“He Goes Off and I Think He Took the Child”: Narrative (Re)Production in the Courtroom

(2006). “He Goes off and I Think He Took the Child”: Narrative (Re)Production in the Courtroom. King's Law Journal: Vol. 17, No. 1, pp. 71-95.

struggle to have their view and interpretation of events accepted by the court. 4 This struggle can lead to the mobilisation of different interpretations based upon alternative discourses. Discourse in this sense can be taken to refer to "a specialised language, a particular jargon"; "a more or less coherent set of categories and theories of action"; 5 a body of knowledge on a particular subject-such as criminology-or sequences of statements that form conversations and narratives; 6 or a system of communication that utilises "common reductive terms". 7 Whatever the terminology used, what is important to grasp is that, either implicitly or explicitly, courtroom actors may draw upon different discourses at different points in the proceedings when defining action.
The choice of discourse used to describe events affects the manner in which an event can be seen or categorised. 8 While different discourses can be mobilised in support of pre-existing orientations to problems, some actors are pre-disposed to utilise favoured discourses. Lawyers are trained to fashion conflicts and problems within the discourse of law, while a social worker may perceive an event through the lens of an alternative or competing discourse. A brief fictional example will suffice in explaining the importance of discourse in interpreting events. A neighbour calls out the police to a late night disturbance in a block of flats. Upon arriving, the police find a husband and wife in the middle of a heated exchange and both parties have resorted to violence. Different observers of (or participants in) this event will define it in different ways. The neighbour may see this as a disturbance that has shattered the peace and tranquillity she expects in her living space. She wants this disturbance to end so that she can continue to read her book, so she calls the police because they possess the powers to end the disturbance. The police, however, may define the situation in numerous ways and each suggests possible solutions. The incident may be seen as a domestic with public order implications (the disturbance to the neighbour) but justifies no other action. The police will therefore act to end the disturbance as soon as possible with no perceived need to initiate further action. Alternatively, the police may view the problem as a situation where a criminal offence has been perpetrated and the offender may well be arrested and criminal proceedings initiated. The participants may view these activities as symptomatic of marital breakdown and either seek guidance and counselling or decide to end the marriage. One partner may view the actions as resulting from the drinking activities of the other and request that help be sought to treat the addiction. The local authority may view this disturbance as part of a sequence of anti-social behaviour and may evict them from the flat, or initiate civil proceedings such as an injunction or an anti-social behaviour order to address the problem. Social services, relatives, medical professionals and others may all interpret these events in differing fashions. The main feature, however, is that each position acknowledges a discourse that names the problem, interprets its meaning and "points to a solution". 9 Consequently, no one approach has any overall claim as truth. The claims that the protagonists make as viable solutions make sense within each discourse and, furthermore, no discourse makes a more appropriate validity claim than another. Indeed, the claims made really only make sense when examined from the perspective of the implied discourse utilised. If the police officer were to say to an addiction worker that "this person has committed an offence and needs to be placed before a court", the response would presumably be that the police officer has "missed the point", that the behaviour is conditioned by a disease that needs to be treated. Each protagonist is correct on their own terms, but communication between the disciplines is difficult, as the statements of each discourse make sense within their own realms. Legal discourse, however, holds a privileged position as it is supported by the power of the state, meaning that legal truth claims have a special status. 10 This, according to Goodrich, is evidenced in how legal discourse relates to other discourses: In terms of legal dialogue, the 'imagined community' of the legal genre . . . evidences its discursive power through an explicit superiority of attitude towards and exclusion of other discourses and usages. 11 It is to law's relation to alternative discourses that we now turn.

C. LEGAL AUTOPOIESIS
Autopoiesis, when applied to law, is concerned with how law, as a system of communication, makes legal communications and interprets the communications of other systems and in so doing, offers insights into how law copes with a plurality of voices. At its basis, autopoietic theory sees law as a system of communication in a society "made up of multiple systems of communication", where "[c]o-ordination is made possible by the use of common reductive terms, self-referential communication, and (at the level of discourse) widely shared values". 12 The legal system is described as a "normatively closed" yet "cognitively open" system. 13 Law is cognitively open in that it is receptive to outside influences; law as a system addresses issues and debates originating from other disciplines and discourses. Law will (and perhaps inevitably) consider(s) problems that are more appropriately addressed by other discourses. For instance, antitrust legal regimes consider economic theories of competition when addressing the problems of anti-competitive behaviour and the market inefficiencies thereby created. In a similar manner, the judges in Conley and O'Barr's research 14 were required, in some instances, to address the wider social issues that litigants brought to the courtroom. However, because the legal system is normatively closed, the engagement with alternative discourses operates in a manner that transforms these other discourses. Alternative discourses are not merely transplanted into law; rather they are translated into the discourse of law. 15 This is explained as a consequence of the legal system "reproducing itself by legal events and only by legal events". 16 The inputs from other systems into the legal system have to be translated into the legal code.
The insight that legal discourse or legal communications interpret and translate outside discourses is not one unique to legal autopoiesis. Goodrich, for instance, in Legal Discourse, comments how law can "exclude and stigmatise-define out" alternative voices. 17 In Languages of Law, Goodrich explores how the voices of the Haida Indians were largely ignored in Western Forest Products Ltd v Richardson and Others. 18 This translation process is not confined to law, but is rather a feature of social systems; "within each system of communication, one can only judge the validity of cognitions-statements about the world-by referring back to the internal procedures of that system which determine the validity of a particular piece of knowledge". 19 The law, in utilising this abstraction process, reproduces events into a legal/illegal binary structure. 20 The narratives presented to the law by witnesses, defendants or complainants must be either presented in this form or translated to this form. The legal system, qua a system of legal communications, must translate other discourses into legal communications: For to be clients of the legal system, people have to operate within the system. They have to be aware of a legal problem; have to define their situation accordingly and have to commit themselves to advance legal claims or at least to communicate them. They participate in the legal system using its system-reference to give meaning to their activities. 21 Although utilising different language, Conley and O'Barr drew similar conclusions about the translation of narratives by the legal process through the production of a "case". 22 They noted how the translation process leads to a recounting of events that selects legally relevant facts from these narratives in the construction of cases. 23 This process is problematic, because when restructuring "life-world" relations in this manner, 24 legal translations will inevitably involve substantial changes to the participants' narratives. Merry commented particularly upon the "paradox of legal entitlement" and noted that, while courts may offer remedies to problems that citizens face, to come to the law involves subservience to the categories and categorisations of law. 25 Translation also suggests an ideology of law that constructs individuals as free autonomous agents able to participate fully in social relations. The law creates the binary legal/illegal opposition in a manner that abstracts legal cases from a wider social context while hiding the ideological basis of the case as a construct: these conceptions of legal subjectivity express a dogmatic articulation of the image and nature of man and their function in law. In law, human rationality and the concept of a free and autonomous will are the basic repositories of the nature of man. 26 This process not only operates at an abstract level, where we can glean assumptions on social relations in the operations of law, but also works for the subjects of lawdefendants, witnesses, plaintiffs and complainants-who are all legal constructs for the purposes of legal operations. The process of abstraction, and the imposition of the binary code upon action, necessarily leads to a partial and incomplete framing of subjects before the law. Victims are just that: victims. The person has been reduced to an "artefact" whereby the law deems the relationship of subjugation to the defendant, and the harm caused, as the only relevant details. Legal subjects are: not real flesh-and-blood people, are not human beings with brains and minds . . . they are mere constructs, semantic artefacts produced by the legal discourse itself. 27 Or, in the words of Goodrich, the: [p]articularisation or decontextualisation of legal discourse is its most significant ideological hallmark: concrete social relationships and real (social) people are transmogrified into the abstractly free and equal legal subjects of the legal code. 28 For instance, returning to our earlier example, to view the domestic incident as a crime prioritises the legal explanation over alternative explanations and, at the very least, pushes these alternatives to the background. The criminal law, in determining criminal liability for action, is concerned purely with ascertaining whether the defendant committed the prohibited act (actus reus), with a blameworthy state of mind (mens rea) and is unable to rely upon a legal defence.

