SOURCE: Ofuani, Ogo A. “Lexical Cohesion in Okot p'Bitek's A Song of Prisoner.” In The Language of African Literature, edited by Edmund L. Epstein, pp. 205-28. Trenton, N.J.: Africa World Press, Inc., 1998.

[In the following essay, Ofuani explores elements that add to the thematic unity of Song of Prisoner]

Despite numerous studies by critics and literary scholars, the effectiveness and aesthetic use of language in Okot p'Bitek's Song of Prisoner (1971) have remained largely unexplored. Scholars have made contributions to our understanding of p'Bitek's themes, images, historical and biographical background, and traditional poetic devices, but have devoted little time to linguistic/stylistic analyzes of his language (see Ofuani 1985). The issue that tends to dominate all others has been the controversial one of the number of prisoners contained in this Song (see Ogunyemi 1982: Wanambisi 1984). The linguistic clues that should resolve this issue are often neglected, and critical analyzes have suffered from a deplorable vagueness and lack of depth (Ofuani 1988).

By looking at the lexical cohesive properties of a specific text, instead of all of p'Bitek's Songs, this essay will help establish some of the special properties of Song of Prisoner.¹ It ought to be possible to state some of the grounds upon which readers or critics of a literary work base their estimation of the work's stylistically significant features. On a more general level, this essay hopes to supplement the larger corpus of literary critical analyzes in which linguistic explanations of literary impressions, intuitions, and judgments about this text are either completely absent or glossed over in generalizations.²

I

The concept of cohesion is semantic: “it refers to relations of meaning that exist within the text, and that define it as a text” (Halliday and Hasan: 4). Cohesion occurs where the interpretation of one element in the discourse is dependent on that of another. The one presupposes the other in the sense that it cannot be effectively decoded except by recourse to it. When this happens, a relation of cohesion is set up, and the two elements—the presupposing and the presupposed—are thereby at least potentially integrated into a text. In another related definition, cohesion is seen as

the relations obtaining among the sentences and clauses of a text. These relations, which occur on the grammatic stratum, are signalled by certain grammatical and lexical features reflecting discourse structure on a higher semiologic stratum. These features, such as anaphora, subordination and coordination … account for what may be referred to as the textual connectivity of sentences and clauses.

(Gutwinski: 26)

Cohesion is part of the system of a language. The potential for it lies in the systematic resources of reference, ellipsis, and so on that are built into the language itself. In any given instance, however, cohesion depends not merely on the selection of some option from these resources, but also on the presence of some other element that resolves the presuppositions set up.

Like other semantic relations, cohesion is expressed through the strata organization of language. Language can be explained as multiple coding systems, or strata: the semantic (meanings), the lexicogrammatical (forms), and the phonological and orthographic (expressions) (Leech 1969:37). Meanings are realized (coded) as forms, and forms are in turn realized (recoded) as expressions. There is no hard-and-fast division between vocabulary and grammar; the guiding principle is that more general meanings are expressed through the grammar and more specific meanings through the vocabulary. Cohesive relations fit into the same overall pattern. Cohesion is expressed partly through the grammar and partly through the vocabulary.

We can refer, therefore, to grammatical cohesion and lexical cohesion. The distinction is really one of degree, and we need not make much of it here. There is no sharp line between grammar and vocabulary: the vocabulary, or lexis, is simply the open-ended and most delicate aspect of the grammar of a language.

Lexical cohesion is the cohesive effect achieved by the selection of vocabulary. It is not a single homolithic entity or phenomenon but involves a complex of interrelationships in the choice and use of words within a text, the cohesive effect of this choice and use deriving from the various semantic effects they yield (Halliday and Hasan 1976:74).

Lexical cohesion could be achieved through reiteration, lexical relations, and collocation. Reiteration involves the repetition of a lexical item, at one end of the scale; the use of a general word to refer back to a lexical item, at the other end of the scale; and a number of things in between: the use of synonym, near-synonym, a superordinate, or a general word; and in most cases it is accompanied by a reference item, typically the. This kind of reiteration involves reasonable reference to the same referent. But a lexical item may cohere with a preceding occurrence of the same item whether or not the two have the same referent, or indeed whether or not there is any referential relationship between them. The possibilities may be, as far as reference is concerned, identical, inclusive, exclusive, or simply unrelated. At another level, there is always the possibility of cohesion between any pair of lexical items that are in some ways associated with each other in the language. The cohesive effect of such pairs depends not so much on any systematic relationship as on their tendency to share the same lexical environment, to occur in collocation with one another. In general, any two lexical items having similar patterns of collocation—that is, tending to appear in similar context—will generate a cohesive force if they occur in adjacent sentences. However, it should be borne in mind that collocation is simply a cover term for the cohesion that results from the co-occurrence of lexical items that are in some way or other associated with one another, because they tend to occur in similar environments: the specific kinds of co-occurrence relations are variable and complex, and would have to be interpreted in the light of a general semantic description of the English language (see the essays by Halliday and Sinclair in Bazell 1966).

II

Song of Prisoner is a poem about African society after Independence (lo Liyong 1971:59-61). It explores further the theme of post- colonial disenchantment raised in the earlier Song of Lawino (1966) and Song of Ocol (1970). The prisoner's anger is the anger of the class of East Africans he represents: the poor who contributed immensely to the Independence efforts but are now relegated to pre-Independence subjugation and humiliation. They have lost their pride and dignity through the wiles and violence of the new black African masters.

