i) The Background to Foregrounding
Jan Mukařovský, one of the founding members of the Prague Linguistic Circle, explores the concept of foregrounding (aktualisace) in a paper entitled ‘Standard Language and Poetic Language’. Mukařovský is never in any doubt that there is a distinction to be made between ‘standard’ language and ‘poetic’ language, and he writes that ‘the violation of the norm of the standard, its systematic violation, is what makes possible the poetic utilization of language; without this possibility there would be no poetry’. We have already seen that the postulated existence of an exclusively literary language has since been thrown into question, but Mukařovský’s work can be regarded in hindsight as a valuable precursor to the work on linguistic register. The identification of ‘the norm of the standard’ is likewise a contentious issue: any attempt to classify language use according to some all-encompassing norm is bound to involve arbitrary decision-making and sweeping generalisations. However, the advent of the technological age and the rapidly growing role of computers in the gathering of linguistic data means that is it becoming increasingly possible to establish the norms of a given text, as Rebecca Posner points out:
computers…are being used mainly for the purpose of compiling concordances and frequency word lists [which] for students of style…provide the solid ground of the ‘norm’, on which the deviation of ‘individual style’ can stand out more clearly.
Although computers have simplified the task of spotting deviation from the norms established within a given text, it remains problematical nevertheless to measure linguistic deviation within a text against any external standard: first, because a norm of standard language use is practically impossible to document exhaustively, and more importantly, because there are convincing arguments against the existence of a single standard language in the first place.
This dubious literary/non-literary distinction notwithstanding, Mukařovský recognised that foregrounding is possible in both poetic and standard language. In standard language foregrounding is subordinate to the function of communication. Conversely, in poetic language,
foregrounding achieves maximum intensity to the extent of pushing communication into the background as the objective of expression and of being used for its own sake; it is not used in the services of communication, but in order to place in the foreground the act of expression, the act of speech itself.
Mukařovský describes the relationship between poetic and standard language in terms of three positions on a language continuum. At one end of the scale there exists no distortion of the language, although Mukařovský is careful to note that this too represents a stylistic choice. At the mid-point of the scale, the language is distorted to a certain extent, but communication is still the dominant function of the text. At the far end of the scale, the language is distorted, and communication is now subordinated to aesthetic function.
Mukařovský complicates the notion of foregrounding by claiming that in poetry there are in fact two norms, that of standard language, and that of the aesthetic canon: Mukařovský argues that poetic language is not a special brand of the standard because it has at its disposal all forms of the given language in addition to its own lexicon, phraseology and grammatical forms. Mukařovský concludes his paper by noting that foregrounding contributes to richness of expression, and that it subsequently enriches the language. He notes that in standard language, foregrounding brings to the fore ‘the essence of sentence meaning and the dynamic nature of sentence construction…the meaning of a sentence appears as the total of the gradually accumulated meanings of the individual words’. Poetry, on the other hand, can foreground the relationship between the individual words and the subject matter of the sentence, and the semantic interrelationships of the words in the sentence – every word affects every other.
In sum, Mukařovský’s argument runs as follows. There is a distinction between standard and poetic language, and standard language provides the background against which poetic language can be thrown into relief, whereas poetic language employs its own established canon as well as borrowing from standard language. Foregrounding can occur in both standard and poetic language, but its function in literary texts is to highlight the expression itself, whereas in standard language foregrounding serves a communicative function.
Mukařovský’s ideas have since been adapted and revised. The idea of a specifically literary language is no longer popular, and the precise way in which literary language functions has been more rigorously defined than it is in Mukařovský’s three-point scale. It is also no longer fashionable to claim that literary texts do not communicate content, and that they exist purely as aesthetic objects; there is an element of truth in the claim that the function of foregrounding in literary texts is to highlight the expression, but we are perhaps less likely now to dismiss the idea that literary texts can communicate or that non-literary texts can have elements of literariness. Finally, the claim that standard language acts as background for literary language has to be re-evaluated: instances of foregrounding can only confidently be identified within the confines of the given text, because the only standard that can be established is that of the text. Any comparison to norms of language usage outside of the text will ultimately rest on subjective assumptions and inconclusive data. This final point can, I think, be cited in support of the claim that ‘form’ must relate to the whole text: the text, not the language as a whole, is the background against which instances of foregrounding can be identified and from which they derive their meaning.
It should be noted that Mukařovský refers specifically to poetry in his paper, but according to Leech and Short, foregrounding can also act as an aid to clarifying the distinction between transparent and opaque prose writing. The greater the number of foregrounded elements, the more opaque the text:
the aesthetic theory of foregrounding or de-automatization enables us to see the references to TRANSPARENT and OPAQUE qualities of prose style…as more than vague metaphors…prose is opaque in the sense that the medium attracts attention in its own right; and indeed, the interpretation of sense may be frustrated and obstructed by abnormalities in the use of the lexical and grammatical features of medium…opacity can be equated with the extent to which the reader is required to be creative.
The distinction between transparent and opaque will not always be a clear one, however; in reality, it is likely that texts will occupy positions on a sliding scale. Leech and Short begin their exploration of foregrounding and its implications by distinguishing between literary relevance (foregrounding), psychological prominence and statistical deviance. Deviance is defined as a ‘purely statistical notion…the difference between the normal frequency of a feature, and its frequency in the text or corpus’. Psychological prominence refers to those features which register in the mind of the reader. Foregrounding is not the same thing as either prominence or deviation, because the foregrounded items must in some way be artistically relevant to other foregrounded items and finally to an interpretation of the text as a whole: ‘we should be able to see a prominent feature of style as forming a significant relationship with other features of style, in an artistically coherent pattern of choice’. A prominent feature may not have any literary function; prominence may be due to a writer’s preference for short sentences, or monosyllabic words, for example. Therefore, ‘the dividing line between foregrounding and unmotivated prominence must be drawn in principle: where it is drawn in practice depends on a coherent literary interpretation of style’.
