In 1664, Sieur Rausset sold the Caribbean Ile de la Tortue, off the coast of Saint-Domingue (now Haiti), to the French West India Company, stating that although he was “ordinarily residing in that Isle” he was at present a prisoner in the Bastille.Footnote 1 Rausset had traversed the Atlantic and extended the span of France's early modern empire, but by 1664 he had reversed course and now sat in Paris in what would later become that empire's most famous prison. In the 1740s, Jean Clavier worked as a bailiff in the Saint-Domingue community of Léogane, just south of Port-au-Prince, where he helped manage court proceedings.Footnote 2 Convicted of illegal trading, he was sentenced to prison but escaped, in one administrator's telling because the prison walls were falling down. The cases of Rausset and Clavier feature a familiar and ubiquitous punishment, imprisonment, against the specific backdrop of European colonization and empire building. Both began careers as state agents only to have the state turn against them. Both stories point to broader questions: how, specifically, did prisons play a role in the built environment of empire? Where did imprisonment fit within early modern logics of punishment? And how were such logics manifested in practice, in place?
I theorize that coercive strategies existed together within a “punitive matrix,” in which imprisonment operated as one among many methods of state control, alongside others such as galley labor, banishment, and corporal punishment. All of these methods were unevenly applied; all of these methods had gaps that could be pried open, sometimes literally. Many of those gaps stemmed from the material nature of each punishment. Prisons fell down. Galley ships leaked (they were often old leaky ships to begin with). Banishment required authorities and those sentenced to play a game of keep away (plenty of banished subjects sneaked back across borders).
This conceptualization accomplishes two goals. First, it keeps the material and spatial aspects of punishment central. States sometimes controlled subjects by keeping them close to centers of power in prisons, which often abutted court complexes or military forts or imperial capitals (like Rausset's stay in the Bastille). Or states could prioritize distance between disorderly subjects and civil society through banishment. Galleys were a compromise. Famous as “floating prisons,” they channeled incarcerated labor into imperial engines. Imperial subjects responded to these methods with their own strategies that countered or undermined state control. Some, like Clavier, ran away or escaped prison. They—and sometimes banished subjects—started new lives in new imperial outposts.
Second, a punitive matrix orientation contextualizes prisons amidst other common premodern solutions to crimes rather than overdetermining the eventual development of a modern carceral state. Ample scholarship from Foucault onward has probed the origins of prisons as a key technology for modern state control, but it overemphasizes the revolutionary era as a moment of invention.Footnote 3 The prison existed much earlier and, as this research demonstrates, gradually became a favored tool among many widely utilized alternatives. Other historians have explored early modern strategies such as exile, used most often to punish recalcitrant subjects when other strategies did not work.Footnote 4 This evidence points to a general pattern whereby these forms of control coexisted and even overlapped in an array of punitive options. Subjects could be sentenced first to imprisonment, then to exile, and sometimes even recalled from exile, with corporal punishments and fines possible at all stages. In slave societies such as Saint-Domingue, criminal authorities also constantly measured—and dictated—freedom and bondage as they meted out punishments.Footnote 5 Slippages ran across and between these forms of control at key pressure points as the large and growing literature on marronage has demonstrated for slavery specifically.Footnote 6
To begin to map these multiple dimensions in tandem, this article begins with evidence about prisons from Saint-Domingue in the first half of the eighteenth century, building out from Jean Clavier's Léogane in the 1740s. It then turns to the archival sources themselves to understand how colonial administrators described prisons in two types of documentation: legal records and building and town plans. These sources are themselves material records of prisons, but often occlude efforts to recover the past built environment because they tend to be prescriptive, whether for punishment or construction, rather than descriptive. Read together, this evidence highlights the limited power of prisons, and imprisonment, to control imperial subjects, while elucidating some of the pathways chosen by those subjects.
