "Structuralisme et critique litt6raire," L'arc,
26 (1965), 40-41. It is hard, however, to imagine how any structural divisioning
can be made without some interpretation of the text.
4 Discussed by Eugenio Donato in "The Two Languages of Criticism," in The
Languages of Criticism and the Sciences of Man, the Structuralist Controversy,
ed. Richard Macksey and Eugenio Donato (Baltimore and London, 1970) pp. 89 ff.,
especially p. 96.
5 "Metacommentary," PMLA, 86 (1971), 9. Yet others are more concerned
with it than ever. One example of this interest is this present issue of NLH. In
1959 Hans Lipps tried to lay down some principles for a hermeneutic logic. See
his Untersuchungen zu einer hermeneutischen Logik, Philosophische Abhandlungen,
No. 7 (Frankfurt am Main, 1959).
304 NEW LITERARY HISTORY

sign recur again and again. To separatethese sensesis almost a super-

human task. One example sums up the whole problem. Theology has
been concernedfor the past thirty years with the notion of demytho-
logization,ever since Rudolph Bultmannraisedthe issue. He said that
in order to get to the core of the Christianmessage,the mythological
elements in the New Testament must be eliminated. We no longer
live in a world of demons,of astrologicalpowers,of magic, of miracles.
These elementsare stumblingblocksfor moderns. In orderthat modem
man understandthe permanentmeaning of Jesus' life and death, we
must get to the heart of the Chritsianmessage and eliminate the his-
torical paraphernalia.
Now to the demythologizers,myth means the fake beliefs, the his-
torical contingents,of any epoch. Myth is a bad word. In some mod-
ern literarycriticism,however, we find just the opposite use of terms:
we are urged to find the myth in literature. The myth, especiallythe
archetypal myth, is the core of literature, its permanent underlying
meaning. It is the eternal truth of literaryworks, the meaning to us
and to all men in every time and clime, extractedfrom the historical
accidentsin which it is embedded. Supportersof this view are saying
the same as Bultmann and his followers,but they are using the exact
opposite word to describeit. Instead of "demythologizing"literature
they are "mythologizing" it. But whether "demythologizing"or
"mythologizing,"both the theologiansand these literarycritics are at
one. Yet the word myth is bad to the theologians and good to the
critics.6
Some critics like C. S. Lewis have attempted to separatethe terms
allegory and symbolism,reservingthe former for what we would call
personification-allegory.However, few keep to this distinction, al-
though some critics, following Coleridge, use allegory as a bad term
and symbolismas a good one. Symbolismis imaginativeand allegory
is fanciful to these critics. I shall not attempt to keep these terms

6 Jean Dani6lou, "La d6mythisation dans l'Ecole d'Alexandrie," II problema

della demitizzazione, Archivio di Filosofia, I961, Nos. I and 2 (Padua, 1961),
PP. 45-49, argues as we have that Alexandrine allegory is supposed to have been
demythologizing, i.e., modernizing the Bible a la Bultmann. For the centrality of
the problem of myth and the closely related problem of interpretation in much
contemporary theology, especially Protestant, see Pierre Barthel, Interpre'tation du
Langage mythique et thdologie biblique, Etude de quelques itages de l'dvolution
du probleme de l'interpretation . . . (Leiden, 1963). Yet note Hermeneutique et
tradition, Actes du colloque international Rome, to-i6 janvier 1963, (Paris and
Rome, 1963). Enrico Castelli in "Herm6neutique et tradition," ibid., pp. xi-xvi,
writes, "La d6mythisation est toujours le r6sultat d'une herm6neutique" (p. xi).
ALLEGORY AS INTERPRETATION 305

separatehere. Some distinctions,however,should be made.