Leech and Short also differentiate between qualitative foregrounding and quantitative foregrounding. Qualitative foregrounding is to be found in deviation from the language code itself, when the writer breaches some rule or convention; quantitative foregrounding is the deviation from some expected frequency. It should be noted here that an unusual or noticeable frequency does of course include examples of absence as well as presence: for instance, the absence of a rhyme in a line of poetry where there should have been one – thereby defeating reader expectancy – or the sudden shift away from a prominent pattern of parallel structures in a prose text. Of course, one form of foregrounding will eventually shade into the other: ‘the quantitative foregrounding…of a prominent pattern of choices within the code shades into the qualitative foregrounding which changes the code itself’. Is the lack of finite verbs in the Peake passage discussed in the following section an example of qualitative or quantitative foregrounding? It is more likely that it begins as one and ends as the other. One expects to see finite verbs in prose text, after all, and their initial absence will undoubtedly strike the reader as odd; as the passage progresses, however, their absence becomes the norm, and any finite verb introduced at this stage will itself be foregrounded as deviant from the code already established by the writer. I follow my discussion of the passage from Gormenghast with an exploration of a passage from a relatively more transparent text, Emily Brontë’s Wuthering Heights.
ii) Semantic Density in Gormenghast
It is clear from the opening lines of Gormenghast that the reader is dealing with a unique text. It is a text that quickly establishes its own norms, and these norms can be briefly summarised as follows. Extra levels of meaning are created through a concentration of parallel structures, as detailed in Appendix B. Lexical items are linked through sound patterns, notably alliteration and pararhyme. Peake regularly breaks grammatical rules; for example, his text contains examples of the omission of expected items, and the use of grammatical ellipsis or deletion creates a sequence of fragmented sentences, particularly in the opening paragraphs. In addition, instances of nominalisation account in part for the noticeable lack of finite verbs. At word level, content words are shifted out of their more usual function, or they are employed in a slightly unusual sense or in a sense that is tricky to place. Also prominent is Peake’s idiosyncratic use of punctuation, the colon and semi-colon in particular. And yet I would argue that almost all the above constitute instances of foregrounding, because there is little here that cannot be justified in thematic terms. To alter the form of Peake’s text is to lose something of its content.
In the following analysis, I have examined each of the seven paragraphs in question individually. Each paragraph is numbered and reproduced in full preceding its discussion for ease of reference, and the text in its entirety is reproduced in Appendix A. As noted above, Appendix B is an attempt to represent visually the parallel structures within the text; Appendix C is the result of my attempt to paraphrase these opening paragraphs as part of an investigation into whether the form could be altered without affecting the content. I would argue that the semantic density of this text renders futile any attempt at paraphrase.
(1) Titus is seven. (2) His confines, Gormenghast. (3) Suckled on shadows; weaned, as it were, on webs of ritual: for his ears, echoes, for his eyes, a labyrinth of stone: and yet within his body something other – other than this umbrageous legacy. (4) For first and ever foremost he is child.
This paragraph consists of four sentences numbered (1) to (4) above. The first and second sentences contain three words, the third, 34, and the fourth, eight. The opening and closing words of this paragraph are parallel in structure (proper noun/pronoun – copula verb is – adjective/enumerator):
Titus is seven S V C (subject –verb– subject complement)
he is child S V C
The omission of a determiner in the final phrase (‘he is child’, not ‘he is a child’) has a twofold effect. The parallelism thus created between this phrase and the phrase which opens the novel – ‘Titus is seven’ – links the two statements in the mind of the reader, thereby reaffirming Titus’ status as a child. The word ‘child’, which usually functions as a noun, arguably functions here as an adjective, as does the enumerator, the cardinal number ‘seven’, the effect of which is that the word ‘child’, italicized in Peake’s text for emphasis, describes Titus rather than simply telling the reader what he is, almost as if Titus is an embodiment of, or is representative of, every child in the world – an effect which becomes significant later in the extract. It is Titus’ status as child which forms the basis of his rebellion against the oppressive atmosphere of the castle. Peake’s use of parallel structures foregrounds this important information for the reader.
The first sentence also forms a parallel with the second, in that both consist of just three words. The second sentence is elliptical: ‘His confines, Gormenghast.’ The second word ‘confines’ could be interpreted in either an abstract or a concrete sense, depending upon whether the confines referred to are the actual stone boundaries of the castle or alternatively the restrictions and limitations in Titus’ mind, in that Gormenghast is all he knows or all that he can conceive of. However, the third sentence would appear to reinforce the abstract interpretation of ‘confines’ in its description of Titus’ upbringing and the way in which he interacts with and accesses the world available to him through his physical senses.
Peake’s rich prose does not lend itself easily to paraphrase, not only because several differing but equally valid interpretations of a sentence or phrase are occasionally possible, but also because any paraphrase is likely to lose the ‘poetic’ qualities of the original text, finishing as merely a pale reproduction. I have noted in Appendix B the many parallel structures which contribute to the rhythm of Peake’s text, and in Paragraph 1, as in the other paragraphs under analysis, there are many subtle phonological effects which one would also not wish to lose. Note the alliteration in ‘Suckled on shadows; weaned, as it were, on webs of ritual’, and the echo in ‘for his ears, echoes, for his eyes…’. The italicized phrases echo each other, and the meaning of the text is thereby reinforced in its form. The word ‘other’ occurs twice, providing another echo in the passage, and the sound pattern of the word ‘for’ is repeated in the final phrase ‘For first and ever foremost he is child’.
In writing my very literal paraphrase of this opening paragraph, I noticed that many deep structure verbs had been suppressed in their surface structure manifestations. The only verbs remaining are the copula ‘is’ – which, in its insistence on Titus’ state of being, perhaps serves to foreshadow Titus’ rebellion – and ‘suckled’, ‘weaned’, both of which place Titus in the role of beneficiary instead of agent. Titus is not allowed to be active: he does not hear the echoes or see the maze of stone – they are provided by the castle for his ears and eyes. Titus is, but he does not act.
(1) A ritual, more compelling than ever man devised, is fighting anchored darkness. (2) A ritual of the blood; of the jumping blood. (3) These quicks of sentience owe nothing to his forbears, but to those feckless hosts, a trillion deep, of the globe’s childhood.
Sentence (1) consists of twelve words, and a subordinate clause is embedded within the main clause: ‘more compelling than ever man devised’. This clause informs the reader that the ritual referred to in this paragraph is not, or is unlikely to be, one of the rituals observed at Gormenghast, all of which have been devised by Titus’ ancestors. The present continuous verb form ‘is fighting’ locates the passage firmly in the immediate present; a battle is at this moment raging against the enemy, the ‘anchored darkness’. This enemy is fixed and immovable, ‘anchored’ in the liquid of Titus’ blood, which by contrast moves and jumps.
The nine words of the fragmented second sentence tell us that this ritual is ‘of the blood’; this phrase is paralleled in the words following the semi-colon, ‘of the jumping blood’. The parallelism takes the form of the repetition and embellishment of a noun phrase; this repeated phrase ‘of the blood’, therefore, is the pulse of the sentence, and the adjectival participle ‘jumping’ represents one of the ‘quicks of sentience’ mentioned in (3). This effect is reinforced by the rhythmical stress of the phrases:
of the blood
of the jumping blood
where the regular heavy stress on ‘blood’ is interrupted by an additional stress on the first syllable of ‘jumping’.