Along the way, this survey incorporates examples from other French colonies for context and comparison. All of these colonies reported to the same metropolitan center from which also came royal edicts. Some, like Martinique, lay within the same Caribbean circuits of trade and information. Until 1713, the regional government that oversaw Saint-Domingue's was based in Martinique. Others, like Ile de France in the Indian Ocean, were similar slave societies. Imperial planners designed prisons for Louisiana and Pondichéry from the same offices where those for Saint-Domingue originated. Colonial subjects also communicated with each other, whether formally via regional governments or informally via global trade routes and correspondence.Footnote 7
Prisons of Rubble
Jean Clavier, a bailiff in a Saint-Domingue admiralty court, had been convicted and imprisoned in the early 1740s on charges of illicit foreign commerce, an ubiquitous crime, and in 1742, he escaped prison with two of his suspected accomplices. One administrator blamed “these frequent evasions [as] the effect of the bad state of the prisons of the whole colony.” It was nearly impossible to keep prisoners inside.Footnote 8 The prisons were continually falling down. Prisoners often just stepped over the rubble to escape. Clavier danced back and forth over these physical and legal boundaries as he managed court proceedings in his role as bailiff, and then became a target of such court proceedings.
Bailiffs like Clavier acted as gatekeepers for court proceedings, preventing disruptions and maintaining the order of cases as plaintiffs, defendants, and witnesses took their turns in front of magistrates. It could become a lucrative position, too. A man named Baudu held the slightly higher-status huissier audiencier position, meaning that he was a bailiff who called court hearings to order in addition to managing court proceedings. He received a pension of 2,400 livres for service over just 2 years (1770 and 1771).Footnote 9
As an admiralty court bailiff, Clavier would have overseen maritime cases dealing with precisely the kind of crime that carried the longest prison sentences: trading with the enemy, the crime he committed himself. Foreign trade (as got Clavier into trouble) and, in general, communication with foreigners remained a dominant theme among crimes that warranted imprisonment. A 1680 edict in Martinique forbade the imprisonment of any resident (aucun habitant) with the exception of anyone who communicated with enemy forces. Authorities were required to write to the king about those whom they imprisoned, specifying why.Footnote 10 Martinicans were also forbidden to house foreigners under any pretext, facing imprisonment in the nearest location if they failed to comply.Footnote 11
It is hard to tell exactly where Clavier and his associates ran away from, in part because the prison is not mapped well. Maps drawn in 1742 (the same year as Clavier's case) and 1785 do not name any buildings in the town, despite delineating blocks in great detail and a hospital just outside the perimeter (Figure 1).Footnote 12 A main road connected the town to the seaside fort. Most noticeable are the polygonal plantations marked all around the town, with only a band of undeveloped land around the town and road. With small tributaries and no elevations marked, it is easy to imagine the local prison amidst a flat landscape surrounded by few obstacles for enterprising convicts who dared to leave.
Detail, “Plan de la Ville et des environs de Léogane dans l'Isle Saint Domingue.” John Carter Brown Library Map Collection E791 P793r \3-SIZE. (Paris: Sr. Phelipeau, ingenieur geographe, 1785).
A survey of Saint-Domingue town plans reveals only occasional marking of prisons, without a clear pattern, such as larger towns having larger prisons.Footnote 13 The adjacent town of Petit Goave, a little larger than Léogane and depicted in 1742 as the seat of provincial government, housed a prison on the same block as its palais (which held courts and administrator offices).Footnote 14 In a 1731 plan of Cap, the northern capital and increasingly the gateway to the most prosperous planting region, only one building appears marked as a cachot, apparently as an isolation space separate from a prison, which tended to comprise a complex of different spaces for incarceration. It was surrounded by military structures, such as the barracks and a small battery, that would have forestalled potential escape (Figure 2).Footnote 15 Prisons were sometimes, but not always, constructed as separate buildings, so they do not always appear on city plans. Standalone prisons, as described by a 1774 plan for a civil and criminal prison complex in Cap, could contain separate holding areas for men and women, free and enslaved, in addition to a section devoted to the cachots.Footnote 16 Sometimes they were incorporated into government buildings like law courts, where defendants could be held before trial or after conviction.Footnote 17 Sources consulted for this project do not mention barracoons, which were barracks-like structures that held enslaved Africans in port awaiting transportation along the Middle Passage (or tributary routes), similar to port-side prisons that held individuals awaiting transportation.Footnote 18 But some do mention bagnes, penitentiary buildings, that closely resembled and overlapped in usage with prisons.Footnote 19
Detail of Cap with prison/dungeon (cachot) circled, “Plan de la Ville du Cap, a la Côte Septentrional de Saint Domingue.” John Carter Brown Library Map Collection E730 C478h. (Paris: Hippolyte-Louis Guerin, 1731).
