Modern legal experiences in Europe and the United States immerse their participants and observers in an ocean of paper. Most legal acts involve paper and signatures, and in litigation, from the written summons through written evidence, written verdicts, and a written transcript of the trial, paper is ubiquitous and unremarkable – unless, in a moment of drama, handwriting experts need to be called in or the record needs to be read back. Writing on paper is a tool and a technology, a neutral facilitator of the procedural and probative goals of the law and of the courts. By contrast, writing in Roman legal acts was not consistently ubiquitous, and Roman trials incorporated writing far less until the late-antique period. Before then, therefore, different questions about writing used in Roman legal contexts should dominate the discussion. What physical forms did such writing take? How were different types of written document valued when they were used? And what could these forms of writing have meant to those who used them? For centuries these were not, for Romans, legal questions at all. Instead, the legal documents of Roman citizens were, through the classical period, generated with the help of all-purpose scribes, not official notaries; their ultimate legal weight was determined not by any ‘law of evidence’ but by their impact in court, in which traditional assumptions about their authority as well as rhetorical deftness in circumventing those assumptions played a role. It is a modern assumption that writing is functional, and a similarly modern verdict that legal documents, even Roman ones, almost always serve only as proof.¹ The Romans, their legal world imbued from an early date with religiosity and scrupulous ritual, saw writing both in documents and in procedure as powerful in different and (to the modern reader) unexpected ways.²

Legal documents of Roman citizens, written in Latin, survive only by chance, and as a consequence the more than 1,000 preserved documents tell only a stop-motion story of developments well underway by the time they can be studied. No documents from the Republican era survive, although we know from Cicero that they existed;³ from the imperial period there are substantial collections from Pompeii and Herculaneum, Britain, Germany, Vindonissa in Switzerland, Dacia (Romania north of the Danube), and Egypt, with a smattering coming from other locations.⁴ Physical circumstances had to be favourable, since these collections make clear that wood was the preferred medium for such documents, and the survival of wood depends on very specific climactic conditions. Indeed, it appears that Roman legal documents were almost always written on wood-and-wax tablets; even in the late Empire, when the formal requirements of (unspecified) materials and language were officially relaxed, wooden tablets were still used, as a cache from fifth-century North Africa demonstrates: the form was chosen even when not apparently required.⁵ This legal use of wooden tablets was therefore significant and special, and distinguished Roman legal documents from their Greek contemporaries, which were of papyrus and on scrolls.

The physical form of this wooden tablet, and the treatment of the text on a wooden tablet, changed over time. The two earliest known examples show two different forms. The simpler form belongs to the second oldest,⁶ which seems to have had merely a single copy of its text written horizontally, parallel to the long side, into the wax on the interior faces of two tablets hinged together. These were then closed face-to-face as a form of protection, with a string wound around both together. By contrast, the very earliest so far found, from 8 BC and published only very recently,⁷ already had its string fixed in place by seals, with the names of the eight sealers (three partially preserved) written next to their seals. A copy of the interior text was (simultaneously with this development) written on the exterior of the tablet, in ink and parallel to the short sides, making this what is called a doubled diptych. To protect the seals better, in the next phase a wide groove called a sulcus was built into an exterior side of the second tablet, and the seals were placed in this channel over the string. Gradually, for some documents a third tablet was then added to the two that were sealed shut. The third carried the exterior copy (instead of, or sometimes in addition to, the text being copied on the exterior of the first two tablets)⁸ while also, when shut, giving added protection to the seals in the sulcus on the back of the second tablet.⁹ This is a triptych. This general format – two or three tablets, two copies (interior and exterior) of the text, the seals in a sulcus – is the one most commonly found (although not always or even usually with all tablets intact). Polyptychs, with more than three tablets, also existed, and were used for especially lengthy documents like wills. In AD 61 a senatusconsultum Neronianum¹⁰ required that tablets (of the will of a Roman citizen) be pierced and that the string be threaded three times through the holes, thereafter to be sealed in place by seals in a sulcus. Soon most documents on tablets followed this practice (a late-antique text actually claimed that the senatusconsultum applied to all legal documents).¹¹ It is thus deducible that, at and after this point in time, four physical aspects of a legal document had come to be thought important: the use of wooden tablets; the existence and protection of a written original text (the sealed interior version); an accessible copy of the text (the exterior version); and the attestation of presence and weight of social standing provided by the sealers whose names were written next to their seals. Sealers were listed in order of social prominence and were lending their authority and their fides (trustworthiness) to the document so sealed.¹² For Pliny the Younger, the performance of this vital social task of sealing (especially for wills) was one of the ways he spent his time when in the city of Rome.¹³