The third sentence in Paragraph 2 is once again the longest, consisting as it does of 21 words, a coordinating construction (‘but to’) and a subordinate clause. There are some difficult words here: ‘quick’ is not listed as a noun, but in Peake’s text its function is clearly that of a collective noun. I have assumed that Peake meant the reader to understand the ‘quicks of sentience’ as referring to the jumps in Titus’ blood mentioned in the previous sentence, if one is to assume that Peake has followed the rules, or guidelines, for the construction of a coherent text – namely, that the subject should not be randomly altered without giving the reader some indication of the change. My assumption is reinforced by the use of the determiner ‘these’ as an anaphoric reference: ‘these quicks of sentience’ refers back to ‘the jumping blood’ of the previous sentence. I have paraphrased ‘quicks of sentience’ as ‘flashes of self-knowledge’, to indicate Titus’ growing awareness of himself as a child, rather than as the Earl of Gormenghast. By contrast, the word ‘feckless’ (‘feeble, futile, ineffective, aimless, irresponsible’) seems to undermine the force of these ‘quicks’. The hosts are ‘irresponsible’ because in encouraging Titus to rebel, they are working against the castle. Alternatively, the hosts are ‘feeble’ because Titus is as yet immature – or there is always the possibility that we are meant to understand both senses. Another tricky word to place is ‘hosts’; but the phrase ‘a trillion deep’ would seem to suggest that ‘hosts’ should be interpreted in the sense of a large multitude.
(1) The gift of the bright blood. (2) Of blood that laughs when the tenets mutter ‘Weep’. (3) Of blood that mourns when the sere laws croak ‘Rejoice!’ (4 ) O little revolution in great shades!
In common with the preceding paragraphs, this paragraph consists of a series of fragmented sentences. It ends with an exclamative, and is perhaps the most ‘poetic’ of the paragraphs to be analysed. The second and third sentences in particular are poetic in that they are almost a mirror image of each other in their structure: Of blood that – verb phrase – when the – noun phrase – speech verb – imperative in direct speech. This whole paragraph presents the reader with six verbs in four lines, a relatively dense concentration of verbs, presented as forms of opposites and contrasts (laugh/mourn; weep/rejoice). This sudden proliferation of action verbs may be emblematic of Titus’ growing state of defiance: the imperatives issued by the laws and tenets are disregarded and mocked by the rebellion in Titus’ blood. Titus’ rebellion – referred to as a ‘revolution’ in (4) – is still ‘little’, opposed as it is to the ‘great shades’. And how are we to understand ‘shades’? Is Peake referring to the shadowy corners of Gormenghast castle, or to the ghosts of Titus’ numerous ancestors – or both? And are we to understand ‘revolution’ as a rebellion, a change in the state of affairs, or as directionless turning in a circle? For Titus cannot escape Gormenghast – having left its boundaries, he is later compelled to return – albeit briefly – at the end of the trilogy. As the Countess prophesies at the end of the second novel, shortly before Titus leaves the castle: ‘You will only tread a circle, Titus Groan. There’s not a road, not a track, but it will lead you home. For everything comes to Gormenghast.’
(1) Titus the seventy-seventh. (2) Heir to a crumbling summit: to a sea of nettles: to an empire of red rust: to rituals’ footprints ankle-deep in stone.
This paragraph consists of only two sentences and the structure here is very similar to that of the first paragraph in that both begin with the proper noun ‘Titus’. The first paragraph told us Titus’ age, but now we are given his title: ‘the seventy-seventh’. Leaving aside the obvious parallelism in the chosen enumerators (seven: seventy-seventh), the emphasis switches from Titus’ status as a child to his status as the Earl of Gormenghast. The focus of the text has likewise shifted away from Titus’ growing feelings of rebellion, and moved toward the castle itself – that which Titus has inherited from his ancestors, as opposed to that which he has been given by the other children of the world. The whole phrase forming the first fragmented sentence serves as a proper noun: it is an example of ellipsis – the phrase in its entirety should read ‘Titus is the seventy-seventh Earl of Gormenghast’. The phrase reinforces Titus’ role as Earl, immediately following two paragraphs which describe his rebellion as child. Titus is seven years old – but he is also the seventy-seventh Earl of Gormenghast.
There is a strongly marked parallelism in the second sentence:
Heir to a…
to rituals’ footprints
The images which follow in each of the succeeding phrases are those of decay and things overgrown. The image of the sea (‘a sea of nettles’) is recalled when the reader reaches the word ‘footprints’ a little later, but the footprints are not those in sand which are quickly erased by the sea – these footprints are ‘ankle-deep in stone’. The rituals have left their mark indelibly on the castle.
This fourth paragraph is followed by the single word ‘Gormenghast’, the proper noun forming a paragraph all on its own. In this short extract, then, three paragraphs have begun with proper nouns, as follows:
Titus (is seven)
Titus the seventy-seventh
The parallel structure exemplifies the power-hierarchy. Titus the seven-year old child is subject to Titus the seventy-seventh Earl of Gormenghast, and even the Earl himself is subject to the timeless power of the castle. The structure demonstrates where the power lies – with Gormenghast, a force so powerful that it is granted a paragraph all to itself.
(1) Withdrawn and ruinous it broods in umbra: the immemorial masonry: the towers, the tracts. (2) Is all corroding? (3) No. (4) Through an avenue of spires a zephyr floats; a bird whistles; a freshet bears away from a choked river. (5) Deep in a fist of stone a doll’s hand wriggles, warm rebellious on the frozen palm. (6) A shadow shifts its length. (7) A spider stirs…
This is the longest paragraph of the opening passage, consisting of seven sentences. The focus is still on the castle. The first sentence contains a combination of animistic and humanizing metaphors to personify the castle, which is given a sulky and faintly malevolent personality (withdrawn/broods). The internal parallelism of (1) after the first colon is reinforced by the repeated ms in ‘umbra’ and ‘immemorial masonry’ and the (almost!) pararhyme in ‘towers’ and ‘tracts’. The answer to the question posed in (2) is seemingly anomalous because the question – ‘Is all corroding?’ – is ostensibly about the castle, but the answer is concerned with forces of nature (zephyr/bird/freshet) unconnected with the castle – until, that is, the reader comes to Titus once again, wriggling his ‘doll’s hand’ within the ‘fist of stone’. The question and answer imply the presence of a narrator and a narratee; Gormenghast itself is undoubtedly one of the characters of the trilogy, perhaps the most interesting character of them all. Peake’s narrative voice is often situated within the consciousness of one of the characters, and more often than not the narration is given from the castle’s point of view. Manlove comments that ‘Gormenghast, considered both as place and society, is the most important character’. In considering this short extract alone, it is interesting to note that the number of fragmented sentences rapidly diminishes once the subject turns from Titus to the castle. The castle is complete – hence the full sentences – but the immature Titus is not. Peake’s syntax is in this respect iconic.