D. NARRATIVE REPRODUCTION, LANGUAGE AND LINGUISTIC NEGOTIATION WITHIN THE COURTROOM
As has been explained, the courtroom is a site of specific discourses and languages where the regular participants feel at ease with these discourses. Law, as a system of rules, as an institution or as a cultural practice, necessarily concerns the manipulation of language: "language penetrates the legal system, and the law perhaps more than any other is a profession of words, ultimately and utterly dependent on some form of linguistic negotiation". 29 On a very basic level, site specific languages can create problems for those not versed in the language, in addition to problems associated with not understanding the rules of procedure, evidence and the informal assumptions concerning courtroom interaction. 30 But the importance of language in the courtroom is not only to be found in the confusion created when participants are not familiar with implicit courtroom conventions; it is also a valuable window into the cultural world of the participants. When courtroom participants perform speech acts they are engaged in a process of production that reflects cultural positioning. For instance, Goodrich noted that: Linguistic structure itself encodes inequalities of power and is also instrumental in enforcing them. The linguistic structures of a text or of a particular institutional practice are thus a matter for critical interpretation. In so far as the text reflects and expresses the roles, purposes and ideologies of its participants or subjects, these implicit or unconsciously regulated operative meanings are accessible to study through their expression in the lexicon, syntax and semantics of the text. 31 Similarly, Harris commented on how magistrates' language offers valuable insights into the ideological basis of the legal system and the particular interaction. The courtroom interaction is said to offer a "tangible connection with more abstract concepts inherent in the legal process". 32 This proposition suggests that an examination of courtroom discourse illuminates the ideological assumptions that support and legitimise law. 33

THE IMPORTANCE OF STORIES
One method of inspecting courtroom utterances is through an inspection of courtroom narratives. Narrative is a specific language form that carries general significance in addition to carrying specific value in the legal process. Storytelling is a pervasive cultural activity 34 whereby we make sense of the world: Our very definition as human beings is very much bound up in the stories that we tell about our own lives and the world in which we live. We cannot, in our dreams, our daydreams, our ambitious fantasies, avoid the imaginative imposition of form on life. 35 Narrative is important in the legal context as people come to the law with problems. These problems can be presented as a narrative, with the presenter as the central character; as someone who has been harmed or suffered a violation of personal interests. A legal audience must listen to these narratives and then sift through them for legally relevant facts while applying the law to these facts. This is the nature of the construction of legal cases. 36 The lawyer must then reconstitute the narrative in the courtroom in a manner appropriate to that particular setting, all the while being aware that an opposing lawyer may be constructing a counter-narrative, either on the basis of a different version of events or by interpreting the agreed events in a different manner. 37 The courtroom lawyer, in constructing a narrative, has to deal with the contradictions and gaps that appear in our narratives, in addition to attributing legal relevance to these events. The courtroom lawyer therefore, "must at once elicit and construct a story, and the distinction between the elicited and the constructed is by no means clear". 38 There can be little doubt that courtroom participants are engaged in the process of either constructing or interpreting narratives. 39 This leads to a question of how courtroom participants construct or hear stories.