It is “a generalized satire” (Roscoe 1977:54), almost allegorical: Prisoner has no name, the politician he murdered is unnamed too. Prisoner has no identity; he is one of a type. This makes him pathetic and symbolic. He is married with children and comes from a minority tribe. He has played an unspecified role in the Independence struggle. Now, he is a spokesman for the disillusioned majority: the rural poor, neglected minority groups, the unemployed, the underdogs. He is arrested and imprisoned for that and now talks to us from his cell (which he describes as “hell”). He does not expect to be and is not fairly treated by the judge and his jailers, whom he addresses as “Brothers.” At times he talks like a man from “the wrong clan,” a man outside the system he criticizes, but at others he talks like a man who once belonged. He accuses and condemns the politicians and their thugs, the prison guards, and his own weaknesses.

But his motives are suspect. He has plenty of reasons to be bitter: he has fought for “Uhuru,” and those now in power have betrayed what he fought for. He is still poor, his children are still sick, hungry, and uneducated; his matrimonial bed is desecrated by the “Big Chief” (44); and he sees nothing but violence, corruption, and tribalism. His anger seems justified. But he is also motivated by hatred and despair, and we do not find it easy to separate his personal motives from his social vision. His description of the man he killed (66-74) lacks social significance; we do not immediately see whether he killed him to avenge his own disappointment. His descriptive tone turns into a sadistic venture: as if he killed for killing's sake.

There is a general tone of complaint in the poem as Prisoner catalogs, in each section, the ways he and his class have been maltreated.³ In the enraged state of mind in which he finds himself after being beaten and tortured, he digresses now and again and repeats what he has said before in his effort to elaborate, substantiate, and prove his allegations or justify his actions. He looks for analogies to his physical situation. In spite of his digressions and repetitions, and in spite of the length of this monologue, the poem is a coherent unit. Cohesion, particularly lexical cohesion, helps to make the poem what it is.

III

Song of Prisoner, in the tradition of p'Bitek's other poems, is in fifteen movements or sections.⁴ Consequently, an analysis of the type intended here would be cumbersome if all its movements were subjected to analysis. The first (under the subtitle “dung of chicken”) is representative enough of the other movements and so provides the corpus for the present analysis. Where necessary, incursions will be made into other movements to depict the relationship between all of them. This restriction of data is, however, not a defect if a detailed microstylistic analysis is to be achieved.

The title of the poem contributes significantly to its general cohesive pattern. Song of Prisoner is a nominal phrase of the Head (“song”) and Qualifier (“of prisoner”) type,⁵ in which the article, whether definite or indefinite, has been deliberately omitted. The Head and the Qualifier are both capable of taking the article, the addition of which should have changed the form and, therefore, the significance of the title. The omission is paradigmatic. There are four possibilities: “the song of a prisoner,” “the song of the prisoner,” “a song of a prisoner,” and “a song of the prisoner.” If the qualifying element of the title is crucial, then two possibilities are likely: “song of a prisoner” and “song of the prisoner.” If the meanings of these two latter versions are considered, the omission of the article will be seen to be stylistically significant.

If the indefinite article a is taken as given, the prisoner will have an almost allegorical quality, referring to any one prisoner who finds himself immersed in the kinds of experiences narrated in the poem. Phonological cohesion can also help here to suggest only one prisoner known to the poem (“Song of a Prisoner”).⁶ The choice of the definite article the, on the other hand, suggests a specific prisoner. The will make the title lose its allegorical tone by exophorically referring to a prisoner already known to the poet but not known or assumed not to have been known to the reader.

Okot p'Bitek's deliberate omission of the article is a grammatically cohesive device that operates at two levels: the level of exophora, where it is assumed the reader has foreknowledge of which prisoner, and at the level of ellipsis, involving the omission of the deictic element of the noun phrase. It is also lexical because it involves the omission of a lexical item of the closed set whose absence consequently ambiguates the meaning of the title. This omission could, therefore, have been partly responsible for the different interpretations given to the persona(e) in this poem. Some critics believe there is only one prisoner who on some occasions involves himself in the confusion and self-deception that create the impression that there are two prisoners (especially in the seventh movement, “voice of a dove,” which seems out of tune with the others). Other critics feel there are two prisoners (a poor one whose voice re-echoes throughout the poem, the misused political assassin and hero of Uhuru who is angry and pleads to all kinds of repression, the defenseless and badly beaten prisoner of all the movements, except the seventh; and the rich, satisfied, unperturbed prisoner of the seventh, whose detention is only momentary). (See Gathungu 1973; Heron 1976; Marshment 1972; Ofuani 1985; and Ojuka 1973.) Edward Blishen (1971: 6-35) even suggests that there are three voices and personalities, though he fails to establish the existence of three separate masks. It is worth mentioning here that in Two Songs, p'Bitek reveals a significant consistency in the non-use of the article in headings. In Song of Prisoner the word “Prisoner,” as used, makes concrete reference, and the feeling of cohesion is achieved through the play with the semantic features of nouns. “Prisoner” is a common noun but the omission of the article and its capitalization give it the status of a proper noun, make it a name, thus giving it specificity, and identity. Prisoner is now a person, not a class. The same goes for the persona in Song of Malaya.