We must distinguishbetween individualsymbols,images or emblems
referredto or describedin a work of art, and symbolismas a mode of
writingwhich is in principledetachablefrom the literarytext and which
consciouslyaims at a universalrelevance. It is clear that individuals
and objects can be used as symbols in literature. Diogenes and the
cross provide examples. The raven croaking, "Nevermore,"the letter
in The ScarletLetter, the Grecianurn are other examples. These have
a wide extension in literatureand serve a variety of purposes. They
function differentlyin differenttexts, but they are intimatelybound up
with the text. It is not always possible to distinguish them from
imagery.
Symbolic or allegoricalworks of art, however, usually mean some-
thing more than texts which containsymbolsor emblems. They possess
a level of significancein the work deliberatelyemphasizedand mani-
pulated by the writer and in principledetachablefrom the text.
Anothercomplicatingfactoris a long traditionof secretteachingthat
goes back to Pythagorasand Plato and possiblybeyond, which we may
call the hermetic tradition. It is probablytrue that some great spirits
taught in their writings an esoteric doctrine to which only the wise
could penetrate. Leo Strausshas stressedin the past thirty-fiveyears
the importance of this aspect of philosophy and theology. Certain
subjectsor aspectsof subjectsare not for the profane, for the general,
or for the masses. The words written down are for the wise and able.
Some philosophershave or hint at a secret doctrine above the plain
meaning of the text. Maimonides,in the introductionto his Guide to
the Perplexed, speaks of the allegorical parables and figures under
which he is writing.' Even more frequently, prophets and disciples
have claimed to find a secret meaning in the writings of others. The
Baconians have tried to find the secret messagesof Bacon in crypto-
grams in Shakespeare'sworks. Such interpretationsof the Bible are
legion.
Then there are workswhich openly presentdifficultieson the surface
-Gnostic, Kabbalistic and Hermetic (with a capital H)-writings
which are meant to be read only for their secretsenses. Allegoryin the
hermetic sense, then, is deliberately secret and concerns itself with
mystical or semi-mysticaldoctrines. Moreover, sometimes individual
symbols as discussedabove may have hermetic meanings. Athanasius

7 Trans. M. Friedliinder, 2nd. ed. (London, 1904), pp. 4 ff.

306 NEWLITERARY
HISTORY

Kircher in his ObeliscusAEgyptiacus,(1650) writes that a symbol is

"the significantsign of a hidden mystery."8
Some of these hermetic works speak openly about their secrets or
even give a key to the readerso that he may see throughwhat to the
writer are obvious symbols. Some readersare hermetically-mindedin
any case, to which the numerousromansa'clef interpretationsof literary
works bear evidence. Some are convinced, as Sir Philip Sidney was,
that "thereare many mysteriescontainedin Poetry, which of purpose
were written darkly,lest by profane wits it should be abused." Some
literature may have an inner political meaning-"aesopic"-as the
Russianscall it.
Hermeticallegorydiffersfrom Biblical,Classical,and other allegories
largely in the attitudeit createsin its adepts and readers-one of awe
at a hiddensecretmeaningwhich presumablyis not open to the profane
and vulgar. It also differs in its vaguenessfrom normal Biblical and
Classicalallegory. Even when the latter is contradictory,as sometimes
happens, we at least know what the alternativesare. With hermetic
allegory,one can neverbe sure.
Hermeticallegoryis often taken as the only kind of allegory. Michael
Murrinin his recentbook, The Veil of Allegory,9writes that "allegory
is preeminentlyan obscure form of poetry" (p. 8) and that "the
allegoricalpoet ... expendsmuch of his energyin protectinghis truth
from the multitude,ratherthan communicatingit" (p. 9). This view
of allegorywas especiallywidespreadin the Renaissanceand is based
on a senseof an ancient,restrictedbody of knowledgeaccessibleonly to
an elite, and at the same time on a desireto defend poetry as a worthy
peer of philosophy. Poetry is a more hidden kind of philosophy,of
higher worth and not for the profane. Allegoryin this sense desiresto
restore,as Paul Ricoeurputs it, some lost, forgottenmeaning, a process
which he calls negative hermeneutic.10
Anotheruse of allegory,as we have suggested,is appliedto a special
kind of writing which employs personifications.Some medieval alle-
gories are of this type, in which abstractionssuch as Truth, Holy
8 Quoted in Don Cameron Allen, Mysteriously Meant, The Rediscovery of Pagan
Symbolism and Allegorical Interpretation in the Renaissance (Baltimore and
London, 1970), p. 128. This whole book is a valuable collection of Renaissance
material on the hermetic tradition and its allegories (as well as allegorical in-
terpretations of mythology). See also Edgar Wind, Pagan Mysteries in the Renais-
sance (New Haven, 1958, rev. ed. 1968) and Liselotte Dieckmann, Hieroglyphics,
The History of a Literary Symbol (St. Louis, 1970).
9 (Chicago and London, 1969).
io De l'interpritation (Paris, 1965).
ALLEGORY AS INTERPRETATION 307

