Professional Documents
Culture Documents
An introduction to
literary stylistics
T o Norman Denison
__________.
Linguistics
and Literature
An introduction to literary stylistics
Raymond Chapman
Senior Lecturer in English, London School of Economics and Political
Science
Edward Arnold
( Contents
Preface
vi
Note on Reading
vii
Allies or Opponents?
21
32
Syntax
44
58
72
85
100
11 3
Index
118
Preface
Note on Reading
In ad dition to the books listed at the end o f each chap ter, the
follo w in g are recom m en d ed :
Allies or Opponents ?
which has four possible interpretations that can be sorted out only
b y a deep analysis leading perhaps to rephrasing. In our everyday
use o f language we rightly regard such uncertainty o f m eaning as
undesirable and do our best to avoid it. Y e t critics continue to
argue about the interpretation o f the last lines o f K eatss O de on a
G recian U rn :
W hen old age shall this generation waste,
T h o u shalt rem ain, in midst o f other woe
T h a n ours, a friend to man, to whom thou sayst,
Beauty is truth, truth beauty, that is all
Y e know on earth, and all ye need to know.
W hat exactly is the U rn supposed to be saying the phrase
Beauty is truth, truth beau ty, followed by the poets own comment,
or the whole o f the last two lines? In the absence o f quotation marks
from the original edition, we m ay be in doubt. T h e argum ent must
certainly take into account the question whether K eats elsewhere
uses y e as a singular pronoun. T h e point is that the uncertainty is
a source o f stim ulating discussion, adding to the interest o f the whole
poem rather than detracting from it.
T h ere exists, then, a type o f discourse in w hich we can apply
evaluative words like clever, interesting, and even brilliant to
usage w hich m ight evoke very different adjectives if it occurred in a
different situation. W e m ay justify such discrimination by assigning
Thom as, Hopkins and K eats to the realm o f literature when we are
discussing certain written texts which they have left to posterity.
T h e same tolerance m ight not extend to other aspects o f their
writing, such as their letters or their critical essays, and would
certainly not be extended to an encounter w ith them in polite
conversation if such were possible though in practice w e m ight
carry over some o f the respect attached to their literary reputation.
W hat is literature? T h e question m ay find an answer, or rather
a num ber o f answers, as we exam ine some specimens o f language in
the course o f this book. A t this stage it is easier to say w hat literature
is not. First, it is not sim ply that w hich is w ritten as opposed to that
w hich is spoken. It is true that w e speak loosely about the literature
which a m anufacturer sends out to promote his product. Y et none
o f us w ould include such areas when considering the possible scope
o f a syllabus in English Literature. N or w ould w e include recipe
books, telephone directories, Acts o f Parliam ent or guidebooks to
Allies or Opponents?
Allies or Opponents?
Allies or Opponents?
F U R T H E R R E A D IN G
io
12
13
14
show selection and arrangem ent o f items that contribute to the total
cffect; elements that w ould be absent or incidental in other styles
are im portant for the fulfilm ent o f purpose. Poetry shows such
patterning devices as metre, rhym e, assonance, alliteration; prose
m ay contain similar devices, less regularly arranged. Both types o f
literary discourse w ill have careful and often unexpected selection o f
words and syntactic constructions. Figures o f rhetoric w ill give
unusual prom inence to certain items. W e m ay therefore add a third
to the two distinguishing marks o f literature suggested in the previous
chapter: the use o f special devices w hich heighten the effect o f
linguistic acts through patterning.
Literature m ay be m uch m ore than w ould norm ally be understood
as a style, but there is value in attem pting to treat it as one. A n y
profitable approach through linguistics must deal with literature
as an exam inable part o f the available realization o f langue. Special,
heightened and prestigious as it m ay be, it cannot deviate too far
from the expectations o f the speech-comm unity i f it is to find any
readers at all. Such deviations as occur can be discerned and de
scribed by methods applicable to more fam iliar and hum bler
paroles. L ike other styles, it has features not shared by all users at all
levels; but, as in all styles, these features can be utilized only in
association with com mon core features.
Extremes are generally dangerous and distorting. In some periods
and cultures literary language has gained such prestige that other
styles have been jud ged good or bad according to their resemblance
to it. Non-literary users have tried to incorporate literary features
into personal com m unication. T h e developm ent o f national
languages has been affected b y the prestige o f a dialect used for
literature; individual writers have left their m ark on common
speech. This kind o f influence, by no means undesirable in itself,
has had the unfortunate effect o f isolating literature from regular
methods o f investigation. T h e other extreme, w hich has already
been mentioned, was the dismissal o f literature by some m odern
linguists as too deviant for their attention.
It is now generally accepted that any description o f a langue
must take account o f all its different realizations. T h e present
investigation sees literary style as deriving its strength from the
com m on core , even in respect o f features w hich are usually thought
to be distinctively literary . Its deviations do not break down com
m unication with com mon core users. C ertainly it is sometimes
necessary to m ake the kind o f adjustment or allowance which in the
/5
p.ist was vaguely named poetic licence . A t the same time, the very
extension o f literature into all aspects o f hum an experience means
11 iiit its style is less exclusive than some others w hich are the preserves
of smaller groups.
T h e literary writers have themselves been divided about the par
ticularity o f their style. T h e course o f English poetry, for instance,
shows a succession o f swings between belief in the special nature of
poetic diction and insistence that the criterion o f poetic greatness
was its closeness to everyday speech. A t the end o f the sixteenth
century, Spenser tended to the first view and Donne to the second.