The kind of cohesive pattern traced in the poem's title is also present in the subtitle of the first movement: “dung of chicken.” It is a noun phrase, made up of a Head (“dung”) and Qualifier (“of chicken”). The interpretive significance is that, as a heading, it carries the main statement about the state-of-affairs described in the section: corruption, brutality, mercilessness, and violence. All these associations are signalled by the connotations of foulness, fetid odor, putrefaction, repulsion and so on that “dung of chicken” brings to the reader's mind. We are, therefore, psychologically prepared for the brutality the prisoner says he is subjected to.

The first movement has, for convenience, been divided into eighteen stanzas determined by the poet's use of the conventional breaks between strand so on.hes of verse. This movement starts with a description of the prisoner's discomfort:

1) The stone floor
Lifts her powerful arms
In cold embrace
To welcome me
As I sit on her navel.

(11)

The discomfort described here is conveyed in the rest of the movement through a combination of the various facets of lexical cohesion. There is, for instance, a concentration of phrases with different kinds of collocation: words with negative meanings whose effects contrast and tend to negate other positive ones with which they co-occur (“cold embrace … welcome”); personifications involving suggestions of animacy for inanimate objects (“stone … lifts … powerful arms”; “stone floor … her navel”), and so on. The stone floor's “cold embrace” is negative and immediately contrasts with “welcome me” (positive). The “cold embrace,” if explained as a sexual image, has all the connotations of detached and unresponsive relations and so negates the effect of warmth suggested by “welcome.” Human associations are significantly used to describe the effect of the bare floor (“her navel”) on the prisoner's bare body. The several attributes in 1), whether anatomical or merely attributive, are unnaturally ascribed to the different portions of the prisoner's cell, giving them volition, a degree of deliberateness, a sense of purpose that is almost human in their effects on the prisoner. It is as if Prisoner believes the cell has acquired the physical features of the human beings whose invention it is. Most of the images are erotic and female, suggesting the extra softness and tenderness expected of a woman that starkly contrast with the hardness and coldness of the cell.

Other human associations given to inanimate objects include their ability to exhibit emotions. For instance, in the third stanza the prisoner describes “the brow / Of the weeping stone floor” (12). The wetness (and cold comfort) of the stone floor that increases the prisoner's discomfort is likened to a human face and given the capability to express human emotions (to weep). Most of these metaphors arise from a deviation in lexical collocation at the syntagmatic level, and the negative interpretations we give to their associations derive from this deviational pattern. For example, stone walls have no arms they can lift, irrespective of the effect of these arms, whether they are “cold” or really intending to “welcome” the prisoner, or to increase his discomfort. “Arms” should collocate with “man.” This kind of deviant collocation heightens, in its absurdity and unexpectedness, the unnatural and unjust atmosphere of the prisoner's situation. The inappropriate collocations bolster this distortion of the natural order of pity and mercy. In addition, cohesion is aided through the provision of one consistent imagery. There is a progression in noun reference from inanimate to animate; from animacy to personification; and from human to proper.

The sense of discomfort aroused by the contrasts created by the collocational patterning of the personifications and metaphors is also carried by the similes. In the ninth stanza, the prisoner describes the profuse bleeding of his nose as follows:

2) My nose
Is a broken dam
Youthful blood leaps
Like a cheetah
After a duiker …

(14)

Passage 2) is significant at several interrelated linguistic levels. First, is the prisoner's use of the equative copula verb be (line 2) such that we have the structure x = y, “My nose” (x) “is” (=) “a broken dam” (y). But x is not exactly y. “My nose” would normally make us expect a trickle since nostrils are small; but “a broken dam” carries the suggestion of a deluge. So there is exaggeration (hyperbole) in the discrepancy between “nose” and “dam” (especially as the declarative nature of lines 1 and 2 makes them look like a statement of fact). The prisoner also exaggerates in the simile following “broken dam” (“Youthful blood leaps / Like a cheetah”). “Blood,” because of its liquid nature, does not “leap” but Prisoner states so as if it were true. “Leaps” and “cheetah” seem to form a lexical set, stressing intense agility. The cheetah is known for its speed and lethal accuracy in tracking down its prey. Its agility ties with “youthful” and “leaps.” These associations are blood-curdling when used to describe the flow of the prisoner's blood. For his situation, the comparison is apt, more so as it describes one aspect of the outward manifestation (the result) of the cruelty and manhandling he has gone through.

A similar simile is in the tenth stanza:

3) Brother
How could I
So poor
Cold
Limping
Hungry like an empty tomb …
Inspire you
To such heights
Of brutality?

(15)

Line 6 above is significant. The attributive/stative adjective “hungry” is animate, and if used for a “tomb” suggests personification as the question that arises is “Could the tomb be hungry?” But in the context of Prisoner's starvation, the comparison describes the insatiability of graves (their unending reception of the dead), likened to his hunger. “Tomb,” therefore, ties anaphorically with the description of his imminent death through loss of blood in 2).

Lexical cohesion is also achieved by a concentration of items whose semantic description involves the contrast between want and avoidance (approach-avoidance), as evidenced in the following pairs:

cold embrace-welcome

(11)

colorless-rainbow

(12)

Several such pairs occur in other sections. The significance of the first pair is explained here. The approach-avoidance syndrome can be related to the sexual images explained above (in relation to passage 1), and describe the kind of situation where the erotic man-woman effect that pervades the movement is equivalent to the effect of the cold on the prisoner. His initial acceptance by the stone wall (“embrace,” “welcome,” “kisses”) contrasts with the numbness and pain the wall offers him after the seduction (“shoots freezing bullets / Through my bones”).