NARRATIVE CONSTRUCTION IN THE COURTROOM
On a very basic level, stories or narratives recount a series of events with a beginning, a middle and an end, and these events are organised into a coherent whole that makes sense and gives events meaning. When we tell stories, we do so to make a point; narratives carry an interpretative force that allows the audience to make sense of the story, 40 and a "primary way individuals make sense of experience is by casting it in a narrative form". 41 For Bennett and Feldman, the story is the means by which lawyers in the courtroom organise events in a meaningful manner, resulting in the story being an almost universal feature of courtroom interaction. 42 Lawyers need to grasp disparate events, illuminate a sense of purpose and motivation within human actions, and gather all these events and motives into an understandable, believable and plausible whole. This is achieved through the narrative. 43 For Bennett and Feldman, the success or otherwise of a line of reasoning depends upon how it fits within the overarching narrative that has been constructed and whether or not this narrative is plausible. 44 However, this narrative must be constructed by trial lawyers through the turn-taking sequences associated with courtroom interaction. 45

NARRATIVE (RE)CONSTRUCTION
We are beginning to see that narrative construction is an important activity and one that is crucial within the courtroom. However, the act of narrative (re)production is not simply the recounting of experience; narrative performance involves the (re)construction of events and that inevitably invokes a creative process. This involves omitting irrelevant details ("flattening"), exaggerating important points ("sharpening") and polishing other features of the narrative so as to remove unsuitable material ("rationalisation"). 46 Whenever we tell stories, we do so at different times, to different audiences, to make different points. At each of these different times, the focus of the narrative will change, leading to a different emphasis and the removal and addition of details. This process is not merely a reaction to the audience; it is creative, whereby reality is readdressed on each telling. Riessman, for instance, noted how some narrative theorists see speech acts (and therefore narrative (re)production) as constituting the world: Skeptical about a correspondence theory of truth, language is understood as deeply constitutive of reality, not simply a technical device for establishing meaning. Informants' stories do not mirror a world "out there". They are constructed, creatively authored, rhetorical, replete with assumptions, and interpretive. 47 Narratives are created with the assistance of "schemata": "an active organisation of past reactions and past experiences which organises elements of recall into structured wholes". 48 They are organised around a simple structure that we implicitly use and acknowledge as a framework when we tell stories. The schemata perform a number of functions that allow for the organisation and recall of information: they represent a "prototypical abstraction" of a concept; they help to organise information through the use of "variables or slots that can be filled" whenever we receive information; they guide "the interpretation of incoming information" so as to help us make sense of the world; and we can fill in the gaps when the "expected information does not appear". 49 In short, "[m]emory load is minimised by stripping away inessential details" and "[o]nly sufficient detail of the original event or story is kept to allow a reconstruction on recall". 50 Schemata, therefore, help storytellers to fill in the gaps, they assist in unfolding plots and allow characters to be understood in already existing roles. 51 Schemata, in addition to being important in the (re)production of narratives, also assist in the understanding of narratives by audiences. When we hear stories, we subconsciously attempt to fit characters, their actions and motivations into already existing categories within schemata. When we watch a film or a television programme or read a book, we are tuned to the signals that allow us to assign moral meaning to actions that help us to identify "the good, the bad and the ugly". 52 Barthes, for instance, shows how "The World of Wrestling" is not simply a sport, but is akin to "ancient theatre"; the characters are organised around themes that support moral messages within the spectacle. The "excessive gestures" of the participants help the audience to identify the characters and their roles: Each sign in wrestling is therefore endowed with an absolute clarity, since one must always understand everything on the spot. As soon as the adversaries are in the ring, the public is overwhelmed with the obviousness of the roles. As in the theatre, each physical type expresses to excess the part which has been assigned to the contestant. 53 The audience, through the use of such obvious characters with clear characteristics, can see the play of good versus evil or truth versus dishonesty unfold clearly within the spectacle. When observing the characters, already existing schemata assist in the interpretation of events.
The examination of meanings that audiences place upon stories is closely linked to the work of Van Roermund; this examines what is described as the hierarchy between events and interpretation. 54 The events of the narrative are self-explanatory; these are what happened in a story. The interpretation is the meaning of the narrative or its "discourse". 55 Hierarchy suggests the prioritisation "of one term over the other, in the case Event . . . over Interpretation". 56 Where the event is prioritised over the meaning or interpretation within a story, the meaning is said to originate from an analysis of the events. In other words, when (re)constructing a narrative, we describe the events and then interpret them. However, the event/interpretation hierarchy may be inverted to give the interpretation priority. Rather than events suggesting an interpretation, the interpretations (or schemata) influence the selection of relevant events. In short, narrative (re)production reflects individual world views: According to this presupposition, the narrative can only be understood if one acknowledges that the 'events' referred to are not independent of and prior to the Interpretation, but are rather the products of discursive forces, restrictions and requirements. 57 This takes us back to Riessman above, and the scepticism surrounding a "correspondence theory of truth". 58 Rather than truth being found from observable phenomena, we creatively produce truth from our theories about the world: our epistemic claims are not picturing reality, it is rather the other way round: reality is a picture of our epistemic claims (to be disguised as power claims rather than truth claims). Thus the pole of Event can be reduced to the pole of Interpretation. 59 For Van Roermund, either hierarchy privileges one pole at the expense of the other in a manner that fails to represent narrative (re)construction, both events and interpretation being crucial for (re)construction. Returning to schemata, Cortazzi noted how narrative (re)production was a top down and bottom up process, whereby the audience constructed meaning from the events and the already existing schemata. 60 Acceptance of the strong thesis that interpretation is prior to the event, or the weaker compromise that the interpretation is merely influential in (but not prior to) the framing of events, has serious consequences for the (re)construction and interpretation of narratives within the courtroom. This implies a limitation upon the range of courtroom narratives, with those outside of a dominant form either lost or reformulated to fit with dominant conceptions. Alternative narratives may be either dismissed as fanciful, unbelievable or implausible, or (re)constructed to fit with already existing schemata that flatten and translate the participant's narrative. Bennett and Feldman would even go so far as to state that most instances of bias within the courtroom result from narrative interpretation. 61 Rather than blatantly acting on prejudice, courtroom participants are said to discriminate through a rejection of narratives that do not fit with pre-conceived ideas, thereby leading to a rejection of culturally different interpretations: If legal facts are reconstructed as stories whose plausibility depends on understandings drawn from experience, then jurors who come from differing social worlds may disagree about the meaning and plausibility of the same stories. 62 The result of such processes of narrative (re)production is that the narratives that are delivered in court and accepted are constrained by culturally dominant ideologies, ideas and concepts.
Jackson, in a critique of Bennett and Feldman, points out that the insights of a narrative approach are applicable not only to the story in the trial, but also to the story of the trial. The participants within the courtroom are also actors with motivations and goals and within this context we could also view the trial as a narrative. Jackson therefore calls for an understanding of "the relationship between the narrativisation of semantics and the narrativisation of pragmatics". 63 Nevertheless, Bennett and Feldman's insight into the importance of the story in the trial remains. 80 T