As these changes in the physical format of the tablet suggest, this is a story of increasing protection of both interior text and of seals, but also of a development caught at a particular moment in time and with an earlier history all but invisible to us. The axis provided by the types of legal acts found on first-century tablets similarly suggests change and development caught in mid-stride. Those tablets specifically from the area of Campania (before AD 79) permit some assessment of the relationship between physical form, date, and type of act. There are, on the one hand, tablets of the older formal, ceremonial acts of Roman law, especially those based on the acts of mancipation and (as I have argued) stipulation¹⁴ and those related to the formalities of Roman legal procedure, all only accessible to Roman citizens: these are written in the third person in careful and often archaic legal language, have between seven and eleven sealers, and consistently use the older diptych form through the middle of the first century AD. But then there are also tablets of informal or bona fides (‘good-faith’) acts: these are written in the first person in freer if also mostly formulaic language (and often called chirographs – ‘hand-writtens’), have between three and five sealers (including the author of the act himself, who sometimes seals twice), and consistently use the triptych form as early as AD 35.¹⁵ Formality and bona fides are, in a sense, two different tracks in Roman private law, and the legal acts based on them differ not only in who can use them, but also in the origins of their powers (formal acts from the efficacy of their correctly performed ritual, informal acts through enforcement by the praetor). The different rate of adoption of the triptych form for the two categories of act, along with the differing number of sealers, suggest that tablet-documents and their sealers initially played different roles depending on the type of act, even though they were on track to become much more similar by the 70s AD. Some documents that combined individual legal acts of the two different types (like a formal mancipation and an informal pact,¹⁶ or a formal acceptilatio [release from obligation] and an informal chirograph¹⁷) suggest a similar trajectory towards amalgamation of traditions formerly (and in the law) treated separately. For neither type of act, formal or informal, was the prevention of forgery the first or only raison d’être for the complex physical form, since all tablets would otherwise have been constructed and sealed in the same way from the very beginning. The senatusconsultum of AD 61 is the first and last official indication before the late-antique period of an interest in the techniques of preventing forgery and can help to explain why many tablet-documents came to look much more similar after that date; before that date, however, tablets in the two traditions were different, and two hundred years before that it is likely that tablets for bona fides acts did not exist at all, since ‘chirograph’ implies importation from the Greek tradition and bona fides acts themselves were recognized by the praetor only in the late second century BC.¹⁸

The trajectory of development in physical form and content visible even in what survives therefore suggests that both diptychs and formal acts – and formal acts on diptychs – were older; that the use of a wooden tablet was sufficiently characteristic and weighty as a ‘Roman legal document’ that a newer type of act would adopt it; that sealers brought social weight to both types of act but had different primary functions in sealing; and that ‘good-faith’ acts and their physical format initially emphasized (and protected) the fides of author and sealers to a greater extent. It would seem, therefore, that an understanding of the role such a wooden document played is rooted in a time earlier than that of the surviving documents, and in the formal acts with their performative rituals and their attesting witnesses. The complexities of form and sealing suggest, too, that the original role of wooden tablets in formal acts was more than that of mere proof. So it should be no surprise that when, in the only apparently generalizing statement about written documents from the classical jurists,¹⁹ Gaius said that ‘the purpose of writing [was] to prove the transaction more easily’,²⁰ he also specifically limited the scope of his observation to two of the ‘consensual’ informal acts – mortgage (hypotheca, an informal good-faith contract) and marriage – two of the later acts that migrated on to tablets to share in their value. Indeed, ‘writing’ in legal acts was never denigrated as such by the classical jurists, who (this quotation aside) paid no generalized attention to it at all.²¹

Even if not intended to be only proof, wooden documents were also very useful as proof, and their exceptional contribution to a court case was especially acknowledged by orators. For Cicero and Quintilian, tablet-documents were a wonderful kind of super-proof: they were happy to wield them when such tablets supported the case they were arguing and recognized the need for feats of special rhetorical agility when they did not. Tabulae had the special and potent quality of auctoritas (‘authority’), said Cicero,²² and were ‘difficult’ to get around;²³ for Quintilian, arguing against them required ‘the greatest power of eloquence’.²⁴ Witnesses were very important in court too, but witness-testimony written on wooden tabulae seems to have combined an excellent type of proof and the best form of proof into one, transforming testimony into a contribution that, like a legal document on a tabula, could be challenged only with great difficulty.²⁵ To a Cicero or a Quintilian, there was some special quality about wooden tablets, some authority, that was unmistakable and virtually unassailable, and this special quality must also have helped to perpetuate their use as the form to be used for legal documents through the imperial centuries. Doubling the text protected the writing and sealers added their own weight, but it was writing on wood that fixed the act or the testimony and made it authoritative. It may, indeed, have been the very existence of a tablet-document that was most important, since even when adduced in court there is no one clear example of their actually being opened: they could do their work without their strings being cut.²⁶