Prison Plans
Prisons were built—and unbuilt—as part of the material fortification of that empire, which included stone and wood defense works, powder mills, and fortresses that most official maps emphasized.Footnote 20 Incarcerated and indentured laborers built these structures as did many enslaved people. Many of these plans were drawn up by engineers who worked for the division of Bridges and Roads (Ponts et Chaussées) in collaboration with the Marine, the Navy Ministry that administered all of France's overseas colonies. In each of the examples described below the evidence does not specify which were built and to what extent plans were followed. But these hand-drawn designs, sometimes water-colored in detail, offer clues to the physical appearance of structures taken for granted in legal documents assigning prison sentences.
Despite a lack of visual evidence, prisons figured in the earliest decades of colonial settlement, such as a royal missive that detailed a recent treaty with the English and enumerated local judicial powers to be employed by the island's court (conseil supérieur), including imprisonment. This appeared early enough that the paragraph following prison instructions discusses war with the indigenous Caribs.Footnote 21 Maps, as described above, infrequently specified the location of prisons for a potential variety of reasons. They were less important to military strategy than forts and roads. They were possibly—as Clavier's case implies—not always built to last (even when made of stone). And they seem to have been portable in some cases, at least in the sense that administrators could house prisoners in court complexes or near military encampments as needed.
Prison plans from the Marine archives reveal at least a series of proposed prisons in the 1730s, much earlier than one might suspect from the carceral state literature, running well into the nineteenth century. Empire-builders craved new territory; they required new human-scaled prison cells within those territories. A 1731 Louisiana prison showed brick and masonry construction with two prison cells (cachots) separated from the outside by thick walls on three sides. Internal doors opened toward a series of rooms for guards and soldiers, with a porch on two sides of the guard rooms.Footnote 22 An example from 90 years later (1821), from France's Indian Ocean colony Pondichéry, outlines eight cells (cachots) arranged symmetrically along a central courtyard, aligned with the (former) tribunal courtroom.Footnote 23 These buildings differed in size and had different features such as a porch, but both had designated cells for solitary confinement called cachots. These small windowless rooms sometimes occupied subterranean space—more akin to a dungeon—while others were ordinary prison cells. These could also be found on plantations, where masters directed punishments.Footnote 24
Other visual evidence draws attention to the dual role of colonial prisons in serving military and (civilian) legal matters. A British drawing from 1 year after the Louisiana plan, 1732 (Figure 3), presumably created for military intelligence purposes, depicts a prison in Martinique. The low, small building sits right along the shoreline of the island's main, militarized port Fort-Royal at the end of a long palisade.Footnote 25 Such a location created easy access for ships to offload prisoners or for local courts to house subjects sentenced to exile, galley labor, or military service. For prisoners meant to stay on the island, military surveillance would have undermined opportunities to escape by sea, in contrast to the empty environs that seem to have enabled Clavier's escape.
Detail, I. Kip, The Prospect of Fort Royal of Martinico, as It Sheweth from the Entry of the Harbour Call'd Cul de Sac Royal. 1732 Engraving. Archive of Early American Images. John Carter Brown Library.
Plans sometimes appear completely de-contextualized, unmoored from any material setting. A 1782 design for “civil buildings” to be constructed within the fort complex at Cayenne, Guyane, marks three buildings: a prison, powder magazine, and guard house. Precise black lines, colored in varying shades of pink, mark the prison's location parallel to the page with the other two buildings arrayed along an imaginary curved line. But there is no compass rose or other clue to the buildings’ orientation within the fort.Footnote 26
Prisons on Paper
Legal documents similarly occlude the physical manifestations of empire, but in ways that contrast with drawings of towns and buildings. Despite constant references to prisons, French colonial archival sources that document law-making institutions and courts convey remarkably little detail about the prisons themselves, whether their social or physical inner workings. And they rarely discuss the collapse of prisons, as referenced in Clavier's case. Instead, these sources documented and served as key material products for a globe-spanning imperial legal regime while neglecting the physical spaces that made it possible.