E. THE STUDY
The research, which was based in one CPS area but covered an urban (City) and small town (County) court, began in March 2001. In total, the fieldwork was completed in nine months. The researcher gained access to the CPS for the research; 12 prosecutors were shadowed in the City court and nine in the County magistrates' court. 64 The majority of the fieldwork was spent within the City court; 70 mode of trial hearings were observed in this court and 30 in the County court. 65 An accurate record was made for each of these observations and access to the CPS allowed for detailed notes to be taken from prosecution files. A data capture form was developed that enabled details of the prosecution case and the court hearing to be systematically recorded. Finally, the City magistrates' court was a large court centre with 10 different court rooms that were predominantly used for different purposes. For instance, two courts were dedicated youth courts, a number were dedicated trial courts and others handled administrative work. Within this rota, the administrative courts were divided between early first hearing courts, early administrative hearing courts and custody courts where bail decisions were processed. In order to ensure a representative sample of mode of trial hearings, time was spent in different administrative courtrooms. In the County magistrates' court, all administrative business was conducted in one courtroom, enabling the researcher to remain in that court for the duration of the fieldwork.

F. AN INTRODUCTION TO NARRATIVE ANALYSIS
During the fieldwork period it became apparent how administrative hearings, and in particular mode of trial hearings, failed to follow the structural regularities identified by Atkinson and Drew. 66 Atkinson and Drew approached the courtroom process from the perspective of conversation analysis. This led, inter alia, to a concern with how adjacency pairs, such as question-answer, operated within the specialist setting of the courtroom. The mode of trial hearing, however, followed a different regularity; the prosecutor delivered what appeared to be a standard statement followed by a brief statement from the defence solicitor 67 and a decision by the bench. 68 This suggested that, not only was the prosecutor's statement the main thrust of the hearing, but this Macmillan, 1979). 67. Usually indicating that she had no observations to make. 68. This decision was usually presented without the giving of reasons. followed a regular pattern, not unlike a narrative. In short, the prosecutor seemed to be telling a highly regulated story. 69 As a means of analysing narrative structure, the evaluation model developed by Labov (1977) was utilised to examine the themes and structure of the courtroom narratives observed. 70 Labov suggested that narratives share a number of common features. At their most basic, narratives consist of a sequence of at least two narrative clauses that chronologically describe a series of events. While there must be two narrative clauses at a minimum, it is usual to encounter many more than two. In short, narratives usually have a beginning, a middle and an end. Next, narratives address a number of potential questions; the substance of any narrative answers these questions before the audience asks them. Finally, a common narrative structure can be identified with different sections performing different tasks and addressing different questions.

NARRATIVE STRUCTURE
Labov suggested six elements that may be present in a narrative 71 ; these are listed below. However, while a complete narrative may consist of all six parts to the structure, they are not necessarily required. As explained earlier, all that is required are two narrative clauses.
The Abstract: This is the optional introduction that encapsulates the crux of the narrative. It also creates a space whereby the narrator makes it clear that a narrative will be delivered thereby suspending usual turn-taking conversation conventions.
The Orientation: This sets the scene and provides answers to questions such as "who", "what", "when" and "where?" The Complication: This is the main text of the narrative; it addresses the question, "what happened?" The Evaluation: This informs the audience of the rationale of the narrative and addresses the question, "so what?" The Result: This section may precede or follow the evaluation and is intended as an answer to the question, "what finally happened?" The Coda: This informs the audience that the narrative is at an end. Oral narration, as explained above, involves the narrator, through the abstract, creating the space for storytelling and the suspension of conversation conventions. The coda operates so as to notify the audience that the story has ended and the usual turntaking conventions are reinstated.
. In what appears to be an almost complete role reversal, Heffer, above n36, expected to find narratives clearly within his data but instead found these embedded in the complex turn-taking sequences of lawyerwitness interaction. Each of these elements addresses an as yet unasked question. For instance, the evaluation addresses the question, "so what?" When we tell stories, we do so in order to make a point, address a concern or illustrate an argument. It is clear that narratives need to address this question from the response seen when stories fail to explain their pertinence to the interaction in which they are placed. One just has to think of a social situation where someone tells a story and the audience either misinterprets the thrust of the narrative or fails to understand. Good stories avoid such awkward social situations by addressing these questions before they are asked.

G. NARRATIVES AND LEGAL STORYTELLING IN THE DATA
As explained above, typical narratives consist of different sections that address specific questions before an audience asks them. Narratives can contain an abstract, orientation, complication, result and evaluation, and all of these sections were found within the narratives delivered by prosecutors. However, each narrative section took a particular form within the mode of trial hearing, and this form gave some important insights into the work of prosecutors and criminal justice agencies.