The real effect of the contrast is that there is a significant play on words. Irony is expected here because the stone wall as an inanimate object has no volition and is incapable of initiating the kind of actions ascribed to it. The ironic intention is to show the similarity between Prisoner's suffering (coming directly from the cold wall and floor) and that meted out to him by his torturers (the guards). The referent involved (the stone wall) is static, fixed, and passive.⁷ Its effects on the prisoner can therefore be blamed on the jailers. But the “rejection,” symbolic of the rejection and lack of protection from those who paid him to assassinate others and who have not kept him out of jail, makes him rigid with anger, his throat burns, his heart gets riddled “with the arrows / Of despair”; he is “drowning / In the deep lake of hatred.” Bruised and bleeding, he marvels at the extent of the jailers' brutality against a poor defenseless man in spite of his services to Uhuru:

4) A young tree
Burnt out
By the fierce wild fire
Of Uhuru.

(15)

His despair is heightened by the concatenation of words that have associations of lifelessness: stone, cold, whitewashed, shoots, freezing, and so on. Shoots in the third stanza, for instance, refers forward through cataphora to elements of pain and destruction in the fourth: fiery hailstones, punching holes, and so on., all of which reveal harshness and brutality.

Cohesion is achieved also in the first movement with words suggesting motionlessness, and there is progression in their occurrence from the first stanza: sit (1), rests (2), dizzy (6), bed (12), sleep (13) and asleep (17). These words arouse in our imagination the difference between pain and death. The contrast is sharper because psychological torture is increased by a combination of physical pain, the glare of the harsh lights in the prison cell, the restricted movement (“tie my hands / And feet / With this rope”), and the numerous images through which the prisoner describes their effect (bullets, hailstones, white, hot, melting, drowning, break, killed, cracked, broken, fire, slap, ram, red-whirlwind, venom, mamba). Sleep is impossible, but that image (sleep) and then death keep coming to Prisoner's tortured mind, as if death should logically follow sleep in this situation. Hence the sequence of “my father / Is asleep” (penultimate stanza) and “my father's / Grave” (last stanza). In his present state, he is a contrast to his father, who is in a kind of restful sleep in his grave,

5) Unseeing
Unhearing
Undreaming.

(17)

There is a cohesive effect in the repetitive pattern of the lines in 5), in which the senses of sight, hearing, and cognition (perception) are concentrated. The cohesion is syntactical (morphological) and also lexical. There is a kind of Cummingsian touch in the role of morphology here, in the derivation of stative nouns through a combination of the use of the negative prefix un- and the gerunding suffix -ing. The un- prefix is reversitive, implying that the states of seeing, hearing, and dreaming have been reversed. These -ing forms are stative nouns rather than progressive verbs suggesting ongoing processes. There is a degree of permanence in the new states created with the aid of the prefix, thus giving the prisoner's lament a tone of finality, futility, and irredeemable sorrow. His father's states contrast with his own.

Most significant of the words in 5) is undreaming which, in this context, has several possible meanings. Literally, it means his father is “not dreaming” (he is not conscious, completely dead, unfeeling). At another figurative level, it means his father no longer has delusions—he now knows more about reality, hence his detached objectivity, unlike the prisoner who still has delusions about release, rewards, Uhuru and its promises, revenge, and so on. At a third level, the un- prefix could be interpreted to mean that in his new state of detachment achieved at death, his father is now “changing his old dreams, re-living his past,” taking stock and seeing his past from a new perspective (perhaps the proper one!). He has achieved a moment of proper focus that is not and cannot anymore be distorted by pain, or the senses.

The last stanza is a single sentence that begins with a line of command:

6) Listen to the footsteps
Of the wizard
Dancing on my father's
Grave.

(15)

The verb “listen” draws attention to the wizard's desecration of the prisoner's father's grave, overturning those sacred things he holds dear. This verb has no immediate subject and so applies to both the singular and plural you (the tormentors as well as the reader). The prisoner is, in prison, more perceptive than his tormentors to imaginatively “hear” the desecration of his father's grave. His views are no longer distorted by dreams of Uhuru that have been “burnt out.” He no longer has illusions. If the analogy is drawn between the “unseeing,” “unhearing,” “undreaming” state of his dead father and his own apparent “life-in-death” state of pain and torture, he may be implying some kind of cathartic effect of the pain for himself, responsible for his sensitive nature. He has experienced a purgation of his earlier trust for his “brothers” in the fight for Uhuru, for whom he no longer has any regard. This could explain the sarcasm with which he attacks their misdeeds, in his own admission of being guilty of the murder of the politician but confessing to other “crimes,” such as impotence within the new order, deafness, blindness, fear, joblessness, landlessness, sickness, orphandom, and hatred for those who exploit the poor—the deprivations caused by the new African post-Uhuru leaders. The last stanza ends on a note of pessimism struck by the grammatical form of the verb that, at the paradigmatic level, is a lexical item.