The themes of rebellion and escape are quickly picked up again after the interrogative of (2), and here we find a succession of action verbs: ‘floats’, ‘whistles’, ‘bears away from’. The wind, the birds and the streams do not answer to Gormenghast: they are not subject to the castle’s influence, unlike Titus, whom we meet again in (5). He is no longer Titus the seventy-seventh; he is no longer even a person, but a wriggling ‘doll’s hand’, tiny and ineffectual against the ‘fist of stone’, the larger hand which holds him fast. But he is active – he ‘wriggles’ – and he is ‘warm’ and ‘rebellious’ where the castle is ‘frozen’. But the castle is clearly not going to give in without a fight; (6) once again reinforces the images of stagnation and inactivity within the castle grounds: ‘A shadow shifts its length.’ The length of the shadow changes because time is passing – the shadows grow as the day wears on and the sun moves through the sky; what appears to be a verb of movement here (‘shifts’) actually is not at all – it is the sun that moves, not the shadow, nor whatever it is that is casting the shadow. Finally, the paragraph ends with a verb of movement after all – the spider ‘stirs’ – but with the image of the spider comes the association of cobwebs and dust. By the end of this paragraph it seems that the two adversaries are evenly matched. Titus is diminished in physical stature and located deep at the heart of the castle, but his eventual rebellion has been clearly marked in the text; Gormenghast is immeasurably huge, an inexorable power, but the freshet made its escape from the ‘choked river’, just as Titus will make his escape from the vast decaying mass of the castle. The final word is perhaps reserved for the castle: ‘And darkness winds between the characters.’ How is one to paraphrase this? Leech and Short suggest that ‘characters’ could refer to the characters of the text – the letters of the alphabet. But for the moment it would seem that the castle continues to exert its power over its occupants.
At the end of the opening passage, the reader has been introduced to the two adversaries: the enormous sprawling environs of the castle, with its weighty phrases and lexical items concerning stone and shadows, and the child Titus, with his action verbs and his tiny wriggle of rebellion. Peake foregrounds the theme of rebellion by a number of means. In the very first paragraph the omission of an expected article emphasises Titus’ childlike status and subsequent phrases relating to Titus’ various roles are placed in a series of parallel structures emblematic of the power hierarchy. At this stage, the power lies with the castle. Gormenghast is described in prepositional phrases separated by colons and semi-colons which give each phrase the weight of a sentence: ‘to a crumbling summit: to a sea of nettles: to an empire of red rust’. The effect of this is to reinforce the castle’s immense physical weight and its heavy influence on the lives of the inhabitants. Titus’ immature status is embodied in the elliptical sentence structure associated with him but there is a sense of growing rebellion in the description of his ‘jumping blood’; here the form arguably enacts the content. Peake’s use of foregrounding therefore clearly signposts Titus’ eventual rebellion at word, sentence and discourse levels.
iii) Foregrounding in Wuthering Heights
The opacity of the opening passage of Gormenghast requires the reader to actively construct meanings. I turn now to a comparatively transparent text, Brontë’s Wuthering Heights. I hope to demonstrate that foregrounded items in this text also require the reader to be an active participant rather than a passive recipient. In the following analysis I have focused on anaphoric reference and parallel constructions, the impersonal pronoun coupled with the use of body parts as actor, and finally the use of parenthetical constructions.
The passage from Wuthering Heights I have chosen to explore is volume one, chapter three, in which Lockwood passes the time before falling asleep reading the words Cathy has written; first those carved into the couch itself, which take the form of her given name followed by the three surnames she imagined herself to possess, and second, the words she has scrawled into the margins of those books which constitute her library. These words recount the events of an ‘awful Sunday’ (p. 20), in which Cathy and Heathcliff are compelled to listen to Joseph’s three-hour homily.
Lockwood’s first dream is a re-enactment of Cathy’s ‘awful Sunday’, in which Joseph is confused with the preacher Jabes Branderham and Lockwood himself takes the place of the child Cathy, forced to listen to a sermon consisting of ‘four hundred and ninety parts’ (p. 23). At the commencement of the dream, Joseph and Lockwood are likened to pilgrims: Joseph carries a ‘cudgel’ which he refers to as a ‘pilgrim’s staff’ (p. 23), and he admonishes Lockwood for not having one. In parallel constructions separated by a semi-colon (pronoun + auxiliary verb + main verb), Brontë changes the main verb from ‘going’ to ‘journeying’:
I was not going there;
we were journeying to hear… (p. 23)
Joseph, described as Lockwood’s ‘guide’ (p. 23), was initially supposed to be showing him the way home, but Lockwood is now a pilgrim on a journey. In the paragraph which follows, the pilgrims become potentially condemned men: ‘either Joseph, the preacher, or I…were to be publicly exposed and excommunicated’ (p. 23). These events are reminiscent of the children’s admonition at the hands – or tongue – of Joseph, and his assertion that ‘owd Nick’ (p. 22) would come to fetch them. Joseph’s three-hour homily is magnified in Lockwood’s dream to a sermon of prodigious length, each part of which deals with ‘odd transgressions’ (p. 23) or sins of which Lockwood was unaware; perhaps this reflects Cathy’s mystification at Joseph’s anger with the children for making themselves ‘snug…in the arch of the dresser’ (p. 21).
The Lockwood-Cathy/Jabes-Joseph confusion is made clear to the reader in the following section of Lockwood’s narrative:
Oh, how weary I grew. How I writhed, and yawned, and nodded, and revived! How I pinched and pricked myself, and rubbed my eyes, and stood up, and sat down again, and nudged Joseph to inform me if he would ever have done! (p. 23)
The pronoun ‘he’ could be understood to refer to either Jabes or Joseph himself, the previous reference to Jabes being sufficiently distant from this passage to render it possible that it is Joseph to whom Lockwood refers. Halliday and Hasan note that ‘where the cohesive element is something like he or one, which coheres by direct reference to, or substitution for, another item, the presupposed element is typically a specific item in the immediately preceding sentence.’ It is significant, I think, that the last reference to ‘he’ which refers directly to Jabes Branderham is four sentences distant from the ‘he’ of Lockwood’s complaint; ‘Jabes’ appears five sentences distant. It is plausible, therefore, that Joseph and Jabes could be confused in that ambiguous pronoun, and that this effect is intentional on the author’s part. Joseph has taken the place of the preacher, and Lockwood, his head still full of Cathy’s words, has taken the place of the recalcitrant child. The parallel constructions
how weary I grew
How I…-ed, and…-ed, and…-ed, and …ed!