THE ABSTRACT
The abstract is most notable by its absence in the data. As the abstract functions to create space for the delivery of a narrative, it is not strictly necessary in the mode of trial hearing. Conversational analysis suggests that everyday conversation is marked by the use of adjacent pairs, such as question-answer, and complicated turn-taking rules. The abstract is the means by which the narrator informs (or requests of) those listening that normal turn taking conventions are suspended and an extended turn will follow where the narrator tells the story. 72 In everyday conversation we might say "let me tell you a story". However, in the mode of trial hearing, such requests are unnecessary as institutional conventions determine the order and selection of speakers. Courtroom participants know the sequence of utterances within the courtroom and come to expect that certain speech acts will take place, in certain places, at certain times. The prosecutor therefore does not need to begin with an abstract, as all the repeat players know what will happen next. The magistrate will deliver a standard statement asking the defendant for her plea and explaining the consequences of that plea. If there is a plea of not guilty, the prosecutor will then deliver their mode of trial observations, followed by the defence solicitor's statement. Nevertheless, even though an abstract is unnecessary for the court regulars, prosecutors did occasionally introduce their observations with an abstract. For instance, the prosecutor in case 25 opened with a typical abstract: "as far as mode of trial is concerned, the prosecution say that this is a matter which you can try here." Similarly, the prosecutor in case 40 started by saying, "the prosecution say that this matter is not suitable for summary trial". These are just two examples of standard abstract clauses that can be found, albeit infrequently, within the data set. Although mundane, they evidence two considerations regarding the work of lawyers within the legal process. Firstly, they do suggest an expectation that stories begin with a beginning. Secondly, these abstracts show something of the nature of legal storytelling: whilst we might start an everyday story by saying, "let me tell you a story that explains this", or "I know another story about that", lawyers start (and tell) stories differently. This can be seen most clearly in the abstract to case 18: "Sir, there is an aggravating feature". All three abstracts are focused upon the legally relevant details of the case.

THE ORIENTATION
Orientation clauses in narratives set the scene; they introduce characters, the setting and the time, before proceeding to what happened next. Within the mode of trial hearing, it is normal for the orientation clauses to be grouped together at the start. For instance, in case 59, the prosecutor opened with an abstract (the recommendation) and then introduced the characters: Prosecutor: Both the defendant and the complainant at the time of the matter are inmates at [a young offenders institution]. They were at the gym, five-a-side football match. They were on opposing teams. This orientation introduced the central characters to the narrative, explained the location of the incident and what the participants were doing at the time. However, there is little of the character development associated with mainstream narrative forms; the audience was given the minimum detail needed to interpret the complication that follows. Unlike everyday narrative forms, there was no subtle introduction of the characters and their role; the prosecutor was much more explicit on this point, labelling each character-defendant and complainant-in a way that left little to the imagination or interpretation, the roles for each character being fixed from the outset. The institutional site of the narrative heavily influences form. The mode of trial hearing is not an opportunity for storytelling per se, but is a forum for making a particular legal decision.
However, on some occasions the prosecutor expanded upon these minimum details, such as in case 17. The abstract, orientation and complication are reproduced below: Abstract/Orientation Prosecutor: Sir, the ABH is upon his partner [. . .] Orientation What seems to have happened is that they have, she's told him that the relationship is over, she doesn't want him at the premises.

Complication
He forces his way into her address-that's the summary only offence of using violence to secure entry-and she then describes how he assaulted her there so whilst he pushes the door in or kicks the door in, he punches her to the side of her neck which she says causes her to move past. He goes into the living room where the child is-I think the child is about six months old or thereabouts-he picks up the child in his arms and kicks the child's bouncer towards her so that he kicks the outside of her right thigh. He then pulls the phone out of the socket, he takes the handset, she says he hit her with the phone handset two or three times to the back of her head. She puts her arms to push him away, he grabs her hand and bites her, she says, on the left forearm. She says he's holding it for about five seconds-she's got a circular purple mark on her left arm as a result of that. He shoves her with the side of his body causing her to move out of the way. He then picks up the baby's milk, pushchair and what have you. Tries, she tries to take the child back and he uses his hand to push her right shoulder, causing her to fall onto the bed. Then she says he takes her tee-shirt and drags her into the hallway and says he shook her causing her to fall to the floor. She says he then kicks her to her left outer thigh when she turned around. He's shouting about the child, not being able to see the child. She goes into the living room. She says he comes in and stamps on the phone in the living room, runs into the bedroom, he tries to pick the child up, he tries to take the child as well. He then punches her to the left temple. She says she's feeling dizzy. She goes into the living room, she collapses on the floor. She says she's there two or three minutes but she doesn't blank out. She goes back into the bedroom at that stage. He's got the child and the pushchair by then. She says he runs towards her, pushes her back again into the living room, causing her to fall to the floor, and then he gets the phone wire, she says, and puts the telephone cord over her head and onto her neck. She says, "I realised he was going to choke me with the telephone cord, I grabbed it and pushed it away to stop him". She describes herself as feeling very frightened. She says he then grabs her arm and swings her, causing her to fall to the bed, and at that stage he goes off and I think he took the child.
While the orientation section in this narrative is more complete than that outlined as typical in case 59, the mere details given are still insufficient for a full understanding of the interaction. While the three details outlined-the assault is said to take place against the defendant's partner, the relationship is now over, and the complainant no longer wants the defendant at the address-provide some context to this incident (this is a case of domestic violence and the relationship between the parties is breaking down), many more questions need to be addressed to enable a thorough understanding of the incident. For instance, we could ask why the relationship is over. Who, if anyone is to blame for this? Why does she no longer wish to see him at the premises? Does she have something to hide, or is she afraid of him? Does he have a right to be at the premises or an interest in being there? 73 The orientation section in this narrative failed adequately to address any of these (and other) questions, yet there was an indication of a wider context in the text. Throughout the complication, orientation clauses told another tale (these have been italicised in the text) and these referred to the existence of a child who was present during this encounter. 74 Taken as a whole, it is most likely that this incident resulted from an argument over custody of the child; both parties picked up the child on at least one occasion, he complained about not being able to see the child and he eventually left with the child. An appreciation of this possible context, however, simply leads to more unanswered questions; does the defendant have a right to take the child and does he have good reason for doing so?
The analysis so far suggests that narrative (re)productions in court are partial and fail, therefore, adequately to reflect what has occurred. While this is to some extent inevitable-as a (re)counting of events, narratives flatten reality when describing the important features of an event-legal narratives are prone to intensify this process. The mode of trial hearing, being concerned with one particular legal process, the venue decision, situates narratives within this institutional structure. Much of the human interest in case 17 is lost; the courtroom participants do not need to hear this information when assessing the defendant's case. All that is needed is an outline of the legal case against the defendant, along with an evaluation of how this fits in with an understanding of the mode of trial decision.