Whatever the meaning read into the end of this movement, it tallies with the note of negativisim with which it started. The first movement as a whole sets the general tone for the other movements, which are composed along the same lines, cataloging the ills the prisoner laments about. Lexical cohesion is not internal to this movement but is tied with the other movements. Most of the patterns discussed in the first recur in the others.

IV

Examining a text, especially one as long and intense as Song of Prisoner, can often reveal sources of cohesion we might not otherwise notice and can help us discover recurrent themes and images. In any discourse, lexical choice is intimately associated with—and partly limited by—choice of subject matter. Lexical choices that result most directly from choice of subject matter are likely to be the least striking stylistically. This means that the choices, to the extent that they are likely in the context (i. e., if they are normally associated with the subject matter), are not foregrounded, not attention-drawing, and therefore not as stylistically significant as the foregrounded elements.

Song of Prisoner illustrates some of the usual kinds of lexical foregrounding. A number of words, word-groups, and even clauses stand out by virtue of their rarity or their novelty in the context in which they are used. These elements are often striking because of the associations they carry. In the first movement, the following occurrences are some of the most striking:

i) “freezing bullets” (12). The bullets here are not real bullets but bullets of cold; cold (“freezing”) contrasts with bullets, which are usually propelled by the heat of gunpowder combustion; here bullets refer to the piercing effect of the cold on the prisoner's bare body.

ii) “fiery hailstones” (12). Hailstones are drops of iced rain; they should be cold but the ones dropping from the roof of the prisoner's cell are unusual—they are “fiery,” not in their coldness but in their glare and persistence (they refer to the harsh electric light on the roof). Hailstones can never be fiery, hence their unusual collocation. The continuous drop of these masses of iced rain can literally cause discomfort and pain.

iii) “colorless rainbow” (12). Another unusual collocation that is reminiscent of Chomsky's (1965) “colorless green ideas,” describes the blood-spattered walls of his prison cell. Could it be a reference to the monochrome effect of his blood on the walls and floor?

The sense of ii) and iii) could be seen if placed in the full context of the utterance in which Prisoner exhibits his usual predilection for exaggerations:

7) That giant firefly
On the high ceiling
Rains fiery hailstones
Into my closed eyes
Punching holes
Through the thatch,
There is a colorless rainbow
On the bleak white walls …
And on the brow
Of the weeping stone floor …

(12)

A poet's use of deviant expression and unusual collocations is usually called “poetic license.” As Geoffrey Leech explains (1969: 37), in the phrase “poetic license” we concede the poet's right to ignore rules and conventions generally observed by users of the language. But in the grammar of a language, there are usually constraints imposed on the use of words in certain environments. Such “selectional restrictions”⁸ govern the selection of lexical items for insertion into underlying structures (Chomsky 1965). The three examples above are instances of apparently meaningless collocations involving contradictions because they say something is both x and not-x at the same time (Traugott & Pratt 1980:206). In iii), for instance, the contradiction lies in the fact that rainbow is colorful but is here being claimed to be colorless.

Examples iv) and xi) below are in a different class as metaphorical forms, with their areas of semantic transference divergent. Metaphors may be anomalous but they do not fail to communicate-they are to some extent “meaningful.” When our knowledge of language and the world will not let us take a meaning literally, we do not give up, but rather “make sense” of the anomalies by allowing certain features to override others in particular contexts. These examples further illustrate Prisoner's hyperbolic disposition; their contexts are provided to make their senses clearer:

iv) I am drowning / In the deep lake / Of hatred …

(13)

v) My heart is riddled / With the arrows / Of despair …

(13)

vi) I am engulfed / By a red whirlwind / Of pains …

(17)

vii) Do you see / The beads of blood / On my legs and feet?

(14)

viii) Look at the laughing wound / In my head …

(14)

ix) The dark silence / Urinates fire / Into my wounds …

(16)

x) My brain is melting …

(13)

All the examples can be given different explanations, and have been italicized to show the areas of transference. Texts iv) to vi) are concretive metaphors that attribute physical existence to abstractions. The abstract “hatred,” “despair,” and “pains” that describe sensory perceptions or feelings are equated to the physical forms “lake,” “arrows,” and “whirlwind,” respectively. The semantic transference lies in their equations (lake = hatred; arrows = despair; whirlwind = pains). Example vii) involves the equation of two physical entities that are in different states (“beads,” solid; “blood,” liquid). Beads are ornamental but these are not jewelry, only the clotted blood on his lips. The brutality he received is nauseating, yet he describes it with an ironic matter-of-fact tone that makes the result more telling (as if he says, “see how they decorate me with my blood”).

The situation in viii) is equally serious. His “wounds” are made to “laugh.” Maybe in personifying his “wounds,” he really wants them to laugh at human callousness. He sees a parallel between the shape of a laughing mouth and the depth of his wound. Text ix) is deviant at two levels. The abstract “silence” is ascribed an animate characteristic to “urinate,” an impossibility made more tenuous by the nature of the product of this excremental process: “fire.” The verb “urinates” collocates with and predicts/anticipates water. “Fire” is therefore aberrant here, even if the silence literally can urinate. It is difficult to perceive the grounds of Prisoner's usage here.

Lastly, text x) adds a climactic note to Prisoner's frustration: his physical “brain” is said to thaw, to melt, to become liquid (like ice). The reduction he describes refers to the unbearable pain, mental torture, and pressure he is subjected to, particularly when later he screams, “I am / Mad, / Can't you see?” (24). The ground for his complaint is not farfand so on.hed.