How I…-ed and …-ed…and…-ed…
and stood [verb –ed] …,
and sat [verb –ed]…,
and…! (p. 23)
mimic the fidgeting and complaining of a child. These constructions – the use of ‘how’ to indicate the extent of an emotion or action and the use of ‘and’ as a conjunction to bind together a string of verbs, coupled with the use of exclamation marks to emphasise strong emotion and to provide the reader with an indication of the tone in which the text is designed to be read – are strongly reminiscent of Cathy’s written style in her makeshift journal:
How little did I dream that Hindley would ever make me cry so!…
Hindley… won’t let him sit with us,
nor eat with us any more;
and, he says,
he and I must not play together,
and threatens to turn him out of the house… (p. 22)
Lockwood’s complaint, as Branderham is about to embark on the ‘First of the Seventy First’ (p. 24), is the accusation of a child, that of being bored by an adult who has forced the child to sit still and listen, as Cathy is forced to groan and shiver her way through Joseph’s homily. Jabes Branderham responds with the complaint that Lockwood ‘didst…gapingly contort thy visage’ (p. 24) – an adult’s complaint that a child has been pulling faces, or yawning, instead of paying attention. Once again, the reader is given the impression that Lockwood has changed places with Cathy, and Joseph with Jabes Branderham.
Lockwood falls asleep for a second time, only to dream that he is visited by Cathy’s child-ghost, a visitor he refuses to acknowledge as human; this refusal is made evident in his use of the impersonal pronoun to refer to the figure at the window, and his repeated use of body parts as actor in material processes.
The poor light obliges Lockwood to rely on his other senses. The child’s face is distinguished only ‘obscurely’ (p. 25), and the focus throughout this second dream is on that which Lockwood can feel and hear. Lockwood uses the impersonal pronoun ‘it’ to refer to Cathy’s voice, ‘ “Catherine Linton,” it [‘a most melancholy voice’] replied shiveringly’, her wrist, ‘I pulled its wrist on to the broken pane, and rubbed it to and fro’, the creature, ‘still it wailed’, and Cathy’s hand, or the grip of her hand, ‘and maintained its tenacious gripe’ (p. 25). Not once does Lockwood refer to ‘she’, or ‘her’, or even ‘the child’. Cathy is an amalgamation of disassembled face, hand, wrist, fingers, and voice; she is kept distant from the world of the living by Lockwood’s refusal to acknowledge her as anything other than a ‘creature’, and this effect is compounded when coupled with Lockwood’s use of elegant variation to describe Cathy once Heathcliff appears on the scene. Cathy is, in turn, ‘the little fiend’, ‘that minx’, ‘a changeling’ and a ‘wicked little soul’ (p. 27); she is also linked with the witches of Macbeth in Lockwood’s later reference to the ‘brindled’ cat ‘Grimalkin’ (p. 29). In Lockwood’s eyes she is a malevolent sprite, in spite of Cathy’s description of herself as a ‘waif’, homeless and helpless, and lost on the moors.
To recap, then: a calculated use of foregrounded anaphoric reference is in evidence once again, but the intention this time is not to confuse the characters in the reader’s mind, but to place as much distance as possible between the disturbed sleeper and the wanderer on the moors. Lockwood’s persistent use of the impersonal pronoun it to refer to Cathy’s ghost, or parts of her ghostly body, has the double effect of distancing the ghost from the human beings – Lockwood and the reader – and of underlining Lockwood’s callousness towards a child in obvious distress, a female child who has lost her way on the moors, and on a cold and wintry night, at that.
Parenthetical constructions are very much in evidence in both dream-sequences, and their role in each is to remind the reader that Lockwood is simply recounting his nightmares, and that the events described have not actually taken place in the fictional ‘waking’ world. In his dream of Joseph and the preacher there are two such constructions, the second comprising two lengthy sentences (106 words in total). The first of these interruptions to Lockwood’s narrative is as follows:
Alas, for the effects of bad tea and bad temper! what else could it be that made me pass such a terrible night? I don’t remember another that I can at all compare with it since I was capable of suffering.
I began to dream… (p. 22)
The reader is notified, therefore, that the events to follow are imaginary. The second interruption is the longest, consisting as it does of a description of the chapel: ‘I have passed it really in my walks…’; the adverb ‘really’ is also intended to act as an indication to the reader that Lockwood’s walks are part of the ‘real’ world, whereas the particular journey he is currently relating is not. This description comes to an end with the words ‘one penny from their own pockets’ (p. 23). Although the digression is long, Lockwood’s words do not destroy the gloomy mood of the passage, referring as he does to the ‘embalming on the few corpses’ lying in the swamp. The parenthesis is brought to an end when Lockwood brings us back to his narrative with the words, ‘However, in my dream,’ (p. 23) and the narrative continues until finally he is woken by the fir-tree’s taps on the window. In his dream of Cathy’s child-ghost, the reader is kept conscious of the fact that Lockwood is dreaming by the scattering of parenthetical phrases such as ‘This time’, ‘I thought’, ‘when awake’, ‘the intense horror of nightmare came over me’, and, most tellingly of all, ‘why did I think of Linton? I had read Earnshaw twenty times for Linton’, until finally we come to ‘I discovered the yell was not ideal’ (pp. 25-26). Brontë, through these parenthetical constructions, has gone to some trouble to distinguish between Lockwood’s dream-world and the ‘real’ world of Wuthering Heights, and the reader is discouraged from confusing the two during Lockwood’s narration of his nightmares. However, it is a different state of affairs once Lockwood is awake, and the effect is rendered more startling by these earlier attempts to convince the reader of the fictional nature of his dreams. Heathcliff, of course, is beside himself with anguish, believing as he does that Cathy’s ghost has returned. Lockwood, too, seems to be convinced that his dream was real. He tells Heathcliff that the room is ‘swarming with ghosts and goblins’, and in reference to Cathy’s ghost he asserts that ‘If the little fiend had got in at the window, she probably would have strangled me!’ (p. 27). In addition, let us not forget that the ghost identified herself as Catherine Linton – a fact already remarked upon by Lockwood. After all, Cathy did not become Catherine Linton until she married Edgar, but Cathy’s child-ghost clings to the name she bore when she died. As Catherine Linton, she is denied both the Heights and Heathcliff’s presence.