Church,Mr. Everyman,Falssemblantappearas charactersand act out

their roles. The use of personificationsin parts of works of literature
and sometimes throughout them is widespread. Personification is
perhaps the most common, as it is certainly the most open, of all
literaryfigures.1'
As a figure in Classicalrhetoric, allegory means "saying one thing
and meaning another." Allegoryis a general word in this context for
a commontype of irony.
Somescholars,moreover,have triedto distinguishbetweenhorizontal
and vertical allegory.12 Horizontal allegory is sometimes called
prophetic allegory,but more commonlyfigural allegory. Figural alle-
gory, or typology, is common in the Bible when one event or person
foreshadowsor is the figure or type for another. Auerbachis especially
associatedwith this type of allegory,for he believed that it respected
the letter or surfaceas well as the future meaning. He did not wish, as
many allegoricalcriticsdo, to despisethe surfacefor the inner meaning
or significance.13Auerbach disliked those critics who ignored the
literal level as the husk or shell, but at the same time he wished to
recognize medieval allegoricaltechniques. He felt that the "figural"
method took seriouslyboth matter and sense. Although Biblicalfigural
allegorywas certainlyused in the Middle Ages, I doubt whether there
was a figural method or composition,14except perhapsfor the Divine
Comedy,as Auerbachargues.
When allegorythen works on historyto project a meaning into the

11 See M. W. Bloomfield, "A Grammatical Approach to Personification Allegory,"

MP, 60 (1962-3), 161-71 (reprinted in Essays and Explorations, Studies in Ideas,
Language and Literature [Cambridge, Mass., 1970], pp. 243-60) and works referred
to there. See also Paolo Valesio, "Esquisse pour une 6tude des personifications,"
Lingua e stile, 4 (1969), 1-21, and Angus Fletcher, Allegory, the Theory of a Sym-
bolic Mode (Ithaca, 1964), pp. 26 ff.
12 E. g., Jean P6pin, Mythe et allegorie, les origines grecques et les contestations
judio-chretiennes (Paris, 1958).
13 His famous article "Figura" appeared first in his Neue Dante-studien (Istanbul,
1944), PP- I1-71, and was translated into English by Ralph Manheim. It is most
easily available in this form in Erich Auerbach, Scenes from the Drama of
European Literature, Six Essays (New York, 1959), pp. I 1-76.
14 "Foreshadowing" is a well-attested method of composition found in Homer
and common in all periods. I do not consider normal foreshadowing figural com-
position. It did not need the Biblical example for its existence as a method of
composition. Unless one is committed to a new prophetic theory (as, say, Joachim
of Flora), it is hard to conceive of what figural composition means. I follow
Anthony Nemetz when he writes, ". . . what is signified does not determine the
mode of signification." (Speculum, 34 [19591, 78). A recent attempt to claim such
a technique is made by Robert B. Burlin in his helpful The Old English Advent,
A Typological Commentary (New Haven and London, 1968). I do not question,
of course, figural allegories in the OE Advent. I am questioning figural com-
position.
308 NEW LITERARY HISTORY

future, we may call it prophetic or horizontalallegory; when it finds

an immediatemeaning in events,real or literary,it is simple or vertical
allegory. The first (when firstenumerated,at least) is only fully under-
standable in retrospection;the second may-but not necessarily-be
understoodimmediately. Accordingto the traditionalpicture,we owe
the firsttype of allegoryto the ancient Hebrews,and the second to the
Greeks. In the Christian interpretationof history, particularlyOld
Testament history, going back to the New Testament itself we find
typologicalor figural allegory (a type of propheticallegory), in which
events have an eschatologicalor future meaning. Sacred History is a
progressivefulfillmentof what is latent or obscurelyindicated earlier.
But vertical allegoryis also found in Old Testamentexegesis,especially
in the Alexandrianschool, of which Origen is the notable commen-
tator.15This latent sense, besidesbeing called typology,has in modem
Catholic theology often been called the "fuller sense" (or sensus
plenior) .16
However, both Harry Wolfson and others have cast doubt on this
picture. Wolfson doubts whether there is any fundamentaldifference
betweentypologyand allegoryin Philo and the ChurchFathers.17After
all, one of the levels of vertical allegoryis called (confusinglyenough)
the allegoricallevel. This is supposedto indicate events in the life of
the Church and its divine founder. Such interpretationswould cer-
tainlybe very typological.
De Lubac in his great study of medievalexegesis18attacksthe notion
that Catholicthinkersever tended to ignorethe literalsense,and argues
that many-leveledallegorypays as much attention to the historic and
literal sense as typology. We need not enter into this disputehere, but
these critics have a point. Some Church Fathers, however, did tend,
in practice at least, to ignore the literal for the spiritualmeaning.