These unusual forms also occur in the second movement, whose preoccupation is with hunger and its effects: “infant pregnancies” (in reference to the prisoner's children's malnutrition and kwashiorkor—22); “their stomachs / Drum sleep off” (reference to the disturbing effect of hunger and starvation—23); “javelins / Inside my stomach” (descriptive of the burning effects of hunger in his stomach—23).

One cannot fail to notice the use of shock words, especially the references to sexual organs, physiological processes, and excrement. This kind of “obscene” language is not a mere attempt at reproducing street language; it serves a fundamental use in the expression of anger and hate. It is often directed toward some person or thing the narrator loathes (see Luce 1977; Collins 1979). Examples abound in the first movement: penis, breasts, bosom, belly-button, buttocks (12); stark naked, in bed with your wife, raping your mother (15). These words and expressions are used in the questions asked by the prisoner in which he suggests (rather naively?) that only such sexual misdeeds should occasion the kind of punishment meted to him. He uses them to advance his argument about his innocence and not because he intended to be vulgar and obscene. His only “outrageous” crime, he suggests, is his loyalty to the Party and commitment to Uhuru.

There are also other references to physiological processes: urinates (16), pangs / Of childbirth (17). These sexual images progress from a mere anatomical description of the effect of the cold on his body couched in the images of sexual foreplay, to those of intercourse (whether cooperatively accepted by the jailer's wife or forcibly imposed on his mother) and then on the finality of childbirth: embrace: kisses: naked: bed: raping: sleep: urinates: childbirth. These words form a lexical set whose meanings are interrelated. In this context, urinates could be a euphemism for the more offensive ejaculation. “Childbirth” in the sixteenth stanza is immediately followed by “my children” in the penultimate stanza, as if the mere mention of “pangs of childbirth” reminds him of the prolonged suffering of his wife and children (who form the subject of the second movement). After the pains of childbirth should come the joys of motherhood but this natural order is perverted as the “lullaby is stuck / In their mother's throat” (17), and he remembers the death of his own father. His children, like himself, will soon be orphans at his own death. This refers forward to the helplessness of his wife and children in the second movement: “The cry of my children / And the sobs / Of my wife” (24).

Homonymy gives rise to another kind of lexical foregrounding and cohesion, the pun. Puns are often described as involving double or multiple meanings, an intentional ambiguity. What puns usually do is introduce more than one lexical item by means of a single phonological sequence. The multiplicity of possible meanings in the context can create layers of unexpected cohesion. One way writers make puns is by manipulating spelling to suggest multiple lexical items by a single word (homography and homophony). One striking instance of such usage is the reference to arms in “the stone floor / Lifts her powerful arms / In cold embrace” (11). The immediate literal meaning of arms here is that used in reference to limbs, though it has already been explained that this is an instance of personification. This meaning of arms collocates with its use in “see the muscles / Of my arms” (13). But arms can refer to weapons and is suggestive of those “powerful guns and rifles” used by the jailers, especially if related to “why do they ram my feet / With the butt / Of their rifles” and “uniformed Brothers” (16). In both instances, the single word arms is homophonic (/a:mz/).

Homophony can also give rise to puns in instances in which two words may not be homographic but are pronounced alike. In the first movement the prisoner's description of his loss of blood contains the possibility of punning the word cheetah that reminds one of cheater: “Youthful blood leaps / Like a cheetah.” Cheetah, a feline, pronounced /t∫i:t[UNK]/, brings to mind the homophonic equivalent cheater (/t∫i:t[UNK]/), which means “one who cheats: an officer who collected the fines to be paid into the Exchequer” (in Shakespearean times; Chambers 1977: 222). The homophony here could be seen in this second meaning to relate later with the words caught, find in 8), where the central issue is “Was I caught cheating?”:

8) I was not caught
Dancing stark naked
Around your house.
Did you find me
In bed with your wife
Or raping your mother?

(15)

The effect of this second pun is not as direct as the first but is equally a possibility in the context of the poem in which crime, discovery, and punishment are re-echoing themes.

Lexical foregrounding can also be achieved with the aid of reiteration involving the repetition of one lexical item, as is obvious in the stanza below:

9) My head rests
On her flat
Whitewashed breasts.
She kisses
My bosom
My neck
My belly button
My back
My buttocks
And shoots freezing bullets
Through my bones.

(11-12)

The first-person singular possessive pronoun my is used seven times. This repetition produces a cataloging effect. The deliberate slowness of the rhythm is metrically evocative of the slow and methodical spread of the cold as it virtually “undresses” the prisoner in their “love tangle.” The use of enjambment here may make the reader want to produce the lines at a faster rate to reveal the urgency of the “spread.” But even then, the end-pauses unmarked by punctuation and the repeated use of my at the beginning of each of lines 5-9 slow down the speed. This repetition of my may be seen as the prisoner's way of emphasizing the effect of the cold floor—emphasizing the degree of personal knowledge of the suffering involved in this “they-affect-me-so-I-should-know-if-you-doubt-me” stance. This personalization of the experience through the use of my ties cohesively with the use of other variants of the first-person singular pronoun (“I,” “me,” “mine”) in all the movements of the poem. Texts 1-3) and 6-9) support this. This usage pervades the song. For instance, in section 2 he pleads insanity (“I plead insanity, / I am / Mad, / Can't you see?” 24) and later contradicts himself (“I am not insane,” section 8, 67). In section 4 he blames his father for marrying his mother (36) and threatens to desecrate his father's tomb for it (“I will exhume your bones / And hang you / By the neck,” 38). In the same way, he gloats over his cold-blooded murder of the politician (“Yes / I did it / And, / My God, / What a beautiful / Shot!” section 8, 66) and later complains that he was not a “thug” and deserved congratulations (67-68). He still pictures himself as a political hero, despite his disillusionment: “I want to raise my hands / And acknowledge / Their cheers / I want to shake hands,” section 10 (77).

