Voices in the Courtroom

The most personal piece of evidence we have about the women and men who are the subject of the Digital Panopticon is their words, as recorded in the Old Bailey Proceedings. But, what was lost in translation in the space between the spoken and the printed account? And how did the defendant experience the trial process? Words spoken in court carried all the significance of an accent – now unrecoverable; all the significance of cadence and tenor – now lost; all the significance of gesture and body language – now gone. The markers of class and origin shaped how words in court were received; as did the architecture of the courtroom itself – a stage upon which your location was all-important, whether in the dock, witness stand or on the judges bench. As importantly, those words were recorded in shorthand and edited for publication; transformed from individual speech into generic text. It is only by understanding the context of the words uttered in court, that their full meaning can be assessed (Hitchcock and Turkell, 2017; Klingenstein, Hitchcock and DeDeo, 2017).

Joseph Hunton at the Bar of the Old Bailey British Museum 1919.0201.1 ©Trustees of the British Museum.

The Proceedings as Evidence of Speech

The "Proceedings", available online as The Old Bailey Online, form the largest body of direct recorded speech by non-elite people ever created and the spine around which the Digital Panopticon has been built. From at least 1778, when John Wilkes as mayor of London ordered that they should contain a comprehensive account of the trials held at the Old Bailey, to the end of the series in 1913, the "Proceedings" form a relatively comprehensive record of at least a proportion of what was said in court. Their great strength lies in their consistency. But the "Proceedings" do not fully reflect what actually happened. Sodomy and rape trials, for example, were censored from the late 1780s onwards. But more importantly, much of the back and forth of the trials was edited out of the text.

A simple measure of this process can be found by comparing one of the few trials for which we have what is claimed by the shorthand recorder to be a full, verbatim record, against the rest. In 1761, the shorthand recorder claimed that one trial had been reproduced, "verbatim, as delivered by the witnesses for the crown and the prisoner, without the least alteration or abridgment in any part of it" (t17610225-18). This trial was lurid in detail and scandalous beyond measure – and revolved around the death by gangrene of Anne Bell following a sex fuelled weekend in a Bagnio during which Willy Sutton stabbed her in the arse. The trial contains 26,600 words of transcribed speech divided into over 1,473 separate questions and answers (utterances), which were, on average, 17 words long. This was a staccato back and forth of query and statement; of defence and cross-examination.

By comparison, among the 21,000 trials marked up for speech acts between 1780 and 1880 by Magnus Huber as part of the Old Bailey Corpus project, the average number of words spoken was just 550 per trial, divided among 13 or 14 separate utterances. Each utterance in this larger sample of Old Bailey trials was on average 40 words long. The difference between 17 words per utterance and forty words is the measure of the intervention of the short hand recorder’s editorial blue pencil; and a clear measure of the distance between the trial itself and the published trial account (Huber, 2007).

Colloquial language was also silently transformed into a variety of standard English. At the final trial of Jack Rann, the celebrity highwayman (Sixteen String Jack) in 1774, the shorthand recorder claimed he had printed Rann’s defense statement "verbatim et literatim" or ‘word for word’. Rann’s defence statement began:

"I knows no more of it than a child does unborn, nor I never seed Mr. Bell before he came to Sir John's…"

In contrast, Rann’s co-defendants had their defence edited for print. William Collier, up next, is recorded as saying:

"On the 26th of September last I called at Miss Roache's lodgings. When I came in I asked if she was within. There was a stranger, who … opened the door… and desired I would walk in…"

The "Proceedings", despite preserving the semantic content of the words spoken, dispensed with the nuance of normal speech (for changing content, see Klingenstein, et al, 2017). And of course, in the process, they also lost the further detail of the by-play with the court. The "Proceedings", for instance, are almost entirely silent about audience reaction. In a vanishingly rare example, again in the trial of Willy Sutton, the shorthand recorder had to explain in a footnote that the evidence was given in direct response to: "a very great hissing in court".

The ‘Old Court’ at the Old Bailey, 1824 Collage 5592 ©City of London.

The Courtroom

One way of understanding the speech of defendants is through an exploration of the courtroom itself. The original medieval court on the site was rebuilt in 1673, and this building remained in use through the late 1770s. Originally open to the elements to the north, the courtroom was enclosed in 1737. This was done for the comfort of its judges and jurors, and for the next forty years – through a series of outbreaks of typhus, that killed at least one Lord Mayor sitting as a judge – this cramped and older courtroom served London (see History of the Old Bailey Courtroom).

But the most significant transformation in the history of the site came in 1774 when George Dance re-designed the Sessions House. Over the next half decade, the rebuilding went on, creating what would become a standard layout for British courtrooms. Dance’s courtroom formed one of the first of its sort, and was designed to accommodate a specific kind of trial, with specific places for each individual actor, including the judge, lawyers, jury, witness and defendant. Up until this time courtrooms had tended to be large open and flexible spaces capable of being used for different types of trial and administration(Mulcahy, 2003). But, from 1774 onwards, British courtrooms increasingly adopted a design geared towards the single drama of a jury trial. From beginning to end Dance carefully adapted his design to the specific form of the trial at Old Bailey – including room to accommodate the multiple judges involved, the two juries, one each for Middlesex and one for the City, and spaces for the shorthand recorders – who uniquely translated the events of the trial into the printed public record of the "Proceedings". The basic outline of Dance’s courtroom can be found reflected in almost every court built in the British Empire for the next 150 years. Attacked and rebuilt after the Gordon Riots in 1780, and substantially remodelled in 1858, it was Dance’s building and court layout that serve as the ‘Old Court’ for the whole the nineteenth century, until it was finally demolished in 1907.