THE COMPLICATION
The complication is the main body of a narrative; it is here that we find the narrative clauses that (re)produce past experiences and describe the events central to the narrative. 75 The complication (or a string of narrative clauses) is the only essential feature of a narrative: "[o]nly . . . the complicating action is essential if we are to recognise a narrative. The abstract, the orientation, the resolution, and the evaluation answer questions which relate to the function of effective narrative". 76 Case 17, above, illustrates how lawyers present events to the court.

THE RESULT
The result clauses inform the audience of what finally happened to the central characters in the story. Labov suggested that the result is usually delayed until after the evaluation clauses; the narrator concludes a story by stating what finally happened, after pointing out the meaning of the narrative through the evaluative clauses. However, within the mode of trial hearing, the result frequently preceded the evaluation; this can be explained on the basis that result clauses (especially for offences against the person cases) often contain information used to evaluate the events. In this sense they are also quasi-evaluation clauses. For instance, case 18 displayed a typical result section for offences against the person: The prosecutor here utilised the victim's words to describe the result of the assault. If a trip to hospital were required then this would be highlighted at this time. In short, when describing the injuries and outlining what happened, the prosecutor was also describing an important feature of the allegations: the harm caused. Most mode of trial hearings for assaults had a section where the prosecutor outlined the extent of the injuries, even if these were not expanded upon for evaluative purposes.

THE EVALUATION
In the evaluation, the narrator informs the audience of the reason for the story. In the mode of trial hearing, the events are retold so that the court can make a decision on venue; the evaluation is focused upon this reason for the narrative. It was most common to find evaluation clauses grouped together at the end of the mode of trial hearing, along with the prosecutor's recommendation. The following, from case 38, is typical of an evaluation section. The prosecutor noted all the main features of the allegation, and how they tied together to suggest the recommendation: Prosecutor: Your worships, if you take the prosecution case at its highest, we have the pulling out of earrings, which could actually cause very serious injury, although fortunately didn't in this case, and a bite to the cheek, so we've got the use of teeth as a weapon, they're complete strangers to each other-they're not known to each other-and of course it's the town centre in a nightclub, and we have, what on the face of it, the complainant is suggesting unprovoked violence. I'd have to ask you, in all seriousness, whether the case is one suitable for magistrates' court trial. 6. THE CODA Labov suggested that the coda operates to "bridge the gap between the moment of time at the end of the narrative proper and the present". 77 The coda effects a clear indication that the narrator has ceased telling the story and therefore normal turntaking conventions are restored. The recommendation therefore usually operates as a coda, because at that point the narrative has performed its function and the courtroom regulars are aware that the defence solicitor is then at liberty to make her representations. Nevertheless, some prosecutors will occasionally make an utterance that resembles a coda. For instance, in case 38, the prosecutor closed by stating, "[y]our worships, there is nothing else really that I can usefully say".

H. NARRATIVE FORMS
In the mode of trial hearing, prosecutors utilised different narrative forms at differing times. These usually fell into one of three different categories: full or complete narratives; truncated or brief narratives; and finally, prosecutors may not deliver a narrative as commonly understood at all. Each of these different forms will be examined in turn.

FULL NARRATIVES
Some prosecutors in the sample displayed propensities for storytelling, delivering complete and extensive narratives, while others utilised this form only for difficult decisions. Case 46 provides a good example of an extensive narrative, where the prosecutor utilised a relatively complete, if brief, narrative so as to outline the alleged events: Abstract Prosecutor: Sir, the charge involves a sum of around £24,000 allegedly taken. The prosecution case is this . . .

Orientation
. . . the defendant is a financial adviser and one of the people that he is advising is [the complainant] who is a private dentist . . .

Complication
. . . and the essence of the case is that he said [the complainant] should invest lots of money in [a company]. He then, the defendant, the prosecution will allege, opened an account in the name of [this company]. The cheque from [the complainant] was paid into that particular account and the defendant, according to his admission, then took the money out . . .

Evaluation
. . . because he actually had some debts.

Result
He hoped to be able to pay [the complainant] in due course, but it would appear that no money has been set aside, not before now.

Evaluation
That, in essence, is the prosecution case. It is a substantial sum of money and a breach of trust. Nearly all parts of this narrative focused upon legal aspects of the allegations. The abstract noted the value of property taken; important in considering venue. 78 The orientation described the relationship between the parties; this displayed the breach of trust inherent in the allegations. 79 The complication outlined how the defendant affected the transference of the property, while the result questioned the defendant's intention to return the property-an implied claim that he intended permanently to deprive the owner of the property. There was one passing reference to an alternative narrative-a narrative that would explain why the events took place-and this was the comment that the defendant took the property to repay debts. This alternative narrative was outlined by the prosecutor in the police station while reviewing the files; she suggested that the defendant probably had drinking or gambling debts that explained the offending behaviour. She made a number of sweeping comments that suggested an alternative evaluation of the narrative. She stated that "men did this sort of thing" to clear debts and that drinking and gambling was prevalent in City's ethnic minority community. 80 The court narrative focused upon the elements of the offence and a consideration of venue, while this alternative narrative focused upon crude social stereotypes and common sense as a means of explaining what was alleged to have taken place. This is a good example of how a legal focus results in different narratives from what might be delivered elsewhere.

TRUNCATED NARRATIVES
Frequently, prosecutors will deliver what can be called a truncated narrative. Narratives can be truncated in a number of different ways. Either narrative sections are omitted by the prosecutor, or while each section is presented, the prosecutor reduces the events to the bare minimum, losing much of the detail seen in everyday narrative practices.
Case 15 is a good example of a case where all the relevant narrative sections were present yet the narrative was truncated: Abstract Prosecutor: The Crown would ask that you accept summary jurisdiction. This involves the theft of a mobile phone . . .

Complication
. . . whether she could borrow it. She took the phone . . .