Restoration poets in turn reacted against the school o f Donne, which
seemed to have becom e artificial and remote from real life. M ost
eighteenth-century poets shared the opinion o f G ray that the
language o f the age is never the language o f poetry, until W ords
worth cam e with his standard o f poetry written from a selection o f
language really used by m en .
O ne task o f literary stylistics is, w ithout taking sides in this
dispute, to determine how far and in w hat respects a poets language
in fact shows deviant features. Another is to note how a writer uses
generally accepted features to special effect. It is necessary to p ay
close attention to particular writers, since literature shows far more
diversity o f individual usage than do other styles. This fact creates
a link between modern stylistics and the m ore traditional w ay o f
discussing a writers style in the sense o f the too-often quoted dictum
o f Buffon L e style est l homm e meme .
Although we have tried to postulate a literary style, as parallel
to a legal, m edical or religious style, it is apparent that we are led
(o something that is far from being homogeneous. In practice, the
examination o f a single writer, or o f the com m on features o f a school
or literary period showing com m on aims and influences, w ill yield
the most satisfactory results. W hatever can realistically be said about
literary style as a whole is w orth saying as a contribution to critical
theory and to understanding o f how language works. In the present
study, examples from particular writers w ill be used as evidence for
general principles, but the aim o f the whole is to prepare readers for
close stylistic study o f m ore lim ited areas.
U nlike other styles, literature does not and cannot exclude any
aspects o f langue. H ere the notion o f register is im portant to stylistics,
as w e approach the w riter as an individual user o f resources poten
tially available to other members o f the speech-community. Like
other users, he is in a relationship o f com m unication, though one o f
iy
i8
ig
FURTH ER
R E A D IN G
20
Language, Literature and History
nil
23
if,/
25
26
27
28
855:
T h e angel o f death has been abroad through the land; you m ay
almost hear the beating o f his wings.
T h e interchange between literary and non-literary language is
not confined to discrete features. W e are accustomed, certainly in
European cultures, to think o f a standard form o f a national
language, local or class variants on it being regarded as dialects.
In m any instances, however, the standard form is traceable back
to a dialect which gained prestige through being the m edium o f
literature. A period o f great literary production, with associated
confidence in execution, can elevate a dialect to superior status and
cause it to be adopted b y educated members o f the com m unity for
written and usually later for spoken com m unication, w ithout
area restriction. This status was gained by W est Saxon in preConquest England and by East M idland at the beginning o f the
fifteenth century. T h e same kind o f thing happened to Francien in
France, Castilian in Spain and Tuscan in Italy. In all these instances,
econom ic and social factors also played their part; it can be argued
that the literature o f these dialects owed its strength to favourable
external conditions. L u th ers choice o f Saxon can certainly be
seen to have a direct influence on the developm ent o f Germ an,
partly through the deliberate aid o f later writers and grammarians.
W ithout over-simplifying a com plex question, it is im portant to
realize that literature helps to affect the diachronic developm ent
o f the langue which is its medium. Ernst Cassirers words are w orth
remembering:
N o poet can create an entirely new language. H e has to adopt
the words and he has to respect the fundam ental rules o f his
language. T o all this, however, the poet gives not only a new turn
but also a new life . . . T h e Italian language, the English language,
the G erm an language, were not the same at the death o f D ante,
o f Shakespeare, o f Goethe, as they had been at the day o f their
birth.1
Prestige, status and high quality are notions which are scarcely
congenial to the m ore descriptive and analytical approach o f
m odern linguistics, but w hich cannot be altogether excluded from
1 An Essay on Man (New Haven, 1944, Y ale University Press).
2g
jjo
axiom atic that English could and should be described in Latinbased gram m atical terms.
Also, and this is a w orthy discipline for all whose study requires
historical perspective, the fact that any theory can no longer be
accepted totally does not m ean either that its originators were fools
or that it contained nothing fit to be remembered. Literature can
still be illum inated b y the judgem ents o f those to whom it was
fresh, the latest expression o f im agination through language. O n e
o f the greatest mistakes that can be m ade is to regard the past as
homogeneous. It is easy to laugh at Lord M onboddo in the eighteenth
century calling G othic the parent o f all the different dialects o f the
T eu ton ic . Y e t everyone today must honour the wisdom o f John
W allis in 1653 com plaining that gram marians had previously
forced our tongue too m uch into the pattern o f L atin .
Everyone well, almost everyone. T h ere are still teachers o f
English whose consciences should be touched by the words o f W allis.
I f they venture the P layers excuse, I hope we have reformed that
indifferently with us, they can be met w ith H am lets admonition,
Reform it altogether . N ot all the wisdom o f the past has been
universally recognized. N ot all the attitudes o f the past need be
eschewed by those who have the wisdom o f the present. T h ere are
ideas w hich can be lifted out o f their shaky fram ework and fitted
into the new. Abuses should never make us disregard right uses.
F U R T H E R R E A D IN G
31
Tennyson has left us a b rief word picture o f how his Morte d'Arthur
was read aloud:
T h e poet little urged,
But with some prelude o f disparagement,
R ead , m outhing out his hollow oes and aes,
Deep-chested music . . .