By first convincing the reader that Lockwood’s narrative is really only a dream, and then confounding the issue by having the characters behaving as if the aforementioned events had really taken place after all, coupled with the anomaly of the ghost’s name, Brontë leaves the reader in some doubt as to whether Cathy’s ghost did or did not come that night. In any case, the effect on Heathcliff is that which is important. John Hagan has commented as follows on this passage:
when Lockwood arrives at Wuthering Heights and dreams of Catherine’s ghost at the window, all of Heathcliff’s anguished yearning is revived…. Heathcliff is convinced that his visitor has really seen the spectre he himself had hoped to see for all those eighteen years. Most significantly, it is on precisely this delusion that Emily Brontë arranges for the dénouement of the novel to hinge, for from this point onward Heathcliff can think of nothing but joining Catherine in death.
It is certainly arguable that Heathcliff loses all lust for revenge once he hears Lockwood’s tale of ghosts and goblins.
Lockwood’s dream of Cathy’s child-ghost is one of several episodes in the novel in which Brontë balances the supernatural with the plausible. We cannot know for sure whether or not Cathy’s ghost did come to seek Heathcliff that night, nor can we be sure that the ghosts of Heathcliff and Cathy do not ‘walk’ at the end of the novel, despite the assertions of Lockwood and Nelly Dean to the contrary. Does Heathcliff really see Cathy herself shortly before his death, or is her appearance simply an hallucination, an effect of his self-imposed fast?
To sum up: foregrounded anaphoric reference indicates that the participants in Lockwood’s first dream have changed places; certainly the Joseph-Jabes switch is made evident, and thus the Lockwood-Cathy switch can be inferred. The anaphoric ‘he’ is foregrounded because the reader will be forced to consider who ‘he’ is. McHale notes that a character, unlike the narrator, is not subject to the cohesive rules of the text and does not have to supply a referent. It is unclear whether Lockwood is referring to Joseph or Jabes, but I have argued that this is not a failure on Brontë’s part but rather an intentional poetic effect allowing the switch between the characters to take place. The resulting confusion of Lockwood with Cathy explains how Cathy and not Lockwood comes to be accused of adultery. The foregrounded parenthetical constructions which represent Lockwood’s intrusions into his own dream narrative support the carefully constructed fabric of the supernatural versus the plausible which is woven throughout Brontë’s novel – a fabric which allows the more romantically inclined reader to believe that Cathy and Heathcliff do indeed walk the moors together once Heathcliff has joined Cathy in the quiet grave.
In this chapter, we have seen many instances of foregrounding, both qualitative and quantitative: Peake’s text in particular abounds in broken rules and unusual constructions which amount to examples of qualitative foregrounding, or breaches of the language code itself. The parallel constructions that constitute quantitative foregrounding, or deviation from an expected frequency, in Gormenghast contribute to the semantic density of the text: new meanings are created as the reader is required to interpret the links forged between the connected items.
We have also seen examples of prominence – sound patterns and parenthetical constructions – which are noticeable to the reader, but which lack thematic significance and therefore do not constitute an instance of foregrounding. We see how a collaboration between linguist and critic could prove useful: a linguistic description of a text can reveal what rules have been broken or what is unusual about a particular utterance; the critic can then use this information in the formulation of a coherent interpretation of a literary text, because it is only those examples of deviation and prominence that show thematic significance which can accurately be described as foregrounding. The theme of rebellion in Gormenghast is marked in foregrounded items such as parallel structures and elliptical sentences. Brontë’s calculated use of anaphoric reference is important thematically because the character-switch between Lockwood and Cathy must be effected so that Cathy, and not Lockwood, can be seen to be charged with adultery: this point will be explored in more detail in chapter four.
Needless to say, foregrounding is lost in paraphrase, as I discovered in my attempt to paraphrase the opening paragraphs of Gormenghast. Foregrounding is not just textual decoration that can be discarded: foregrounded items support, reinforce, or even introduce the ideas or themes of a text, and as such constitute a vital component of content. To disregard foregrounded items, to exclude them from a paraphrase, to change the form in this way, necessarily entails a loss, or a change of content.
CONTINUED IN CHAPTER TWO: SPEECH AND THOUGHT REPRESENTATION
Mervyn Peake’s Gormenghast (1): Extract
(1) Titus is seven. His confines, Gormenghast. Suckled on shadows; weaned, as it were, on webs of ritual: for his ears, echoes, for his eyes, a labyrinth of stone: and yet within his body something other – other than this umbrageous legacy. For first and ever foremost he is child.
(2) A ritual, more compelling than ever man devised, is fighting anchored darkness. A ritual of the blood; of the jumping blood. These quicks of sentience owe nothing to his forbears, but to those feckless hosts, a trillion deep, of the globe’s childhood.
(3) The gift of the bright blood. Of blood that laughs when the tenets mutter ‘Weep’. Of blood that mourns when the sere laws croak ‘Rejoice!’ O little revolution in great shades!
(4) Titus the seventy-seventh. Heir to a crumbling summit: to a sea of nettles: to an empire of red rust: to rituals’ footprints ankle-deep in stone.
(6) Withdrawn and ruinous it broods in umbra: the immemorial masonry: the towers, the tracts. Is all corroding? No. Through an avenue of spires a zephyr floats; a bird whistles; a freshet bears away from a choked river. Deep in a fist of stone a doll’s hand wriggles, warm rebellious on the frozen palm. A shadow shifts its length. A spider stirs…
(7) And darkness winds between the characters.
Gormenghast (2): Figure showing parallel constructions in extract (1)
Titus is seven.
His confines, Gormenghast.
Suckled on shadows;
weaned, as it were, on webs of ritual:
for his ears, echoes,
for his eyes, a labyrinth of stone:
and yet within his body something other –
other than this umbrageous legacy.
For first and ever foremost
he is child.
A ritual, more compelling than ever man devised, is fighting anchored darkness.
A ritual of the blood;
of the jumping blood.
These quicks of sentience owe nothing to his forbears,
but to those feckless hosts, a trillion deep,
of the globe’s childhood.
The gift of the bright blood.
Of blood that laughs when the tenets mutter ‘Weep’.
Of blood that mourns when the sere laws croak ‘Rejoice!’
O little revolution in great shades!
Titus the seventy-seventh.
Heir to a crumbling summit:
to a sea of nettles:
to an empire of red rust:
to rituals’ footprints ankle-deep in stone.
Withdrawn and ruinous it broods in umbra:
the immemorial masonry:
Is all corroding?