II
When we, however,define allegoryin its broadestsense as significance,
we must make room for a number of distinctionswithin this general

I5 This is the picture presented in, say, R. W. C. Hanson, Allegory and Event
(London, 1959). Hanson, like Auerbach, approves of typology but frowns on
allegory.
I6 See Pierre Benoit, "La pl6nitude de sens des livres saints," RB, 67 (I96o),
161-96.
17 See, e.g. Harry Austryn Wolfson, Faith, Trinity, Incarnation, The Philosophy
of the Church Fathers, 3rd ed. rev. (Cambridge, 1970), I, 24 f. especially 39 ff.
x8 Exedgse me'die'vale, Les quatre sens de l'Ecriture, Th6ologie, 41, 42, 59
(1959-63).
ALLEGORY AS INTERPRETATION 309

categoryif the concept is to be usable at all. We have alreadyreferred

to systemsof significations,such as Christianity,which were commonly
used in a periodlike the Middle Ages and Renaissanceas the matrix of
allegory. This is the basis of historicalallegoryin the West. Further-
more, we must recognizethe special uses of allegory or symbolismas
indicatedabove in symbolsand emblems. There is also personification-
allegory,allegoryas irony,and hermeticallegory.
Even more important is to distinguishbetween conscious and un-
conscious allegories. This problem pertains especially to ahistorical
allegoryin the modem period. If a work of art is to have any meaning,
it must in some sensebe symbolic. David Copperfieldis, in this general
sense, as symbolic as Moby Dick. If a work of art says anything, it
must have a generalmeaningin or above the meaning of its particulars.
Particularsby themselvespresent only raw experience. To be under-
stood, they must be conceived of as particularsof something. They
must, unlessthey are propernouns, belong to a class of particulars. In
this sense, experience-present and past and potential, in fact or in
words-has in its ordinaryway a meaning and a meaning of organiza-
tion above and beyond the meaning of its particulars,which I have
called, above, signification. This is what Aristotlemeant by his famous
distinction between poetry and secular history. (Sacred History is,
however,significanthistory.)
Particulars,then, must be significantparticulars. Yet I think we all
recognize a differencebetween Kafka and Golding on the one hand,
and C. P. Snow and Joyce Cary on the other. Of the first two we say
that they are symbolists(or allegorists),althoughusuallyno two critics
can agree as to what is being symbolizedin their novels,whereaswe do
not apply that term to the second one. How can we make precise
this distinction? It does not lie merely in the first group of novels all
having a significanceabove and beyond the surface,for both groupsof
novels share this characteristic,but it must lie in the way in which this
significanceis hinted at or presented.
Here I think the notion of dominance put forward by Angus
Fletcher as a development from Northrop Frye's idea is helpful.
Fletcherwrites, "Whenevera literarywork is dominatedby its theme,
it is likely to be called an allegory. . ... Allegorical stories exist, as it
were, to put secondarymeaningsinto orbit around them; the primary
meaning is then valued for its satellites."19 In allegoricalwriting, Frye
says, the author'swhole technique "triesto indicate how a commentary
19 Fletcher, pp. 22o-21, 304-05. See N. Frye, Anatomy, pp. 89-90.
310 NEW LITERARY HISTORY

on him should proceed." This, he goes on to say, imitates critics,be-

cause it attemptsto prescribehow to interpreta poem.20
In allegoricalnovels,then, the "significance"exercisesa close control
over the plot, imagery,tone, and characters. Every elementis theoreti-
cally there for its significance. The surface tends to exist for the sake
of the significationor significations.The theme controlsin a relatively
close way the story. As we read, we sense this control.
Novels which are not primarilyallegorical,or are only allegorical
in the most general sense of the word, tend to give a freer rein to the
storyand the details.All elementsdo not seem to be underthe tight con-
trol of the significance. Many of them establisha situationor provide
a background which is not strictly needed. They are not "over-
determined."
This distinction,however,must not be taken in too stricta sense,but
in general it is helpful to define the essence of the differencebetween
allegoryin one of its narrowersensesand allegoryas significance. Nor
must we conclude from this distinction that the significance of an
allegoryis necessarilyeasy to determine. It may or may not be so. Just
because a narrativegives the impressionof being closely controlledby
the theme does not mean that the theme is necessarilyunisemous or
easyto unravel.
Furthermore,in a curiousway, the non-allegoricalwork allows for
less detachabilitythan the allegoricalwork. The symbolismof David
Copperfieldis in the surfaceof the work in a deeperway than in The
Castle. K. could be replacedby someoneelse in The Castle. Without
David Copperfieldand all the other charactersand events there could
be no David Copperfield(even with a differenttitle). This is not to
say, of course, that the name K. rather than Hans Neumann, for
instance, does not bring anything in particularto the novel. In fact,
the relation of K. to the author'sown name is a particularlybrilliant
piece of allegorizing. However, the greaterpart of the significanceof
The Castle, whatever it may be, could still be there if the hero were
named Hans Neumann and underwent many different experiences.
An allegory can always be added to.
By making use of the notion of detachabilityalong with that of
dominance,we shall be helped to make the kind of distinctionsneces-
sary if we wish to define allegory as interpretationof the significance
of a work of art. The awarenessof this detachabilityenabled Hazlitt
to make has famous bon mot about the Faerie Queene: "If we don't
meddle with the allegory, the allegory won't meddle with us." It is