Evaluation
Value £50 at most, so therefore summary trial.
Here, there was an absolute paucity of information; there were no details as to the location of the incident, for how long the defendant asked to borrow the phone or how the defendant actually disappeared. There were only three narrative clauses that described the allegations: the request to borrow, the taking, and finally the disappearance. Yet, given the low value and the nature of the taking, the narrative contained sufficient information for the bench to make a decision.
While the prosecutor noted the important case features, at no point did she describe the details of the allegations. This case shows that it is not necessary to describe the allegations in order to make a decision.

I. CHOOSING NARRATIVE FORM
While it is difficult to encapsulate how different narrative forms were utilised by different prosecutors, there are a number of "rules of thumb" that help to explain the different approaches. First, different prosecutors have different propensities; one prosecutor within the sample, for instance, nearly always utilised a storytelling style, while others would frequently abstract from the case its essential features. Nevertheless, depending on the context, nearly all prosecutors could switch between styles. Second, workloads could impact upon style; busy court days demanded that business be expedited as much as possible and, as a result, prosecutors would scale back their observations. Third, different offence types could be treated differently. The truncated narrative was initially identified in a shop theft mode of trial hearing; these nearly always followed a similar pattern, with the prosecutor divulging no details and simply stating that the case would be suitable for summary trial. As shop theft was regarded as obviously suitable for summary trial, there was little reason for the prosecutor to go into any detail. To some extent, therefore, the most truncated narratives would be delivered for the simplest cases, whereas if there was any doubt as to venue, then a more complete narrative could be expected. These three influences interact somewhat, resulting in no obvious rule as to when a truncated or complete narrative could be expected. Some prosecutors would persist with their style, no matter what the case or the court's workload, while others would adapt to the situation.
That full narratives are not needed in simple cases can clearly be seen from an examination of case 45: Abstract Prosecutor: Sir, the Crown would say that this matter is not suitable for summary trial. It involves a taking by the defendant of a police baton, as a result of the police attending the defendant's ex-girlfriend's home address and the defendant striking his ex-girlfriend's father on the head with that police baton, causing eleven stitches to a cut on the back of his head.

Orientation
The defendant had gone to the home address of his girlfriend threatening, demanding the continuation of the relationship and demanding she give him money . . . District Judge: I probably don't need to hear any more.
The prosecutor for this case did have storytelling telling propensities that were a product of her desire to be thorough and well prepared. She gave a thorough abstract (so thorough in fact that for most prosecutors this would suffice for the hearing). We know that this is simply the abstract that introduces the narrative because the prosecutor moved to introduce the parties and the scene of the incident before describing what occurred. Recognising that the prosecutor was about to deliver a full narrative, the District Judge intervened and indicated that he agreed with the recommendation. Obviously, all the District Judge needed in this case was the extent of the injuries and these being so serious the decision was regarded as straightforward. While the representations sufficed for the hearing, the voices of those involved in the incident have, for the time being, been silenced.
It is not always possible to truncate a narrative effectively in this fashion. For instance, in case 21, where the allegations were of making threats to kill, the assumption that the decision was straightforward did not apply. That cases are regarded as easy can be explained on the basis that there is a shared stock of knowledge that assists in interpreting the cases. When this stock of knowledge does not exist, then assumptions on how the case will be understood no longer apply.
Abstract Prosecutor: Sir, it's an allegation that the Crown would submit is suitable for summary trial.

Orientation
Effectively Sir, the complainant in the case, [. . .], and the defendant were business partners. That was ended by the complainant: he was of the opinion that he was effectively doing all the work, paying all the bills and the defendant wasn't effectively paying his way within the business. At the end of it, the defendant took exception to this and he believed that [the complainant] owed him a sum of money: somewhere in the region of £22,000. That's by way of background Sir . . .

Complication
. . . on [a date], [. . .], the complainant, was in [a] shopping centre with some friends. There's a chance meeting with the defendant who approaches him, grabs him by the arm, and effectively frogmarches him out of the [. . .] shopping centre saying, "You're going to take me to your house and you're going to give me the £22,000 or I'll kill you". In the course of the frogmarching, he's taken out of the [. . .] shopping centre and then being walked towards [an] estate, holding him tightly by the arm. During the course of which there is a repeated threat, which the complainant says he took very seriously and was in fear for his safety. Sir, their journey takes them on to [another] street and near to the police station, and as they get there the complainant manages to free himself, runs into the front office and complains to the police. So that, effectively, is the allegation . . .

Coda
. . . I would submit that your sentencing powers are sufficient.
Defence Solicitor: Yes, it's an allegation that's going to be denied quite vehemently, Sir. So far as the court regards mode of trial, matters could be dealt with summarily and my client would be happy with that. In this hearing, there was little in the way of evaluation, although the prosecutor gave some history to the interaction in the orientation. Any evaluation would be expected to draw attention to the seriousness (or otherwise) of the allegations. As seen above, this can be omitted if the case is regarded, or constructed, as straightforward and unworthy of extra comment. However, it is doubtful if the common stock of knowledge important for such cases actually exists for allegations of threats to kill. They do not appear as a specific offence in the Mode of Trial Guidelines 82 and most prosecutors preferred to proceed with an alternative charge. 83 As a result, the infrequency of such cases most probably has not allowed for a common perception to take hold and the usual comfort of working within knowable boundaries no longer applies. As a result, the District Judge had to ask for advice.