3D Model of the ‘Old Court’, c.1808’. ©Ben Jackson.

A "New Court" – smaller but with the same basic lay-out - was then added in 1824, behind the "Old Court"; and a "Third Court" – again on the same plan - was added in 1847. The building was substantially remodelled in 1858, and the layout of the ‘Old Court’ was turned through 90 degrees - increasing the space for barristers and court officers in the process. Along the way, new facilities for judges and juries, flush loos, and an increasing separation between, defendants, witnesses, jurors etc., were all built in to the structure. In 1810, a secure enclosed route from Newgate Prison – fifty yards to the North – was built, connecting directly to the area below the dock. This created the possibility that a defendant could be both "brought up" directly into the dock – and "sent down", and ensured that the victim and defendant would not meet in the process of seeking judgement. The court itself, and its subsequent evolution, represent what Julienne Hanson describes as a new spatial syntax – in which the forms of a new administrative system were gradually built into the workings of the courtroom (Hansen, 1996).

The Trial as Drama

Drawing on original architectural plans and contemporary images, Ben Jackson has built a model of the ‘Old Court’ in 3D Studio Max, as a way of interrogating how the space worked.

This modelling process suggested that the design embedded specific assumptions about the relationship between different actors – defendant, witness, judge and jury. There is a clear transition with the building of this new courtroom form 1774, whereby the defendant was gradually differentiated – both raised up, and set back from the witnesses and victim. The addition of a separate enclosed path from Newgate Prison to the basement of the Sessions House in 1810 re-enforced this transition. As a result, the character of conflict at the heart of a "trial" was re-imagined with the witnesses and victim re-located, ever further from the defendant, but with both facing the judge rather than the jury. The essential equivalence between victim and defendant, both standing before the bar, was replaced by a system in which the defendant was graphically named as the accused, simply by being placed in the dock.

The modelling process also revealed the significance of the different levels in the court. In this courtroom, barristers were forced to speak upwards to the judge, jury, witnesses and defendant from a cock-pit substantailly below the eye level of the rest of the room. Like a theatre audience, the judge, jury and defendant looked down on the stage below. What was created was the real feel of a "theatre" in which, as a barrister, you were forced to perform to the gods – and judging by all the depictions we have of contemporary lawyers declaiming in court – to do so, with all the gestures and conventions of the new theatricality of the eighteenth and early nineteenth centuries. The courtroom demanded a performance (Devereaux, 2013).

All of which simply re-enforces the sense that the words recorded in the "Proceedings", need to be read in light of all the information they fail to encode.

The Defendant's Voice

The question also remains, who was actually speaking? Building on Magnus Huber’s work and The Old Bailey Corpus, we are at an early stage of mapping the distribution of speech in the trials.

The distribution of speech acts, 1760-1878 ©Digital Panopticon.

The marked decline in the percentage of trials with speech text (the yellow line in the graph), reflects the rise of "plea-bargaining" (cases of which include no "speech" beyond a plea of guilty) and which become more common from the 1840s onwards. This line reflects the bureaucratisation of the trial process, and the movement of judgement from the courtroom, to the police interview room. The blue line – the "number of utterances per trial" - effectively measures the role of lawyers and witnesses, as a trials where there is a large amount of conversation and cross-examination will have a larger number of "utterances". Again, as with the percentage of trials with speech, this seems to reflect a dramatic decline in the role of the trial as a complex conversation. And finally, the green line – "average word count per utterance", reflects a major transition in the 1810s and 20s. By this point, when someone did speak, rather than short interrogations, we get ever longer statements.

The distribution of trials where the defendant speaks ©Digital Panopticon.

In parallel with the growing number of trials where no speech is recorded; there is also an apparent suppression of the defendant’s voice; so that by the mid-nineteenth century, when speeches themselves were getting longer, defendants were increasingly mute. Additionally, dividing the same data between those found guilty and not guilty suggests that there was a significant change in the nature of the underlying trials. The voice of those eventually found guilty by the system, was increasingly excluded from the "Proceedings", suggesting that the relative balance of authority in this theatre of justice, was changing.


Conclusions drawn from the kind of indirect evidence interrogated here, are inevitably tentative. But, it is likely that George Dance and the new eighteenth-century courtroom created a new kind of theatre of judgement – where the assumptions of the criminal justice system were encoded in the placing of individuals in a new physical relationship to one another. This was substantially to the detriment of the defendant, and can be thought of as a new "spatial syntax" in which the bureaucratic control of the criminal-defendant was placed at the heart of the system. For the first fifty years after it was created, however, this "theatre of justice" remained available to most defendants, and continued to work as a bureaucratic system in which real decisions about guilt and innocence were made, in real time, in the courtroom. But, the evidence of the distribution of speech suggests that once this theatre of judgement was in place, its workings were gradually restricted to an ever smaller proportion of trials. The rise of plea-bargaining, and of a professional police ensured that an ever growing proportion of trials were determined outside the court. In these instances, the courtroom served as little more than a theatrical rubber stamp. This same evidence also suggests that for the declining number of trials where the guilt or innocence of the accused remained to tested in court, the theatre of the courtroom and the trial became more important; both more long-winded, and arguably more dominated by the voices of the court officers. What begins to emerge from this work is at least the outline of a new story of the development of the criminal trial over the course of the nineteenth century.

Further Information

Author Credits

This page was written by Tim Hitchcock and Sharon Howard, with additional contributions by other members of the Digital Panopticon project team.