20 Ibid., pp. 89-90.

ALLEGORY AS INTERPRETATION 311

easy to detach the allegory of the Roman de la Rose from this work.
It is not easy to detach the allegory from Chaucer's Parlement of
Foules. Furthermore, the theme of the Roman de la Rose (RR)
dominates it in a more involved way than the theme of the Parlement
of Foules dominates that little work. We may say that RR is an allegory
(in the narrower sense of the word) and the Parlement of Foules is not.
However, if we are committed to a theory of interpretation that
claims, for instance, that all medieval literary works exist to promote
the lessons of Chritsianity or that they contain the four levels of Biblical
interpretation which are traditionally found, at least at some periods
in the Middle Ages, then we may find a special allegory in all medieval
literature.
The Frye-Fletcher definitions must be used with caution, inasmuch
as, without common sense and a close reliance on the text, one may
lack the criteria of corrigibility. As C. S. Lewis says, "No story can be
devised by the wit of man which cannot be interpreted allegorically by
the wit of some other man .... Therefore the mere fact that you can
allegorize the work before you is of itself no proof that it is an alle-
gory." 21 Now, works can certainly have a plethora of significations,
but they cannot have an infinity of significations if we are concerned
with the historical situation. If we are not, then any work can mean
many things-but the historical truth acts as a sobering force. It is not
satisfying to feel that anything can mean anything.

III

The crux of the problem of allegory as interpretation is the surface

of the work. The text is the skin which makes possible the existence
of the work. Renoir is supposed to have said that he was not a butcher.
He did not paint flesh, he went on to say, but skin. What is the surface
of the literary work that enables us to interpret it? Here, in the most
stable and enduring part of the work of art, we run into difficulty, for I

21 "On Criticism" printed in Of Other Worlds, Essays and Stories, ed. Walter
Hooper (London, 1966), pp. 57-58, (a talk given to the Cambridge University
English Club, on 24 November 1955). Cf. the story told by Giraldus Cambrensis
(Opera, Rolls Series, I, pp. 409-10) about the allegorizing of the author. "The
archbishop listened diligently to that part which concerned the birds, thir natures,
and the allegories assigned to them.... He inquired whether I had some evidence
from the writings of the Saints and commentators for assigning such allegories. I
answered that actually there was no authority there except that which came from
divine Grace. The good man replied, 'I do not marvel at that, for surely these are
in the same spirit.' "
312 NEW LITERARY HISTORY

shall argue that the surface of a literary work is not simple to determine
and contains at least two, and possibly more, levels. When we are
talking about the literal level, we are already talking about a very com-
plex phenomenon. I hope, in fact, to argue paradoxically that the
most profound aspect of a literary work is its surface such as it is; as
we explore it, it vanishes too.
The achievement of great literary works obviously lies in their being
put together in the way they are. It is the manipulation of words in a
certain order which is the accomplishment. Words are sound or
written symbols which tend to lose their own being in their referenti-
ability, one might say, except for the small number of grammatical
words which exist in order to indicate primarily grammatical relations.
The vast numbers of words in any language are referential or lexical-
moving out of their sentence or word contexts to refer to ideas, things,
movements, etc. The literary artist, in whatever form, will not forget
the symbolic role of his words. He will appreciate their being-ness.
In so far as he "foregrounds" his words, he is an artist. He makes the
referential words lose some of their referentiability so that we can
appreciate them as words. For this alone, which creates the very basis
of the verbal art, we must be grateful for the literal sense. The literal
sense makes the work exist as art.
Now this is all very obvious, but it needs stating because those who
treat art purely referentially or didactically tend to ignore or minimize
the literal sense. With the large number of searchers after archetypes
and moral messages today, "the allegorical berserk" as Arthur Freeman
puts it,22to remind the reader of the magic and power of the literal is,
I think, valuable. Archetypes and moral messages do exist, of course;
but pan-allegorism is another matter.
The literal level, in one sense, is a series of noises and/or marks on
paper, as is all language. Because these noises and marks carry distinc-
tive and contrastive features in terms of a system of meaning exhibited
in recursive syntactic rules, we can interpret them as meaningful noises
or marks-meaningful in the English (or what have you) language
system. They cannot be conceived with any sense as a literal level
without carrying some meaning. The literal level, in a literal sense, is
gibberish. Once the semantic element is introduced, and it must be if
these noises are to be more than a foreign language or the chattering
of a squirrel, then we have to bring in some meaning immediately. If
in spite of this meaning to which native speakers are attuned from early