Through an avenue of spires a zephyr floats;
a bird whistles;
a freshet bears away from a choked river.
Deep in a fist of stone a doll’s hand wriggles,
warm rebellious on the frozen palm.
A shadow shifts its length.
A spider stirs…
And darkness winds between the characters.
Gormenghast (3): Paraphrase of extract (1)
Titus is seven years old. He lives in the castle of Gormenghast, which is both his territory and his prison. He has been reared within the castle, nourished and nurtured on its shadows and traditions. He hears echoes and he sees an endless maze of stone. But inside him there exists something alien, something else, something that is not the shadowy castle with its many rituals that he has inherited. Because the most important thing about Titus is that he is a child.
This ‘something else’ growing inside Titus is a ritual, but it is not like any of the rituals that are observed in Gormenghast; it has more appeal, more power than these man-made rituals. Within Titus there rages a battle between the rituals of Gormenghast and the more commonplace rituals of childhood. The rebellion is situated in Titus’ blood, which pulses with life. These flashes of self-knowledge have not been inherited from his ancestors, but they have originated in the multitude of children all over the world.
The rebellion in Titus’ blood is a gift from the world’s children. This blood laughs when doctrine and tradition demand that it weep, and this blood mourns when the ancient laws command it to rejoice. A small show of defiance in the boundless shadow of Gormenghast’s history and tradition!
Titus is the seventy-seventh Earl of Gormenghast. He will inherit a castle whose uppermost towers are gradually disintegrating: whose grounds are overgrown with weeds; he will inherit a domain of rusting metal: a domain in which the rituals must be observed. These rituals are imprinted upon the very stones of the castle.
The castle sits in shadow, decaying and isolated: the masonry ancient beyond memory or record, the towers, the vast regions within its boundaries. Is everything corroding? No. A breeze floats through the many spires of the castle; a bird whistles; a fresh-water stream flows away from a choked river. Inside the stones of the castle Titus’ tiny hand wriggles; it is warm and rebellious against the cold stones of the castle. The day draws on and the shadows lengthen. A spider stirs…
And darkness winds between the inhabitants of the castle.
 M. Peake, Gormenghast (1950) in The Gormenghast Novels (1995), p. 399. Paragraphs are numbered for ease of reference.
 J. Mukařovský, ‘Standard Language and Poetic Language’ (1958), Linguistics and Literary Style (1970), D. C. Freeman (editor), pp. 40-56.
 Ibid., p. 42.
 Bernard Bloch is frequently quoted in relation to this issue, because Bloch asserts that the style of a particular text can be defined as ‘the message carried by the frequency distributions and transitional probabilities of its linguistic features, especially as they differ from those of the same features in the language as a whole’ (quoted in G. Leech and M. Short, Style in Fiction (1981), p. 43). Leech and Short point out that it is indeed possible to make some fairly reliable statements about the properties of language, but that the norm of ‘the language as a whole’ is not so easily identified as Bloch seems to imagine, and that ‘without some clearcut notion…of what is meant by “the language as a whole”, any sampling procedure is bound to involve subjective decisions’ (p. 45).
 In corpus-based analysis, for example.
 R. Posner, ‘The Use and Abuse of Stylistic Statistics’, Archivum Linguisticum (1963), p. 126.
 Maurice Gross makes some interesting observations on the failure of generative grammar to produce a workable grammar of the English language some twenty years after its initial inception; Gross suggests that generative grammar could have provided this description of the language as a whole that is currently lacking. He proposes various reasons for this failure, among them the fact that no room has been made for diachronic discussion, and that linguistic theory has been privileged over accumulation of data. M. Gross, ‘On the Failure of Generative Grammar’, Language (1979).
 J. Mukařovský, ‘Standard Language and Poetic Language’ (1958), Linguistics and Literary Style (1970), D. C. Freeman (editor), pp. 43-44. Mukařovský takes the position that poetic foregrounding cannot create new means of communication; in the case of poetic neologisms, for example, their aesthetic function is endangered if they are created with communication in mind. Mukařovský’s argument here takes in some of the unwanted extremes of monism, but there is in his words the germ of the idea that literary language functions differently.
 The existence of an ‘aesthetic canon’ can be supported by reference to the fact that certain words and phrases are marked in the dictionary as belonging to a specifically literary register. The following notes can be found on page xxxi of the eighth edition of the Concise Oxford Dictionary: ‘9.3.6 literary indicates a word or use that is found chiefly in literature. 9.3.7 poet. ( = poetic) indicates uses confined to poetry or other contexts with romantic connotations.’
 J. Mukařovský, ‘Standard Language and Poetic Language’ (1958), Linguistics and Literary Style (1970), D. C. Freeman (editor), p. 55.
 Alex Rodgers’ excellent and very thorough essay on W. H. Auden’s poem ‘O Where Are You Going?’ constitutes an illuminating discussion on how this very process works: in working out the links between the words reader, rider, fearer, farer, horror and hearer – words that are placed in similar positions syntactically and which are also linked through phonological resemblance – the reader eventually arrives at a much deeper understanding of what is at first glance a strange and forbidding poem. A. Rodgers, ‘ “O Where Are You Going?”: A Suggested Experiment in Classroom Stylistics’, Language and Literature: An Introductory Reader in Stylistics (1982), R. Carter (editor), pp. 123-161.
 G. Leech and M. Short, Style in Fiction (1981), p. 29. See also introduction, p. 16, footnote 60.
 G. Leech and M. Short, Style in Fiction (1981), p. 48. The problem with this definition, of course, is the difficulty in establishing the ‘normal frequency of a feature’.
 This notion is inextricably linked to literary competence, and is different for every reader, depending on previous training, experience, and so on. See also chapter four of this thesis.
 G. Leech and M. Short, Style in Fiction (1981), p. 50.
 Ibid. It will have been noticed that Leech and Short include the reader in their discussion of foregrounding, which Mukařovský does not.
 Ibid., p. 139.
 M. Peake, Gormenghast (1950) in The Gormenghast Novels (1995), p. 399. Subsequent references to this work are taken from the same page unless otherwise indicated.
 The characters in bold are an instance of pararhyme (CVC), another of Peake’s sound patterns. (CVC: where C = consonant and V = vowel; the characters in bold are those on which the rhyme is founded.) However, the phrase in question here is very common in English. M. Short warns the student of literature against the dangers of ‘over-milking’ the significance of phonetic patterns. He notes on p. 116 of Exploring the Language of Poems, Plays and Prose (1996) that ‘because English only has approximately 45 phonemes, there are bound to be a fair number of accidental alliterative and assonantal patterns in any text as a consequence of the chance distribution of sounds in groups of words. So we should not assume that all sound patterns will be significant in terms of interpretation.’