Tennyson was not the first m ajor poet to be aw are o f the distinctive
features in his own or other voices, but he was the first whose voice
was recorded. It is still possible to hear the old gram ophone record,
too faint and technically poor to give m uch idea o f the living reality,
b u t yet a link between two eras in the study o f language. T h e
developm ent o f efficient means o f recording the hum an voice was
contem porary with the recognition o f speech as the prim ary mode
o f language, both as a m atter o f chronological developm ent and in
its im portance as a corpus o f performance. T h e portable taperecorder makes it possible to assemble synchronic evidence before
change invalidates it.
W ith the application o f phonetic methods to the spoken utterance,
w e are no longer satisfied with such verbal descriptions o f speech as
hollow oes and aes . It would not be true to say that the m ajority
o f linguists regard all w ritten language as unim portant or deviant
though the feeling is not unknown but certainly language no longer
means chiefly written language as it did for the classical philologists.
W riting cannot be esteemed as it was when no other record o f usage,
past or present, was available. T h e student o f literature, still
dependent on written texts, m ay once more seem to be working
against the mainstream o f linguists. I f certain readjustments can be
m ade, however, the result is beneficial to both disciplines.
In the first place, it is unwise to exaggerate the distance between
33
iprerli and writing in a society like our own where so m any people
1 <111 nwitch between the two m edia w ith little or no effort. T h e same
1 lias been felt by the m ore literate from whom the makers and
M illers o f literature were m ainly drawn w ithin post-Renaissance
l umpean culture. T h e present day, more perhaps than any previous
, yields examples o f borderline usage with features characteristic
1if one realization em erging in the other. Broadcasting has produced
I'iniialized speech to be read aloud from a script, a developm ent o f
1In- old m ore limited skill o f recitation or the kind o f literary
ti-.tding aloud that Tennyson described. Conversely, fam iliar
Idlers incorporate spoken features such as contracted verbal forms
iiiid lcxical items generally regarded as part o f the colloquial
register.
W c can speak a text or write down a conversation, but in doing
either we are reminded o f the differences o f m edia. T h e apparatus
speech consists o f bodily organs for articulating and receiving,
with sound-waves for transmission. T h e apparatus o f w riting consists
of visible marks, m ade b y various implements on various types of
surface, transmitted by light-waves and received b y the eye. Thus
wc m ay be led to consider acoustics and physiology on the one hand,
typography and optics on the other. It m ight seem as if there were
two different languages to be studied.
W hat w e in fact encounter are the two ways o f realizing any
given langue, the one phonological and the other graphological. T h ey
both depend on the same available gram m ar and lexicon, but the
selection m ade therefrom m ay be affected b y the type o f realization.
In any stylistic study it is necessary to m aintain sensitivity to the
influence o f the alternative realization w hich is not actually being
examined/ L iterary language, almost entirely written, w ill not be
appreciated in depth if we stop thinking about speech altogether.
It demands little specialist skill to see the m ore apparent differ
ences between spoken and written realizations, quite apart from the
physical m edia used. W riting is generally accepted as more careful
than speech. M an y people feel that they are putting themselves on
more perm anent record w hen they write; they are more cautious
both about w hat they say and about how they say it. Consciousness
o f gram m ar as a set o f prescriptive rules becomes more m arked.
Even in ephem eral modes such as personal letters, most o f us w ould
at least try to divide our discourse into sentences conform ing to the
old and im perfect rule o f m aking com plete sense and containing
at least one finite verb . A covertly introduced tape-recorder,
34
35
Ir.iiures like gesture and facial expression can have com m unicative
value and need to be assessed as part o f the whole.
I n place o f these, written realization has recourse to punctuadon,
paragraphing, and the blank spaces w hich m ay correspond to
mlence. T h ere is also the m ultitude o f graphological devices such as
.ipitalizing, italics, different founts o f type. L iterary creation uses
prosody, scene-division, chapter units and all the other technical
.mis which are well known to literary criticism. It is not difficult to
nrc why written texts, and suprem ely literary texts, were traditionally
1e^arded as the superior form o f language and the source o f rules
lor everyday users. T h e y were stable, accessible, carefully planned |f
liy those who strove to use language to the best effect.
Y et anyone who aspires to the stylistic study o f literature needs to
In- informed about the spoken features o f the langue from which his
texts are derived. Indeed, any student o f literature is missing some
thing if he is entirely ignorant o f what m odern linguistics has to tell
him about phonetics and phonology. Equally, the linguist should
not suppose that literature is irrelevant to this aspect o f his concern.
I or the rem ainder o f this chapter, let us consider the m ain reasons
lor these assertions.
T h e first reason is so simple that it is easily ignored. Every
writer is a m em ber o f a speech-community; the language that he
uses cam e to him in infancy, acquired by a process still not fully
explained today w hich was shared with less articulate contempor.tries. It needs perhaps some conflict o f prestige between languages
lor the deep personal value o f the native tongue to be appreciated.
Milton understood the prim acy o f speech w hen he turned from
I itin to English verse:
H ail, native language, that by sinews weak
Didst move m y first endeavouring tongue to speak,
A n d m adst im perfect words with childish trips,
H a lf unpronounced, slide through m y infant lips,
D riving dum b silence from the portal door,
W here he had m utely sat two years before.
(A t a V acation Exercise )
We have seen how a certain dialect m ay acquire prestige and becom e
the standard for a national literature. It should be remembered that
I he spoken dialect comes first and continues to be used for spoken
realization even after it has gained other and wider distribution.