V

The kinds of cohesive properties of the lexis discussed above are not the only possibilities in this long poem. But they are significant because they show the deliberate poetic design in the choice of words and their arrangements to produce some of the effects described. This attempt, in the tradition of interpretative stylistics, has involved “breaking the poem to pieces” for its different facets of interpretation and interplay of elements. (See Gregory 1974; Handscombe 1970; Leech 1970; and Sinclair 1966.) As. R. J. Handscombe (1970) has observed,

The means by which a poet creates a feeling or body of feelings in his readers' minds are complex and subtle, sometimes so subtle and complex as to pass unnoticed, for in these circumstances it is the emotions that are first required to respond and not the intellect. But it is only when we can understand—and, hopefully, explicate—the means that we can make any progress with the meaning.

(30)

The discussion here has made it obvious that the violence done by and to the prisoner, the brutality with which he is treated, his general discomfort as he laments his fate in his cell in section one, permeates the lexis of the poem. It is to be seen as an important factor in cohesion in relation to the other movements of the song. Sexual images also pervade the text. These tie in with a strong “libidinal” sexual theme that has been traced, especially in relation to Prisoner's delusive recapture, often very vividly, of the desecration of his matrimonial bed by the “Big Chief” (44) in the fifth movement, and his desire to sleep with French girls, drink cognac, and go on a sexual jamboree (in sections 12 to 15, 89-118; Heywood 1980).

Reiteration is a major cohesive factor in the poem's unity. The reiterative pattern varies, and this variety lends intensity to the poem. For example, the clause pattern “Listen to the footsteps …” in the first section is repeated in other movements, but the objects of our attention are varied as the prisoner draws attention to other aspects of the society's ills:

Listen to the Chief's dog

(22)

Listen to the Song / Of the flies

(52)

Listen to the millipede / Whispering a lullaby …

(81)

Listen to the sandy tunes

(105)

Listen to the wailing tunes

(105)

Listen to the orphans;

(114)

and so on.

This structure ties the different movements together grammatically.

This kind of cohesive pattern is also present in the repetitive recurrence of the I plead + direct object structure first used in section one (12). It is repeated with variation in the third movement (“I plead smallness”; “I plead fear”; “I plead hopelessness”; and so on., 33), which coheres anaphorically to the choral pattern “Do you plead / Guilty / Or / Not guilty?” in the first movement (and in most of the other movements: 21, 30, 32, 44-45), and “I plead hunger,” “I plead insanity” (of the second movement, 20) and, cataphorically, to “I plead guilty to pride” of the thirteenth (100). A significant relationship exists between the prisoner's ironic acceptance of being guilty of the various (“hunger,” “poverty,” and so on.) “misdeeds” that readers realize are not crimes, and the satiric intention of metaphorically drawing attention to his plight and the injustice of the authorities who imprison him for non-crimes (Ofuani 1986).

These kinds of cohesion are responsible for the thematic unity of the poem. According to Adrian Roscoe (1977), the poem's theme of post-colonial disenchantment is not new, for many writers have explored it since the early 1960s; what is new is that “not only is the subject weighty” but that “its treatment tells us so” (54). This treatment is partly achieved through lexical cohesion and is done so perfectly and appropriately so as to give the poem its status of being “Okot's angriest poem,” and helps it achieve the “symmetry,” the balance that Roscoe has aptly observed exists in the treatment in this work in which Okot p'Bitek “applies all his strength” (54).

Notes

1. Song of Prisoner is here treated as a single unified text. A text, defined as “any passage, spoken or written, of whatever length, that does form a unified whole,” is the premise for any useful discussion of the concept of cohesion. See Halliday and Hasan: 1.

2. A complementary relationship should exist between stylistic and literary critical analyzes of texts. For instance, Ayo Banjo (1982: 9) sums their interrelationship by saying that “the stylistician is not in competition with the literary critic but rather offers a solid base on which to anchor our literary criticism.” Stylistics is therefore an aid to literary criticism.

3. Heywood (1980: 65-83) has tried to justify Prisoner's rage in this poem. She saw it as an indictment of social abuses and the pervasive sense of collusive pollution, a rage that is occasioned by the oppression and betrayal of the down-trodden by the elitist bourgeoisie.

4. Song of Lawino has 13 sections; Song of Ocol has nine; and Song of Malaya has seven.

5. The linguistic model used here is Halliday (1961: 241-92). In the grammar, the noun group (phrase) has three main constituents: Modifier (m), Head (H) and Qualifier (q). The Head is obligatory; the modifier and qualifier are optional elements.

6. This reading seems responsible for Blishen's title to the American edition of the Song in which he tries to suggest that the Prisoner was Patrice Lumumba of Congo (Zaire). See Blishen 1971.