22 In his poem "Swift, 1740o."

ALLEGORY
ASINTERPRETATION 313

childhood we can still be forced to attend to the noises as noises, then

we are appreciating the discourse (or actively disliking it).
If, then, meaning must be on the literal level, as I am arguing, the
question then arises as to what this meaning must include before con-
scious interpretation (allegory in the broad sense of the word) begins.
When does the area of signification begin? Anthony Nemetz in an
article on the medieval "sensus litteralis"23 argues that medieval phi-
losophy included figures of speech, metaphors, and even parables on
the "sensus litteralis." (He refuses to translate it as "literal level," be-
cause the phrase in modem criticism has normally a much more re-
stricted meaning; he believes this confusion explains much misunder-
standing.) Only, he argues, if what the words refer to refers to some-
thing else, may we then speak of the spiritual sense. In other words,
only if what the words stand for is yet something else, like the events
and beliefs of Christianity, may we speak of a spiritual sense. "... the
sensus litteralis is concerned with the signification of words and con-
cepts, . . . the sensus spiritualis is concerned with the signification of the
things which the words signify" (p. 79)-
This point is certainly made by Augustine (e.g., De Trinitate XV,
9, 15), by Aquinas (S.T. I, I, o108, and In Gal. 4:7), and by others,
but I suspect that it is only partially true. I do not doubt the fact that
many did regard the matter in this light, nor the fact that the more
the literal sense includes, the less allegory (and vice versa); but I do
doubt whether the practice of medieval allegorists consistently followed
this division between the signification of words and the signification of
what the words stand for. For one thing, the four-fold allegorical tech-
nique-which is essentially a monastic technique, not a scholastic one-
did not move, in general, in this way. The moral or tropological level,
for instance, is not a specifically Christian level; it does not refer to
another reality revealed by words, but is actually seen in the meaning
itself. When Isaac displays the virtue of obedience as he willingly
lies on the altar at his father's command, the tropological meaning is
right in the primary level of meaning: obedience to the commands of
one's superiors. Furthermore, in personification-allegory the allegorical
meaning (or most of it) is in the literal sense itself. When Lady Holy
Church appears in Passus I of the B Text of Piers Plowman, her alle-
gorical significance is not the Christian significance of what Holy
Church stands for, it is Holy Church right then and there.
In other words, the problem of the literal level is much more com-
plex, I think, than has been generally admitted. It is not easy to draw

23 See above, note 14.

314 NEW LITERARY HISTORY

the line between the literal and the spiritual level, let alone three
spiritual levels. Robert Hollander in a recent book would find the
four-fold "meaning" everywhere in the Divine Comedy.24 It is encour-
aging to learn from him that, owing to the obtuseness of early com-
mentators on the Divine Comedy, as well as to other factors, no one
until Mr. Hollander came along has been able to understand Dante's
method. He confuses (to me, at least) typology with the four-fold
method, although-as may be seen above-I am willing to admit that
distinctions here are hard to make. But there are obviously some dif-
ferences, including the fact that typology is very text oriented, in a way
the four-fold meaning is not. It is hard to see how a poem like the
Divine Comedy, which is almost completely concerned with anagogy
(eschatology) (the highest spirtual level) on the literal level,25 can
contain four-fold meanings above this literal level. Pietro Alighieri,
the son of Dante, finds seven senses in the work, the first four of which
are literal26-superficial, historical, apologetic, and metaphorical.
These four are not very clearly distinguished, but they do point to an
awareness of complexity of the literal. Some Biblical commentators in
our own time have also referred to the possibility of a double literal
sense.27
The problem of the allegorical level, or allegory as interpretation,
lies essentially in the literal level. Meaning in many senses is intertwined
with the literal level. Furthermore, the over-all signification sometimes
is necessary in order to get the "meaning." These levels and their
divisions are a very complicated matter. In fact, the basic role of
hermeneutics is to distinguish the literal-meaning level from the signifi-
cation level, in order finally to bring them together again if necessary.
We have here something like the movement from whole to part, and
part to whole, in the understanding process.28 The problem is especially
acute in works which are conscious allegories and which reflect
thought and belief systems like Christianity. The literal-meaning level