Legal codes and commentaries affirmed and then reiterated (sometimes with modifications) what would take place inside prisons, who would live there, and for how long. They began as letters sent by the king and his ministers or by colonial administrators. Clerks compiled them into manuscript reference volumes. Enterprising magistrates and planters—such as the Martinican turned Saint-Dominguan Moreau de Saint-Méry—turned them into printed volumes, distributing them widely.Footnote 27 Important laws and court decisions appeared on broadsides posted around colonial towns by town criers, whose role overlapped with the work done by bailiffs within courtrooms.Footnote 28 These sources contrast with the imperial plans, which imagined as-yet-unbuilt fortifications and prisons, while being somewhat ephemeral documents themselves.
Within compilations of laws, local and imperial, a few patterns emerge in the ways that imprisonment applied to different subjects differently, allowing a punitive matrix to come into view again. First, despite military and civilian personnel living side by side, the distinction between those subjects, and between military and civilian authorities, remained important. The king and his ministers worried about the right subjects living within and outside prison walls, a concern that Clavier would have shared. A 1679 royal order forbade local governors from imprisoning local residents (habitants) without approval of supervisory bodies such as the local court or the regional governor.Footnote 29 A 1681 letter from the king to the West Indian governor-general specified a ban on the incarceration of local inhabitants (habitans) in military prisons, emphasizing the role of local courts (ressort de la justice ordinaire) in presiding over civilian cases.Footnote 30 At least to officials’ eyes in France, the intermingling of these subjects in criminal cases indicated mismanagement (perhaps as reported initially by the residents themselves in an unspecified complaint), even if the evidence from maps points to blurred boundaries between the two groups.Footnote 31
Second, as scholars know well, the escalation of punishments and prison terms operated on very different scales for enslaved and free subjects in local ordinances and royal edicts that supplemented the Code Noir. A 1718 Martinique ordinance banning illegal fishing specified 3 months’ imprisonment (plus a 100 livre fine) for a second offense for free people, 1 month's imprisonment (and other punishments) for a first offense by an enslaved person.Footnote 32
Finally, legal records affirm imprisonment—confinement—as a counter to subversion of colonial space by enslaved people through marronage.Footnote 33 A 1707 Cap (Saint-Domingue) conseil ruling specified that runaways should be kept in prison for at least 1 month, without recourse to their masters, and with the possibility of their public sale at the end of the term.Footnote 34 A 1733 ordinance in Martinique prohibited jailers from letting out enslaved people who were imprisoned for marronage.Footnote 35 In 1747, regulations addressed concerns that passenger canoes were “daily filled with runaway slaves” (journellement remplis des nègres marrons).Footnote 36 Although planters and colonial authorities fretted constantly about marronage and other forms of slave resistance, Clavier's case demonstrates that escape could take many forms, and that officials tasked with containing prisoners in courtrooms might themselves abscond.
Conclusion
A holistic investigation of prisons in early modern Saint-Domingue and other French colonies reveals material forms that did not always match paper designs, legal or architectural. Material forms were always being broken apart, renovated, and reimagined. The prison plans signal such constant transformations in the form of administrative, imperial schemes whether ever constructed or not. And the British drawing helps scholars think about how such material forms of law and punishment would have been read by visitors, including imperial competitors.
Turning to the most obvious feature of these colonial societies, slavery, reveals, too, perhaps why violence and policing were ubiquitous, but always in the process of expansion through yet more layers of surveillance and control. That violence and power were aimed toward the free white population, not just the enslaved population. Even bailiffs needed to be managed. Multiple overlapping criminal justice regimes developed together, each oriented toward different colonial features—trade, slavery, the military—but they were all shot through with holes, some jurisdictional, some physical.
Acknowledgments
The author thanks Bianca Premo, Kalyani Ramnath, Sanne Ravensbergen, and Yanna Yannakakis for their ongoing fruitful suggestions. She is also grateful to Gautham Rao and an anonymous LHR reviewer for thoughtful engagement and encouragement.