7. Such a referent is in the language of case grammar performing the function of “Force” in initiating actions in which it has no volition. See Traugott and Pratt 1980: 192.

8. Chomsky, Aspects of the Theory of Syntax. An alternative approach is provided by Angus McIntosh (1961: 325-37), which discusses such “collocational” patterns in terms of the claim that lexical items belong to sets that have certain “ranges,” that is, tolerable extensions.

This paper was presented in an earlier version at the 7th Ibadan International Conference on African Literature, University of Ibadan, August 1982. The author thanks Dr. Niyi Osundare for his suggestions and criticisms.

Works Cited

Banjo, Ayo. “The Linguistic Factor in African Literature: A Keynote Address.” Unpublished paper, 7th Ibadan International Conference on African Literature, University of Ibadan, August 1982: 9.

Bazell, C. E. et al., eds. In Memory of J. R. Firth. London: Longman, 1966.

Blishen, Edward. Introduction to Song of a Prisoner. New York: Third Press, Joseph Okpaku Publishing, 1971.

Chambers Twentieth Century Dictionary. New Edition, with Supplement. Ed. A. M. Macdonald. Edinburgh: Chambers, 1977.

Chomsky, Noam. Aspects of the Theory of Syntax. Cambridge, Massachusetts: MIT Press, 1965.

Collins, Terence. “Self-Image Through Imagery: Black Arts Poets and and the Politics of Excrement.” Maledicta 3, 1 (1979): 71-84.

Gathungu, Maina. “Okot p'Bitek: Writer, Singer, or Culturizer.” In Standpoints on African Literature, ed. Chris L. Wanjala. Nairobi: East African Literature Bureau, 1973: 52-55.

Gregory, Michael. “A Theory for Stylistics Exemplified: Donne's ‘Holy Sonnet XIV.’” Language and Style VII, 2 (1974): 108-18.

Gutwinski, W. Cohesion in Literary Texts. The Hague: Mouton, 1976: 26.

Halliday, M. A. K. “Categories in the Theory of Grammar,” Word, 17 (1961): 241-92.

———. “Lexis as a Linguistic Level.” In Bazell: 148-63.

———, and Ruqaiya Hasan. Cohesion in English. London: Longman, 1976.

Handscombe, R. J. “George Herbert's ‘The Collar’: A Study in Frustration.” Language and Style III, 1(1970): 29-37.

Heron, G. A. The Poetry of Okot p'Bitek. London: Heinemann, 1976.

Heywood, Annemarie. “Modes of Freedom: The Songs of Okot p'Bitek.” Journal of Commonwealth Literature XV, 1 (1980): 65-83.

Leech, Geoffrey N. A Linguistic Guide to English Poetry. London: Longman, 1969.

———. “‘This Bread I Break’—Language and Interpretation.” In Linguistics and Literary Style. Ed. Donald Freeman. New York: Holt, Rinehart and Winston, 1970): 119-28.

lo Liyong, Taban. “Two Songs by Okot p'Bitek: Song of Prisoner, Song of Malaya.” Dhana: Makerere Arts Festival ‘71 (1971): 59-61.

Luce, Stanford. “Increment and Excrement: Celine and the Language of Hate.” Maledicta 1, 1 (1977): 43-48.

Marshment, Margaret. “Reply to Atieno-Odhiambo.” Busara 4, 1 (1972): 63-70.

MacIntosh, Angus. “Patterns and Ranges.” Language 37 (1961): 325-37.

Ofuani, Ogo A. “Okot p'Bitek: A Checklist of Works and Criticisms.” Research in African Literatures, 16, 3 (1985): 370-83.

———. “A Stylistic Analysis of Okot p'Bitek's Poetry.” Ph.D. thesis, University of Ibadan, May 1985.

———. “The Form and Function of Repetition in Okot p'Bitek's Poetry.” META: Journal of Translators 34, 4 (1986): 300-13.

———. “Digressions as Discourse Strategy in Okot p'itek' Dramatic Monologue Texts.” Research in African Literatures 19, 3 (1988): 312-40.

Ogunyemi, Chikwenye Okonjo. “‘The Caged Bird Sings’ African Prison Poetry.” Review of International English Literature 13, 4 (1982): 65-84.

Ojuka, Aloo. “Two Songs: A Discussion, 2.” Standpoints on African Literature, ed. Wanjala.

p'Bitek, Okot. Song of Lawino. Nairobi: E. A. P. H., 1966.

———. Song of Ocol. Nairobi: E. A. P. H., 1970.

———. Song of Prisoner. In Two Songs. Nairobi: East African Publishing House, 1971.

Roscoe, Adrian. Uhuru's Fire: African Literature, East to South. Cambridge: Cambridge University Press, 1977.

Sinclair, J. McH. “Beginning the Study of Lexis.” In Bazell: 410-32.

———. “Taking a Poem to Pieces.” In Essays on Style and Language. Ed. Roger Fowler. London: Routledge and Kegan Paul, 1966: 68-81.

Traugott, Elizabeth Closs and Mary Louise Pratt. Linguistics for Students of Literature. New York: Harcourt Brace Jovanovich, 1980.

Wanambisi, Monica Nalyaka. Thought and Technique in the Poetry of Okot p'Bitek. New York: Vantage Press, 1984.