24 Allegory in Dante's Commedia (Princeton, 1969).

25 Admittedly the eschatological drama is not yet played out-the Last Judgment
is yet to come-yet the next world is a major part of the anagogical level and of
Christian eschatology.
26 See John Paul Bowden, An Analysis of Pietro Alighieri's Commentary on the
"Divine Comedy," Diss. Columbia 1949, PP. Ioo ff.
27 See, e.g., Catholic Biblical Quarterly, II (1949), 20o2-o6.
28 Something very like Leo Spitzer's "philological circle," which, as he points
out, goes back to Plato's Phaedo and was developed by Schleiermacher, Dilthey,
Heidegger, and others. See Spitzer's "Linguistics and Literary History," Linguistics
and Literary History, Essays in Stylistics (Princeton, 1948), especially pp. 33-35
(n. io).
ALLEGORY AS INTERPRETATION 315

is primarily linguistic; the signification or allegorical level is primarily

non-linguistic; or perhaps we may say that this level is the life function
as opposed to the linguistic function.29
In one sense, as Owen Barfield points out,30 there is no literal sense
at all, but few will follow him in his definition of literal, which comes
to this conclusion: that literal nouns correspond to real and wholly
material entities. I do not think we have to worry about this misconcep-
tion. Furthermore, what we are talking about is the historical sense and
sound of words as far as these may be established. From a historical
point of view, the literal sense is changeable in different degrees, de-
pending on the words used, whether they are still understandable and
whether they have changed significantly or not. What was once met-
aphorical may become literal and vice versa, but all this verbal move-
ment happens within the literal sense. Metaphors belong to the literal
sense if they are clearly presented. They differ from the simple literal
sense in the fact that they paraphrase only asymmetrically. That
is-to take the example given by Donald Stewart31-we can say that
"Richard is a lion" means "Richard is brave," but not vice versa. A
metaphor's meaning is very open, and it is determined largely by con-
text. We hesitate to say that "Richard is brave" means "Richard is a
lion," because the metaphor can mean other things out of context. Yet
both sentences are on the literal level.
Etymologies of words, when we know them, show that words may
shift their meanings. However, we are here talking about the syn-
chronic, not the diachronic, literal sense. "Literal" has several mean-
ings, one of which is pejorative. In the sense of "unimaginative,"
"literal" may be condemned by all. In our sense, however, all verbal
communication has a literal sense, because "literal" refers to that which
bears the meaning and the basic meaning borne. Otherwise we are
only talking about barks or marks on paper.
There is also a strong tendency to confuse realism with literalism.
This widespread error, it seems to me, leads to a tendency to see allegory
everywhere. Robert Kaske, who is one of our best medieval scholars,
fell into this error when, for instance, talking of the half-acre scene in
Piers Plowman (B VI), he wrote some years ago, "One inevitably ques-
tions the possibility of coherent and significant literal meaning in a

29 See Isabel Hungerland, Poetic Discourse, University of California Publications

in Philosophy, No. 33 (Berkeley, 1958), p. 36, who provided me with the terms.
30 See his "The Meaning of the Word 'Literal,' " Metaphor and Symbol, ed. L. C.
Knights and Basil Cottle, Proceeding of the Twelfth Symposium of the Colston
Research Society (London, I96o), pp. 48-57, and discussion, pp. 57-63.
31 See his "Metaphor and Paraphrase," Philosophy and Rhetoric, 4 (1970-71),
I 15 ff.
316 NEW LITERARY HISTORY