37
38
39
it1 tn.illy used: slumber, sloth, lassitude, listless, solitude, leisure, all come
In mind.
<)nc more brief exam ple must suffice, with the hope o f guiding
ili reader on his w ay. H am lets indictm ent o f Claudius approaches
Hi 1 1max w ith lines that sound as if they were spat out o f a mouth
ii me with fury:
A cutpurse o f the empire and the rule,
T h a t from a shelf the precious diadem stole,
A nd put it in his pocket.
(IH .iv)
I Iir plosives and harsh consonant clusters are grouped mostly
.1round short close or half-close vowels in a w ay that suggests the
physiological results o f inner tension. T h rou gh the whole utterance,
Ilie Ip/ sound recurs w ith a contemptuous blow ing o f the lips that
eventually underlines the sneer o f the colloquial sneak-thief phrase
'put it in his pocket . T h e fact that our great poets knew nothing
o f phonetic theory and that we do is our good fortune, not their
insufficiency. T h ey seem to have known by their own ears w h at we
can more deeply appreciate b y analysing scientifically. Pope needed
no phonetics to understand that
Tis not enough no harshness gives offence:
T h e sound must seem an echo to the sense.
(Essay on Criticism)
T h e fourth connection between graphology and phonology is the
frequent need to express speech by a literary character. T h e linguist
might w ell think it an ideal if all dialogue in novels and plays were
set dow n in phonetic script, but this o f course w ould cut o ff apprecia
tion from the m ajority o f readers; the contrast with the conventional
orthography o f narrative w ould certainly produce an original and
interesting visual patterning. In practice our writers have had to do
their best with the inadequacies o f the alphabet, so that we must try
im aginatively to interpret the rendering o f dialect or idiophonic
speech. Shakespeare thus shows E d gars sim ulation o f peasant
character by using the contem porary conventions o f rustic speech on
the stage:
G ood gentleman, go your gait and let poor volk pass. A nd
chud h a been swaggered out o f m y life, twould not ha been zo
long as tis by a vortnight.
(IV .vi)
40
41
/*
43
point for stylistics when he said that all art constantly aspires towards
the condition o f music.
FU R TH E R R E A D IN G
Syntax
Syntax
45
Syntax
47
Any native speaker must agree that this sentence is both gram m atic
ally and sem antically well-form ed. A n y oddity w hich it suggests is
48
Syntax
49
which com munication depends and the freedom which the users
unique situation demands. T h e deeper the artistic concern with the
in.inner as well as the m atter, the greater the tension is likely to
become. In everyday discourse, syntax and message co-operate
without troubling anyone very much. In literature and perhaps
in some other styles w ith distinctive features syntax becomes more
onscious and is likely to m ake the user intolerant o f its restrictions.
It is in this area that the difference between literary and colloquial
performance is seen most clearly.
I n one sense, literary language has the greater freedom. Just as it has
been suggested that no register can be excluded from the total concern
of literature, so no choice o f generation from the gram m ar is for
bidden. O ther styles m ay constrict or enjoin: recipes and instruction
manuals m ake considerable use o f the im perative, which is seldom
found in pure science or literary criticism. L itu rgy and preaching
dhow a particular need for the shared im perative, L et us. . . . T h e
style o f Parliam entary debate forbids the use o f the second person
in referring to other members. Literary fashions and the pressures
o f critics m ay indeed lim it the writer if he chooses to heed them,
but literature itself acknowledges no prohibitions.
Y et there are pressures from the gram m ar itself, and it is these
which m ay result in deviation. A n y writer must use, except with
deliberate archaism, the syntax available in his own time. In some
ways we m ay think that English syntax has lost a certain am ount o f
strength over the centuries. T h e present-day writer is forbidden the
emphasis o f repeated negatives that was open to Chaucer:
H e never yet no vileyne ne sayde
In al his lyf, unto no m anner w ight
(General Prologue)
and the doubled superlative o f Shakespeares
This was the most unkindest cut o f all
(,Julius Ceasar, IIL ii)
Poets m ay feel the later language to be overloaded with prepositional
phrases and post-modifying clauses and long for the freedom o f
ellipsis leading to the compressed pre-modification o f O ld English:
H i leton { o f folman feolhearde speru
grim m e gegrundene fleogan
(Battle o f Maldon)
50
Syntax
31
52
Syntax
53
54
Syntax
55
w ill be the deep structure the gram m atical structure o f the base from
w hich the surface structure is generated in which the semantic
m eaning o f the sentence m ust be sought. T h u s although John seems
to stand in the same syntagm atic position in each o f the first two
sentences quoted above, in the deep structure he is shown to be the
subject o f the first sentence and the object o f the second. A similar
result comes from analysing w hat is really being said about
doctor in each o f the second pair o f sentences.
Therefore w hat is violated by an unacceptable sentence such as
*The men is here can be seen to be surface structure; it is at this level
that the everyday judgem ents o f g ra m m a tica lly are made. So also
the variations in word-order that have been quoted w ill affect only
the surface structure: they are o f course none the less interesting
from the point o f view o f literary syntax. D eep structure is violated
b y the appearance o f an item w hich is not generally accepted in
that particular position in relation to the other items. D eviation of
this kind is caused b y the italicized words in the following:
T h e branches shake down sand along a crawling air,
and drinks are miles towards the sun
(Terence T iller, Lecturing to Troops)
D o not go gentle into that good night
(D ylan Thom as, D o not go gentle )
V a lu in g him self not a little upon his elegance, being indeed a
proper m an o f his person, this talkative now applied him self to his
dress.