J. IMPLICATIONS OF DIFFERENT NARRATIVE FORMS
While there may be good institutional reasons for the particular narrative forms utilised by prosecutors, their effect was to silence the narrative's characters. We can see from case 17, examined above in section G.2, how the narrative (re)production practices of the prosecutor lost the context to the narrative that may have explained what took place. If the defendant had a story to tell as part of this process, this was not raised in the hearing. The merits of the child custody dispute that underpinned this incident were absent. As a result, both parties were silenced. The most disturbing feature of the hearing, however, is that the prosecutor's representations could be designed to reflect the incident from the complainant's perspective. The norm that the bench is to assume that the prosecution can prove their case 84 means that the mode of trial hearing is one of the few occasions where the victim can have their story told in court unchallenged. Yet, the process operated in such a manner as to distort their narrative radically. Once again, this was at its most obvious with truncated narratives, but it also operated when the prosecutor engaged in storytelling. The problem has been identified by Goodrich: After the event, the law arrives to reconstruct the discourse of others-after it has been uttered, after the context of its uttering has become cold or alien to its author-and endows it with significance, relevance and meaning, superficially if not always wholly determined according to the various formal procedural and evidential techniques developed specifically for establishing authenticity and degrees of veracity, as well as according to the substantive hermeneutic techniques which enable the imputation of intentions and the more general elaboration of legally implied meanings, narratives and discourses. 85 Goodrich, in Languages of Law, has commented how the legal process is inaccessible to the layperson because of this process of translation. 86 Similarly, Heffer noted how witnesses can be frustrated when they are unable to tell their story in the manner of their choosing. 87 In particular, Heffer shows how witnesses in a trial have their narratives strictly controlled through the questions of the trial lawyer. 88

OMITTING INFORMATION
Another problem with the truncated narrative is what can be omitted; the prosecutor controls access to information. This was a problem in case 33 and was addressed only when the defence solicitor intervened. The prosecutor in this case delivered a standard truncated narrative. She explained that the defendant was acting as a nightclub doorman and the complainant was a guest. She noted how the complainant was punched several times by the defendant and then outlined the injuries; stitches were required and bones were broken in the fingers and wrist. The prosecutor closed with a recommendation that the case was not suitable for summary trial. The defence solicitor, however, noted a problem with the version of events described by the prosecutor; a witness statement that he received from the prosecution suggested that the broken bones might have been self-inflicted as a result of striking a wall. The prosecutor then had to acknowledge this problem; in effect, the truncation process in simplifying events glossed over the ambiguities and problems in the case. 89 K. CONCLUSIONS This paper has theorised on how legal language is constructed and how legal narratives are (re)produced in a particular manner. The narrative (re)production practices of prosecutors have been examined in a manner that highlights the control of information and the legal focus of the professionals. The silencing of defendants, witnesses and complainants is apparent in these practices. Yet, the extent to which legal practice could operate in any other way as presently constituted is not clear. The utilisation of storytelling would be regarded as wasting time and introducing irrelevant details. The frustration observed by other professionals to those prosecutors who adopted a storytelling approach displays the extent to which such reforms would be resisted.
Prosecutors managed courtroom narratives in a way that accorded with legal concerns. The gap between the cases presented in court and the narratives of the participants resulted from professional practice and the objectives of the legal process. The text produced was not designed as an entertaining story or even as a completely faithful account of what happened. Rather, it was a presentation of a case, and a one-sided case at that. This conclusion may be regarded as a little trite: structuralist analyses are often regarded as trivial or banal; whereas a surface analysis relies upon the full significatory power of the language, and is capable of extracting meaning from the multiplicity of small units to be identified within it, a deep 'structural' analysis appears to reduce this specific detail to a few general themes. 90 It is, nonetheless, important for this. As explained earlier, Jackson regards narrative form as almost universal. 91 If this is the case, it could be argued that storytelling is a pervasive activity that we take for granted; it is one of those phenomena that we fail to notice because it is so prevalent. It only becomes apparent when specifically highlighted.
Throughout all the texts analysed, recurrent themes were identified that displayed the concerns of the professionals. This process of legal construction can also be seen in the omission of information in the mode of trial hearing. Much of the context to the disputes that formed the very material of courtroom work was lost in the narrative reproduction process. Brief orientations, for instance, resulted in partial accounts, 94 T H E K I N G ' S C O L L E G E L A W J O U R N A L 89. A further ambiguity was not addressed. The prosecutor noted how the defendant was working as a doorman and the incident took place as part of his employment. This must be an aggravating feature, as the defendant is employed to ensure the safety of the nightclub customers, not threaten that safety. Yet the witness statement makes it clear that the defendant and complainant were friends. This suggests a slightly different interpretation of events but the truncation process masks such fine details. whereby the richness of a narrative was lost. However, this is also functional for the legal system. The prosecutor is able to paint a clear picture of the protagonists and the interaction. Within case 17 for example, we have seen how the prosecutor omitted to describe fully the access and custody dispute that undoubtedly coloured the events. As this was not explained, there was no means of deciding which party had the better claim to custody of the child. Yet the legal decision that had to be made could best be done without this context to the story. The professionals merely had to decide whether the actions of the defendant transgressed the criminal code; the omission of this context helped in the creation of a legal binary, victim/offender or aggressor/defender. By simply describing the actions of the defendant, in the absence of any explanation, the prosecutor constructed a legal case that was focused upon the legal implications of actions; a picture was created of an aggressive male using violence to achieve goals and a passive woman defending herself, her child and her property. While this may be an accurate description of events, 92 the narrative as constructed in court failed to explore any alternative explanation.
Much of the theoretical approach developed earlier can help to explain the legal narratives produced by prosecutors and the silencing of the parties that results from the (re)production process. Comment was made on how autopoiesis suggests that the legal system "listens" to discourse from other systems. The everyday narratives of the participants are produced in a particular style and legal professionals listen to these narratives with distinct purposes in mind. The information that is then selected as relevant is that which fits with the objectives of the legal system. Likewise, an examination of narrative (re)production suggested that narratives were framed with "mental maps" or "frames" in mind and the narratives thereby produced fitted into this pre-existing schemata. Legal narratives are therefore partial and incomplete; they are indebted to a particular perspective that frames the (re)construction of any narrative. Legal sites do not faithfully reproduce a world 'out there', but instead construct the world in the law's image: there is a close, if as yet unsystematised, correlation between institutional and social practices and their typical forms of utterance and discourse. 93 But this reconstruction by legal processes of everyday narrative is not a value neutral process. Law, and the manner in which legal discourse is fashioned, masks the ideological assumptions that help to form it as a species of discourse: Legal language, in reality a cultural construction, is itself made to appear natural, self-evident. 94 The narratives (re)produced in court simply reflect this reality.