narrative embracing elements like the cultivation of a 'half-acre' by so

unlikely and so patently composite a group, and Piers' literally mean-
ingless [sic] act of calling to Hunger and Piers' allegorically-named wife
and children-a difficulty hardly removed by the explanation... ."32
Professor Kaske seems to demand realism, otherwise we can have no
literal level. How can we have a work of verbal art at all without a
literal level? If this scene has no literal level, how can it exist? It is
obvious that behind these remarks is implied a very different concept
of a literal level than I have. If we do not have a literal level in the half-
acre scene, where is it? In the marks on the paper? In the noises we
make when we pronounce them?33
The literal meaning, both in medieval philosophy and in common
sense, includes much more than what many take it to include. Some-
times non-naturalistic elements occur on the literal level, some of which
may be self-explanatory. The poet may explain what his allegory
means. Or personification may be used to avoid ambiguity. In allegory
in the sense we have been discussing it here, there is always the pos-
sibility (even though, on occasion, slight) of misunderstanding what
is really meant.34 If allegory becomes the literal level, then this
normally becomes impossible.
This tendency to see allegory and realism as opposites is the source
of much difficulty in interpreting both these multi-faceted terms. In
terms of allegory as interpretation, such an opposition is unthinkable,
because any work of art must have, at least theoretically, an interpreta-
tion. If we think of allegory as the use of symbols, as irony, as a par-
ticular system of interpretation-Biblical or hermetic, for example-
or as personification narratives, these are all unrealistic, but they all
have literal levels of great significance. We cannot ignore the literal
sense by saying that the "allegory says" when it is the literal sense that
"says."
32 Review of Robert Worth Frank, Jr., Piers Plowman .. . (New Haven, 1957),
in MLN, 84 (1959), 730-3 1.
33 A similar type of assumption which confuses the allegorical (in its sense of
interpretation or significance) with the literal is made by Mrs. Elizabeth Salter in
her Piers Plowman: An Introduction (Oxford and Cambridge, Mass., 1962). In
discussing Piers Plowman B XVI, pp. 51-52, for instance, where the poet tells us
how Grace and Holy Ghost resist the world, the flesh, and the devil, Mrs. Salter
comments, "The allegory [sic] here is telling the dreamer how individual man
struggles against evil with the help of divine examples and agencies." The allegory
is telling us nothing; the literal level is telling us this. Further examples of the
stripping of the literal level, in which the literal level practically disappears, occur
throughout Mrs. Salter's book, and it is especially blatant in a phrase like "the
literal allegorical pattern" (p. 75), in which the literal level is called the allegorical
because personifications are used.
34 See my review of Mrs. Salter's book (see note 33) in MP, 72 (1964-5), 62-64.
ALLEGORY AS INTERPRETATION 7
•

Moreover, the literal is not necessarilyconcrete, nor the allegory

abstract. The abstract/concretedichotomymoves in another universe
of discourse. To equate the two is to make a serious categorization
mistake.
I should say that the greatnessof any work of art lies in the literal
sense: that which gives it shape and being.35 It is the literal sense
which is by far the most profound,becauseit alwaysprovidesus with a
new possibilityof interpretation. Any particular significationis un-
changing. In order to extend significationwe must go back to the
literal sense. The literal gives life and continuity to the text. It acts
as the correctiveforce to misinterpretation.It says that circumstances
are important. What gives the Divine Comedy its greatness is the
narrativethat it presentsand the poignancy of the reaction of a man
granted the divine grace to make or imagine such a journey and be
allowed to record it through his memory, which gives all things sig-
nificance.
Bernard Lamy in the eighteenth century wrote, "Allegoriesindeed
are always most easy. But it requiresa great deal of study and labor
to master the literal sense of the sacred books."36 If we take this
attitudethen it would be impossibleto write the following,which denies
respect to the literal level and indeed destroysthe whole notion of
literary art by minimizing the importance of how anything is said.
"Even though the surfaceor literal level of medievalworksof literature
is by and large designedto take the audienceto a symboliclevel that is
the real reason for the work's existence. ... "" A work of art is not only
what it says, but also what it is.

HARVARD UNIVERSITY

35 "To understand and to preserve a literary work is to be primarily concerned

with its 'letter,' " and "While the modesty of literary studies lies in their concern
with the material and perishable signs on the page, their difficulty lies in preserving
the meaning of the literal statement beyond the passing of what it signifies . ."
[from Wesley Trimpi, "The Definition and Practice of Literary Studies," NLH, 2
(1970-71), 190, 191-92].
36 In the English translation of his Apparatus Biblicus, ist. ed. (London, 1723), p.
363 (Book II, Chapter XI, Rule VIII).
37 Edmund Reiss, "Symbolic Detail in Medieval Narrative: Floris and Blanche-
flour," PLL, 7 (1971), 339.