(James Joyce, Ulysses)
A ll these choices go beyond questions o f the startling (like the
craw ling air o f the first quotation) or the unusual (like the deliber
ate archaism o f Jo yces a proper m an o f his person). T h e y do
something w hich is a liberty not norm ally perm itted in other styles
o f the present-day language. In terms o f syntax they must be called
w rong or mistaken selections. H ere the literary style shows another
o f its unique features: the writer masters language below the surface
level and claims the right o f performance beyond the normal
com petence. W hether we applaud or disallow the performance
depends on judgem ents w hich are not those o f the linguist. But if we
applaud, the insights o f the linguist enable us to understand just
w hat it is that we are applauding.
Syntax
57
M iUTH ER R E A D IN G
6
Words and Meanings
59
ilny English, how do we assess the set teach, teaching, teacher, teachable,
in >1.1 nothing o f the change to taught? I f a foreigner learns the form
Ituch and has some knowledge o f methods o f word-formation, how
111.my words has he learned? Even more im portant, how m any words
11 he learned in recognizing as units the sequence o f sounds which
me written dow n as pipe, match, box, balance? For each o f these, and for
111.my other words, the dictionary offers a num ber o f apparently
ililTcrent meanings.
These are some o f the simpler and more obvious problems
ilirrc are m any others w hich confront those who are trying to deal
linguistically w ith words . It is im portant to recognize that they
exist and not to suppose that words can be treated as isolated linguistic
phenomena. T h e traditional m ethod o f language-teaching was
concerned w ith accidence, syntax and vocabulary; and indeed it
worked well enough in the hands o f a good teacher for the
practical acquisition o f a language. T h e structuralists thought
instead o f phonology, gram m ar and semantics, breaking some of
the rigid divisions w hich prevented deeper understanding o f
language as a hum an phenomenon. Chom sky and his followers
prefer to discuss gram m ar as possessing phonological, syntactic and
Ncmantic aspects.
A ll this is o f the greatest im portance in linguistics and m ay
help our present study if only because there is still no definitive
theory o f semantics and it is exciting to follow w hat is being done
in this field. It will not do too m uch harm, however, i f we continue
to use the term w ord and to pursue words in their relationships to
one another. Literary writers in all ages have experienced w hat
T . S. Eliot called, the intolerable wrestle w ith w ords. A lthough
they m ay have formulated no linguistic theories, they knew well
enough that m eaning is not to be sought only a t the level o f the
single word. It is contained in the smaller units as well: in the
affixes, and in the inflections which are few in modern English
but were once numerous.
R ecognition o f m eaning w ithin a smaller unit than the w ord
makes it possible to compose new units w hich w ill themselves be
more readily recognized in their own right. M eaningful neologisms
depend on competence w hich splits the seemingly atom ic word and
takes from it something that still communicates. H ow ever m uch we
m ay dislike neologisms like motorcade or washeteria, however m uch we
deplore the etym ological inaccuracy o f paratroop, w e cannot deny
their semantic function. T h e y take their places in the paradigms
61
63
64
65
66
67
68
6g
jo
F U R T H E R R E A D IN G
ji
7
The Language of Rhetoric
73
75
insistence that written language was the only type w orthy o f serious
study, and o f the raiding o f past literature for examples to support
names o f figures w hich w ere treated as having reality in their own
right.
W ithout departing from traditional nom enclature, or even
traditional practice in adducing literary examples, w e can find the
figurative expressions alive in d aily usage. T h ere is no difficulty
about agreeing with Bloomfield that poetic m etaphor is largely
an outgrowth o f the transferred uses o f ordinary language . I f we
contend that the language o f literature constitutes a style o f the
langue, any viable gram m ar m ust be able to accom m odate the
usages o f literary writers. Conversely, the writers must be open to
the judgem ent o f the gram mar.
W e w ill look at some not all o f the established figures in the
dual relationship o f literature and common usage. O n e distinction
o f types w hich has sometimes been overlooked is o f considerable
linguistic im portance the difference between tropes and schemes
that was generally made b y the great rhetoricians. Tropes depend
essentially on paradigm atic relationships, schemes on syntagmatic.
Most o f the more fam iliar textbook figures o f speech are tropes.
T h ey take us on from the lexical deviations w hich were discussed
at the end o f the previous chapter and for w hich the nam e metaphor
there becam e necessary. T h e y are the result o f unusual choices
from the items which the gram m ar makes available in a given
pattern.
S im ile is the root-notion o f tropes: the com parison derived from
likeness perceived between two referents. T h ere is clearly a very
wide range o f choice here, and the successful literary simile will
point a likeness not usually discerned yet not so far-reached as to be
purely subjective and therefore uncom m unicative. A t least one
item generally refers to something perceptible b y the senses, which
foregrounds the other item b y its actuality. T h e com parison m ay
be directly between noun and noun:
T h y soul was like a star, and dwelt apart:
T h o u hadst a voice whose sound was like the sea.
(W ordsworth, London, 1802)
She smiled as she saw how big his mouth was, and his chin so
small, and his nose curved like a switchback, with a knob at the
end.