 Gunther Kress distinguishes between transactive and nontransactive structures as follows: ‘events either appear in a transactive form…that is, portrayed as either arising directly as the result of some agent’s action and with a direct effect on a goal (where both agent and goal may be either animate or inanimate), or in a nontransactive form, arising without such action, that is, as either a self-caused action or an action that happens in some unspecified way.’ Clearly, the reader is presented here with the nontransactive form, the portrayal of events in the passive voice. G. Kress, ‘Ideological Structures in Discourse’, Vol. 4: Handbook of Discourse Analysis, Discourse Analysis in Society (1985), p. 34.
 Peake’s use of the present tense for the opening paragraphs of his novel lends the text a sense of immediacy; however, in the wider context of the novel, the use of the present tense also suggests and reinforces the idea of time standing still. Manlove notes that frequently within the structure of the novel’s storyline ‘time seems to be going both backwards and forwards, and the net effect is that the temporal sequence appears frozen’. C. N. Manlove, Modern Fantasy: Five Studies (1975), p. 226. Past, present and future are all one and the same. Change is the enemy of Gormenghast; changes take place over time; time does not move within the castle. As noted elsewhere in this thesis, tense should not always be taken as an indicator of chronological time. Ronald Carter has written that ‘recent studies have contended that where a particular tense pattern dominates a text, what is communicated as a result is not so much a notion of time or chronology as a special modality’. R. Carter, ‘Responses to Language in Poetry’, Literary Text and Language Study (1982), R. Carter and D. Burton (editors), p. 32. In the case of Peake’s novel, the present tense indicates not exactly the here and now, but the sense that what is described represents the general state of affairs, past, present and future.
 Shorter Oxford English Dictionary, fifth edition, 2002.
 C. N. Manlove, author of Modern Fantasy: Five Studies (1975), would most likely agree with this reading, his argument being that Peake became so involved with the world he created in Gormenghast, that he failed to provide Titus with a good enough reason for leaving it: ‘so much does the castle of the Groans dominate [Peake’s] imagination, that where other writers strive to get into their fantastic worlds, Peake struggles to get out’ (p. 217)…‘Throughout Gormenghast he shows his uneasiness in a continual search for some way of accounting for Titus’ quest. He offers a confused variety of motives which still beg the question as to where they come from in the first place’ (p. 238).
 Manlove notes that Peake generally avoids the use of verbs – specifically action verbs – in connection with the castle itself: ‘the power…comes from Peake’s use of verbs: giving none to the castle, he frees its life from any link with the time-bound’, Modern Fantasy: Five Studies (1975), p. 219. While this is an interesting point, I think it has to be noted that Manlove does not provide any quantitative data to support his assertion; without such data – which a stylistician would not, I think, have neglected to provide – his claim, although thematically plausible, is simply a hunch.
 M. Peake, Gormenghast (1950) in The Gormenghast Novels (1995), p. 807.
 Nettles are also plants on overgrown land.
 Leech notes that these categories ‘overlap, because humanity entails animacy’; the animistic metaphor ‘attributes animate characteristics to the inanimate’ and the humanizing or anthropomorphic metaphor ‘attributes characteristics of humanity to what is not human’. ‘Withdrawn’ and ‘broods’ can at a stretch fit both categories, although ‘ruinous’ has a more dubious status. G. Leech, A Linguistic Guide to English Poetry (1969), p. 158.
 C. N. Manlove, Modern Fantasy: Five Studies (1975), p. 217.
 Leech and Short provide an interesting analysis of a section of Gormenghast which immediately follows the one analysed here. This section opens with the words ‘Who are the characters?’, and Leech and Short note that ‘characters can mean either “letters, ciphers”, or “people in the fiction” ’. G. Leech and M. Short, Style in Fiction (1981), p. 140.
 E. Brontë, Wuthering Heights (1847), Penguin Classics edition (1995), pp. 19-32. Subsequent references to this edition are marked in the main body of the text.
 Critics have, of course, discussed Cathy’s negative relationship with books. She makes herself ill on purpose in order to separate Edgar from his library, and from the passage discussed here the reader can see that even from a very early age she was ill-disposed towards literature. Instead of reading the books in her library, she defaces them with scribbled complaints about Hindley’s ill-treatment of Heathcliff and with rude caricatures of Joseph. She hurls a volume into the dog-kennel, vowing that she hates ‘a good book’ (p. 21). Heathcliff follows suit, and his punishment is to be prohibited from eating meals with the family and from playing with Cathy. Heathcliff is later barred from pursuing his education any further, and books symbolise to a certain extent the social world that is represented by Thrushcross Grange and its inhabitants. One can add to these observations the fact that Lockwood piles ‘the books up in a pyramid’ against the hole in the window, in order to keep Cathy’s child-ghost out (p. 25). It is significant also that Cathy’s daughter wins Hareton over by encouraging him to read, thereby achieving what her mother could not – she accepts and sets out to improve the ‘degraded’ Hareton, where Cathy rejects the ‘degraded’ Heathcliff.
 M.A.K. Halliday and R. Hasan, Cohesion in English (1976), p. 15.
 Heathcliff also reprimands Lockwood on this score: ‘your childish outcry has sent sleep to the devil for me’ (p. 28).
 There has been much discussion regarding the morality of Brontë’s novel, and Q. D. Leavis was of the opinion that the book ‘has a very firm moral effect. The technical means…for implying moral criticism without stating it, for making the reader do this work himself, is the technique of contrast and parallelism’. Q. D. Leavis, ‘A Fresh Approach to Wuthering Heights’, Collected Essays Volume 1 (1969), p. 252. Leavis’ argument is that the apparently cruel behaviour of one character is often offset by the brutal behaviour of another, and this parallelism constitutes part of Lockwood’s role: ‘Lockwood’s horror of the household at Wuthering Heights is immediately offset by our horror at him when he then, in a real nightmare, brutally fights off the child begging…to be let in after losing the way on the moor’ (ibid., p. 254).
 By ‘parenthetical constructions’, I mean those clauses or sentences which are not part of Lockwood’s dream narrative, and which represent an intrusion of Lockwood’s voice into the dream-sequences.
 J. Hagan, ‘Control of Sympathy in “Wuthering Heights” ’, Nineteenth-Century Fiction (1967), p. 316.
 B. McHale, ‘Unspeakable Sentences, Unnatural Acts: Linguistics and Poetics Revisited’, Poetics Today (1983), pp. 17-45