Language in Literature

The effect of any text is to a very large extent determined by style. In its broadest definition, style is the way in which language is used (see Leech/Short 1981: 10 for a discussion of various definitions of style).

Style is thus not a phenomenon that is restricted to literature; it is necessarily part of any utterance, because for each context one chooses the way one speaks: One uses different vocabulary and probably different syntax when talking to one's granny than one uses when talking to one's examiner in the final oral exam; a report in the newspaper is expected to display – and in most cases does – a different style than a love letter. Obviously, most people are limited in the range of styles they have at their command, and sometimes style expectations are deliberately flaunted: A newspaper report written in the style of a love letter will no doubt cause a certain amount of surprise and thus possibly increase its effect. A love letter written in the style of a report will most likely cause trouble or appear simply ridiculous.

There are utterances where style is used with more deliberation than in others. Political speeches or manifestos may be counted among such, but also most literary texts. In fact, it is partly the calculated and deliberate use of language, or, the special attention the receiver pays to the use of language, that makes a text literary (see What is Literature?).

There has long been disagreement whether style, or form, can be separated from content or not. To put the question another way, are we saying the identical thing when we use different means of saying it, or is the meaning of the utterance (partly) produced by the way we say it? When A.E. Housman, for instance, says "Wake: the vaulted shadow shatters, trampled to the floor it spanned, and the tent of night in tatters, straws the sky-pavilioned land", would it be the same as saying: "Get up mate, it's daylight"? Or is there a difference in WHAT he says because of HOW he says it? Most modern criticism would agree that form and content are not in fact clearly separable, that one is intrinsically dependent on the other, and that a paraphrase never expresses exactly the same thing as the original utterance. Form, in other words, produces meaning. It is thus worth examining how it does that.

When examining the style of a text, one scrutinises mainly two aspects: diction (the choice and use of words) and syntax (the sentence structure). In other words, one examines which words are used and how these words are put together. Closely related to such questions is the use of rhetorical devices. Particularly in poetry and verse-drama one also focuses on the rhythmical patterns and sound effects. The question at the centre of such examinations, is HOW the use of diction, syntax and rhetorical devices produce certain effects and are aimed to evoke certain responses in the reader.

Geoffrey Leech and Michael Short (1981) provide a list of categories for analysis on which most of the following is based. They also give several illuminating examples, demonstrating the use that can be made of style analysis in literary studies (see esp. their ch. 3).


The analysis of diction involves answering a series of questions, all relating to the use of vocabulary: its origin, its effect, its grammatical categories. These could be questions such as:

Is it simple or complex?
Latinate or anglo-saxon?
Abstract or concrete?
Neutral, evaluative, emotionally charged?
Formal or informal?
Vulgar or refined? Any particular sociolect? Any jargon (subject-specific language)?
Appropriate or inappropriate in the context?
Are words used in unusual combinations?
Which lexical categories are used frequently and which are used little or not used at all: nouns, verbs, pronouns, adjectives, adverbs, auxiliaries, and so on.
What rhetorical devices are used on the levels of individual sound or word?

Most important of all remains the question: What effect does the use of diction have in this particular text?

In classical rhetoric styles were classified into three main levels: the grand style, the middle style and the low (or plain) style. Certain types of diction were thought appropriate for certain stylistic levels. This was called the principle of decorum, which was an influential concept well into the eighteenth century. John Dryden, for example, famously agonised over the appropriateness of the word ‘marjoram’ (the herb) for the middle style (he eventually decided it was too low a word). Poetry, perhaps more than other types of literary texts, tends to use words or phrases that are not current in ordinary conversation, so-called poetic diction.


Just like the analysis of diction, the analysis of syntax involves answering a series of questions relating to the use of sentence structure. These are questions such as:

What kind of sentences are used?

Simple or complex?
Long or short?
Paratactic or hypotactic?
Statements, exclamations, questions, or commands?

Is there a type of clause that is preferred?

Relative clause?
Adverbial clause?
Interrogative clause?
Finite or non-finite clause?

How are sentences linked (sentence cohesion)?

Are there cross references and what type?
Are sentences connected with logical links? Or are they purely associative?
Are any particular sentence structures repeated?
Are any particular words repeated which create cross references?

What rhetorical devices are used on the sentence level?

Once again, the most important question is: What effect does the use of syntax achieve? There are no fixed answers to this question. The effect of stylistic devices will differ from text to text and within texts, depending on the immediate context.


Syntactical Deviations

In literary texts generally, and especially in poetry, syntax can differ from everyday usage. There is, on the one hand, a certain amount of poetic licence which makes it quite acceptable for a poet to deviate slightly from ordinary syntax to accommodate the sentence to the line form and metre. Such accommodations can be, for instance, inversions, that is, a change in word order: "The King's real, or his stamped face / contemplate" (Donne, Canonization) instead of 'Contemplate the King's real or his stamped face'.

A deviation from common collocation, the way words are combined with other words, can achieve interesting effects. Geoffrey Leech (1969: 29-31) illustrates this kind of deviation using the title of Dylan Thomas' poem A Grief Ago. Usually, the expression 'ago' is only combined with time measurements: two years ago, an hour ago, a week ago, etc. To combine 'ago' with 'grief' is a deviation from common usage. It is unusual and thus the expression is particularly noticeable, though we still interpret it as a time measurement. The deviation draws the reader's attention to the importance that grief has assumed in the speaker's life; it has become so dominating that it has replaced other time measurements. A rather different effect would be achieved by an expression like 'two wives ago'. This collocation, though one can easily work out its meaning, makes the speaker sound rather frivolous, perhaps a little callous even, since one expects a wife to take a more central part in a man’s life than merely as a time measurement. Also, in our culture it is still unusual to change wives regularly and often enough to make them convenient as time measurements.

Rhetorical Devices



poetic diction
poetic licence