(V irginia W oolf, The Voyage Out)
77
y8
8o
and see also W yatts poem A n d w ilt thou leave m e thus? which
repeats this line at the beginning o f each stanza and the line, Say
nay! say n a y! at the end o f each.
A n a d ip lo s is links the end o f one stage to the beginning o f the
next b y repetition, as D onnes sonnets La Corona each take the last
line o f a sonnet to be the opening line o f the next in the sequence.
Ernest Dowsons Flos L un ae opens and closes each stanza with
the line I would not alter thy cold eyes .
E p iz e u x is repeats a w ord or phrase without any break at all:
A n d when he falls, he falls like Lucifer,
N ever to hope again.
(Shakespeare, Henry VIII, IH .ii)
Sun is torn in coloured petals on the water,
T h e w ater shivering in the heat and the north
wind
(R ex W arner, N ile Fishermen )
W e are certainly fam iliar w ith repetition in the syntax o f daily
speech; but w e do not dignify it with technical names, in the w ay
that w e recognize the appearance o f certain tropes. Repetition can
be used consciously for emphasis Its cold outside, bitterly cold ;
or to establish a phatic sense o f sharing I t s a shame,isnt it?
Yes, itsreally a sham e. M ore often, it is used unconsciously and
is associated with users w ho have not a highly developed linguistic
skill. A n y overheard conversation in a public place is likely to yield
the kind o f repetition that Eliot reproduced in The Waste Land:
W hen L ils husband got dem obbed, I said
I didnt m ince m y words, I said to her m y s e lf. . .
Nevertheless, patterned repetition is constantly found in literary
language, and also in the religious register: for exam ple, in litanies.
Its appearance in the ritual and incantatory language o f diverse
cultures cannot be overlooked b y the linguist.
It should now be clear that even the seemingly extreme usages
o f the literary style can be approached with the understanding that
they grow from the com mon core o f language. T h e y do not call for
modifications in the explanatory model o f langue. T h e y do, however,
dem and that we recognize the special dimension o f literature in
which these figures do not appear by autom atic response or by
rapid register-choice. T h ey are planned and given performance as
81
84
FURTHER
R E A D IN G
8
Rhythm and Metre
86
Johnm ustgiveyou/adefiniteanswer/byFriday
or, w ith different emphasis but no change o f basic m eaning :
Johnm ustgiveyou/adefiniteanswer/byFriday
A good m any other arrangements o f pause and stress are con
ceivable for this simple utterance within the bounds o f normal con
versation: it is clear that m uch more than a succession o f sounds is
involved. Elements o f stress, pitch and duration make up the
intonation w hich gives distinctive pattern to the dialects and idiolects
o f an ethnic language. It is chiefly through intonation that w e are
enabled to place a person without necessarily understanding all
that he says, or to parody the features o f a foreign language w hich
we speak im perfecdy.
In graphological realization there is practically no indication o f
intonation; it is silently and subconsciously supplied b y the reader,
or inserted b y the actor or reciter, helped to some extent by punctua
tion. A phonem ic transcription cannot rely on ordinary punctuation
which, although originally an aid to oratory, is incomplete and
uncertain in the complexities o f speech. A m ore delicate system o f
m arking is used for these elements, w hich are the supra-segmental or
prosodic division o f phonetics. T h e word prosodic at once reminds
us that literary critics have for centuries been concerned with the
regular patterns o f verse and have tried to represent them visually
b y various systems o f scansion.
For convenience we can use the word rhythm for the distinctive
but variable pattern in the spoken utterances o f a langue; the
deliberate use o f a regular and recurrent pattern in a literary
composition w ill be called metre. General linguistic study is not
concerned w ith m etre as such, but m etre can be exam ined linguistic
ally in relation to the rhythm o f spoken language. T h e latter is an
extensive study and conclusions about it are b y no means definitive;
a few principles are im portant for the present.
W e can recognize in speech the principle o f equal tim ing or
isochronism, w hich breaks utterances into segments o f approxim ately
87
88
8g
go
R hythm can, and most frequendy does, exist without metre: but
m etre draws its being from the existence o f rhythm .
In the divisions o f literature metre is considered an adjunct o f
verse a term which w ill suffice to m ake the technical distinction
from prose without creating any difficulties about the boundaries o f
poetry. Prose shares the quality o f rhythm , in the sense in w hich w e
have been associating it w ith spoken intonation. A nything w hich is
consistendy written in metre, then, ceases b y definition to be prose;
but the appearance o f occasional metre in prose can be the result
either o f chance or o f design: more often, perhaps, o f a spilling over
from exceptionally sensitive awareness o f norm al rhythm . W e are
most likely to become aware o f m etre in prose w hen reading aloud,
but the pattern m ay be strong enough to present itself even in
silent reading and to invite scansion:
X
poor mortals,/ how ye make/ this E arth bitter/ for each other
(Carlyle, The French Revolution, V , 5)
X
gi
g2
93
|xx
g4
55
Poetic tension demands a line w hich is not too far from speech
rhythm bu t w hich can avoid m onotony or bathos. T h e poets have
in fact subjected the iam bic pentam eter and other lines to a great
deal o f variation: it w ill be found that scansion in the so-called
iam bic foot ( x |) is seldom possible for long stretches and that
w hat we in fact find is a line o f five stresses isochronously separated
by an irregular num ber o f half-stressed or unstressed syllables a
pattern closer to speech and to the old accentual metre. Sim ilar
adaptation to speech rhythm is generally m ade b y those who have
adopted unusual or neo-classical lines such as the loose hexa
meters o f Clough quoted above, or Tennyson in Locksley H all or
Byron in T h e Destruction o f Sennacherib .
T h e role o f prosody is in the performance o f a poem not
necessarily or usually in reciting aloud, when the m etrical pattern
becomes most apparent, but in every encounter w ith it as a piece o f
language draw n from the same com mon stock that provides for
our own performances in everyday speech. A reader who is sensitive
to the intonations o f speech m ay gain most from a poem: conversely,
fam iliarity w ith a nations poetry is one w ay o f becom ing fam iliar
with its spoken nuances.
I f m etre can be said to impose rules on the norm o f rhythm , it can
also break its own rules and use deviation for effect to replace or
gy
| x
x -
III
g8
it from prose. O f course not all free verse is disciplined to this extent:
nor was all regularly m etrical verse poetically inspired. Here are
two examples in which freedom and discipline go together. T h e first
is little if any freer than m uch Jacobean blank verse; the second is
looser, but with the freedom o f carefully planned intonation.
I
X X
x | x x
X X X
IX
I I
- x x
H e rode over
-
F U R T H E R R E A D IN G
gg
9
Beyond the Sentence
101
103
105
107
Also . . . (conjunction 1)
. . . the violence o f the nightm are . . . (looser semantic link w ith
a new m enace 8)
A lso . . . (conjunction 1)
A aro n Sisson . . . (opens new paragraph, with semantic and
form al connection m en/m an 4)
H e was late . . . (pronominal linkage 2)
H e was secretary . . . (pronominal linkage 2)
ro8
iog
there
(it)
it
the W ar
there
that
(there)
was
was
was
was
was
was
(was)
a star
early twilight
Christmas Eve
over
a sense o f relief
almost a new menace
a feeling o f violence
had been
another wrangle
was
(was)
was
was
(was)
the last m an
clim bing the hill
late
secretary
nettled
but twice the progression is broken with the pluperfect form which
switches attention back to the w rangle o f the first paragraph:
A aro n Sisson
had
had
attended a m eeting
heard . . . wrangling
no
in
F U R T H E R RE AD IN G
112
Colloquial: inform al, fam iliar u ttera n ce; ge n e ra lly spoken as con
versation but som etim es w ritten , as in personal letters.
Competence: speaker s total know ledge o f a
a cq u a in ta n ce w ith its system (C hom sky).
lan gu a g e
th rough
Complex sentence: sentence consisting o f one p rin cip le clause and one
or m ore subordin ate clauses.
114
final
consonant
w ith
deductions
to
d ep art from
nbrm al
Foregrounding:
utteran ce.
115
Isochronism: prin ciple o f d ivid in g spoken utteran ces into segm ents
o f a p p ro x im ately eq u a l d u ration .
Keneme: em p ty w o rd , existin g on ly as p a rt o f the gra m m ar o f a
lan g u a g e and m eaningless in isolation.
Kernel sentence: one o f the g ro u p o f basic sentences in a lan gu ag e
from w h ich all oth er scntenccs o f that lan gu a g e m a y be d erived
b y the rules o f tran sform ation al-gen erative gram m ar.
Langage':
M in im u m form al un it o f m eanin g.
Morphology:
lan gu ag e.
Neologism:
Onomatopoiea:
Paradigm:
Performance: a ctu al as opposed to poten tial u tteran ce b y a lan gu ageuser (C h o m sk y ); sim ilar to parole.
Phatic: (lan gu age) used to establish relationship rath er than for
com m u n ication o f m essage.
Philology: trad ition al d iach ro n ic
w ritten texts.
study o f lan gu a g e,
based
on
stu d y o f versification.
stud y o f m eanin g.
Semiology:
ny
Structuralism:
system .
Synchronic:
Synonym:
Syntax:
R e la tin g to printin g.
Index
Addison, Joseph, 16
Aldington, Richard, 107
alliteration, 14, 61, 88-9
ambiguity, i, 2, 64-6, 68
anadiplosis, 80
anaphora, 79
Arnold, Matthew, 3, 53
assonance, 14, 38
Auden, W . H ., 77, 83
Austen, Jane, 42
Austin, Alfred, 54
Bally, Charles, 4, 42
Barthes, Roland, 111
Beckett, Samuel, n o
Betjeman, John, 54
Bible, Authorized Version of, 63
Blake, W illiam, 70, 83, 92
blank verse, 91, 96-8
Bloomfield, Leonard, 4,
c;8
Bright, John, 28
Browning, Robert, 21, 51, 94, 104
Buffon, Georges, 15
Bunyan, John, 3
Bums, Robert, 78
Butler, Samuel, 17
Byron, Lord, 27, 95
Carlyle, Thomas, 90
Carroll, Lewis, 37, 60
Cassirer, Ernst, 28
Chaucer, Geoffrey, 21, 49, 91
Chesterton, G . K ., 69, 82
Chomsky, Noam, 1, 9, 46, 55, 59,
72, 104
Cicero, 78
Clough, A . H ., 93, 95
Coke, Edward, 23
110
discourse analysis, 10 0 -in
Donne, John, 15-16, 80, 92, 94
Dowson, Ernest, 80
drama, 17, 34, n o - 11
Drayton, M ichael, 77
Dryden, John, 21, 68, 76, 104
Eliot, George, 65
Eliot, T . S., 18, 59-60, 79-80, 96
Empson, W illiam, 64
epistrophe, 79
epizeuxis, 80
figurative language, see rhetoric,
figures of
Firth, J. R ., 62
Flecker, James Elroy, 82
foregrounding, 48, 51, 63, 78-9, 87,
93. 108
Index
Forster, E. ., 6g, 83
Fowler, H. W . and F. G ., 61
free verse, 36-7, 96-8
Frye, Northrop, 73
Garvin, F. L ., 48
Gascoyne, David, 98
Genet, Jean, n o
genre, literary, 3, 104, 109-1 1
Gilbert, W . S., 79
graphology, 33, 35-7, 39-41, 44, 55,
85-6, 92-3, 103-4, 106-7
Graves, Robert, 90
G ray, Thomas, 15
Green, Evelyn, 27
Halliday, . A . K ., 48, 100
Hamilton, G . R ., 103
Harris, Z . S., 101, 104
Herbert, George, 37, 67, 103, n o
historical linguistics, 21-30, 37, 50,
66
Hobbes, Thomas, 3
Hopkins, Gerard Manley, 1, 2, 26,
50-51, 88-9, 110
Housman, A . E., 17, 24, 26, 67
Hughes, Ted, 7, 97
hyperbole, 78
idiolect, 18, 46, 86
intonation, 34, 36, 86, 98
isochronism, 86, 95, 97
Johnson, Samuel, 6, 16, 23
Joyce, James, 18, 42, 51, 55-6, 58,
106, I I O- I I
Keats, John, 2, 24, 48, 74, 76, 78,
92, 94
Keyes, Sidney, 70, 76
Kingsley, Charles, 16
Kipling, Rudyard, 76
K yd , Thomas, 91, 93
Langland, W illiam, 88
Lawrence, D. H ., 106-9
legal style, 9 -11, 13, 15, 29, 67-8
L6vi-Strauss, Claude, 111
lexis, 16-17, 22-3, 44-5, 47, 53, 56,
58-70, 75, 83-4, 92, 96, 108
I IQ
95, 104
1 20
Propp, V ., h i
prosody, 63, 85-98
punctuation, 35, 41, 86
Puttenham, George, 73
Quiller-Couch, Arthur, 12
Quirk, Randolph, 84
Racine, Jean, 95
Ralegh, W alter, 23
Reed, Henry, 18, 68
register, i i , 15-19, 23, 27, 34, 37
46, 48-9, 62-3, 66-9, 74, 77, 80-81,
96, n o
religious style, -, 13, 15, 26, 29,
49, 74, 80
rhetoric, figures of, 14, 51, 72-84
rhyme, 14, 24-5, 37, 46, 61, 93
rhythm, 36, 86, 88-92, 94-7
Romantic movement, 16, 64, 73, 88,
103
Rowlands, Richard, 91, 93
Saintsbury, George, 90
Saussure, Ferdinand de, 4, 9, 21, i n
scansion, 36, 86, 88, 90, 93, 95-7
schemes, 75, 79, 83
scientific style, 12-13
Scott, F. S., 44
semantics, 22-3, 47, 52, 59, 62-7,
106, 108
sentence, 33, 44-6, 88, 100-07
Shakespeare, William, 21, 23-4,
26-7, 39. 46, 49-50, 52, 58, 60, 63,
65 67, 73, 78-9, 81-2, 91, 93,
96-7, 102
Shaw, George Bernard, 40
Shelley, P, B., 46
Shirley, James, 78
Sidney, Philip, 52
silence, 35, 104, 106
simile, 51, 70, 75-6, 82
Sinclair, Upton, 72
Skelton, John, 91
Smollett, Tobias, n o
sociolinguistics, I I
speech, 18, 32-43, 46, 48, 52, 58, 74,
80, 85-92, 94-7
Spender, Stephen, 76
Spenser, Edmund, 15, 60, 82
Spitzer, L., 102
Steiner, George, 62
stress, 34, 36, 85-9, 92-3
Stubbes, Philip, 23
style, definitions of, n -1 3 , 25, 73
surface structure, 55-6, 64-5
Surrey, Earl of, 91
Swinburne, A . C ., 46
symploce, 79
synecdoche, 77-8
syntax, 44-56, 58-61, 63-5, 68-9,
75, 79-8o, 83-4, 88, 92-3, 95-6,
103, 108, i n
Tennyson, Lord, 32, 38, 95
Thackeray, W . ., 64, 69, 90
Thomas, Dylan, 1, 2, 26, 50, 56, 70,
82
Thurber, James, 103
Tiller, Terence, 56, 76
transformational-generative gram
mar, 55-6, 104
Trollope, Anthony, 41
tropes, 75, 79-81, 83
Verlaine, Paul, 82
Waller, John, 61
Wallis, John, 30
W arner, Rex, 80
Webster, John, 78
Weinreich, H., 100
Wells, H. G ., 36
Whitman, W alt, 66
Wilde, Oscar, 48
Williams, Charles, 98
Williams, W illiam Carlos, 54
Woolf, Virginia, 42, 75
words, see lexis
Wordsworth, W illiam , 15, 54, 64,
69, 75. 9 1 I04> 110
W yatt, Thomas, 91
Yeats, W. B., 37, 53, 69, 82, 102
Zola, Emile, 72
95p net
Edward Arnold