This appreciation of the wooden tablet’s power by orators who wielded or faced them in court is reinforced and in part explained by the wider cultural understanding of such forms. Authoritative finality was also thought to characterize, for example, wooden account-tabulae, tablets announcing repaid vows, tablets of the census, the tabulae of the priests recording religiously significant events of the year, the tablets of the praetor’s edict, and tablets used for prayers read out by magistrates.²⁷ The special rhythmic and formulaic language of legal tablets finds parallels in the language of these other tabulae, again pointing backwards to formulations perhaps as old as the fourth- or third-century Republic, after which the use of such tablets, often as part of a larger ritual, contributed to the creation of social and political order and an appropriate relationship between Romans and their gods.²⁸ The quality of being embedded in larger acts that had to be performed correctly is one of the sources of this authoritative finality: a tablet was a crucial element of such an act – for example, the taking of the census²⁹ – that was not complete until all the writing was done and all the rituals had been performed. Wooden legal tablets were similarly embedded in the old formal acts of Roman law, as Gaius’s description in the second century AD of the ceremony of bringing a mancipatory will into existence makes clear.³⁰ Specific words had to be spoken in a certain order, in front of witnesses; gestures (striking a scale with a piece of bronze, and handing over the bronze) had to be made; the tablets had to be held in the hand of the testator, who then had to speak a specific formula. The Roman-citizen witnesses (testes) were there to judge the correctness of this ritual performance, crucial to its legal validity, and this performance included the tablets themselves, to which they affixed their seals.³¹ Wooden tablets of such acts were generated as part of the act itself, were necessary for its efficacy, and authoritatively embodied and completed it.

These wooden documents, with their acknowledged intrinsic powers buttressed by the social weight of the men who sealed them, were recognized as peculiarly and characteristically Roman by the peoples whom they ruled. Roman citizens travelled with their own wooden documents or drew them up in the far-flung places where they found themselves: hence deposits not just from Campania, but also from the provinces, and especially (although not exclusively) from army camps. Terms were also transliterated, like τάβλα for tabulae (in, for example, a new inscription preserving testamentary dispositions from Cappadocia³²). In Dacia, many of the surviving wooden tablets may have been employed by non-Roman citizens (the status of the participants in these legal acts is disputed and there are anomalies in the execution of the acts).³³ In the eastern Empire, the format of these documents was imitated by non-Romans, producing the (so-called) papyrus double-document. In this, the text of the act was written across the grain of the papyrus at the top of the document, with a second copy written beneath it; the top version was rolled over and sewn shut; and the names of the witnesses were written on the back of the papyrus, next to the knots from the sewing. Provincials who were not Roman citizens could not technically use formal-act legal forms, but could imitate what they thought the Romans valued in the execution of a document: inner copy, outer copy, protection, attestation, and witnessing. Such double-documents are not all that common and seem to be used especially for sales of property, such as slaves, that might be moving from one province to another, or for documents aimed at circumstances in which one could (one imagined) meet up with a Roman official. Such, for example, seems to have been the point of a papyrus double-document of honourable discharge for sailors-turned-legionaries heading for Egypt,³⁴ as well as the guiding assumption behind much of the dossier of documents taken by a Jewish woman named Babatha into the Judaean Desert at the time of the Bar-Kochba Rebellion: she had not only 23 double-documents, but also three copies of an outline of a Roman formula of the actio tutelae, such as a magistrate would issue to a Roman judge to specify the issues to be determined. Babatha was in litigation with the guardians of her son after the death of his father, but was also in some sort of family tangle with the clearly Roman Julia Crispina, and was preparing herself (it would seem) to come off well in an arena where Roman expectations might well reign supreme.³⁵ Roman documentary habits were a Roman pattern that had an impact on the understandings and expectations of provincials, and thus also on legal life in the provinces of the Empire.

In many Roman provinces, especially those in the East, the substantive law that had existed before the Romans remained in place, and so too did the associated documentary habits, especially well-attested in Egypt, but attested also in the epigraphy of Greece and Asia Minor and the papyrus and parchment finds from Mesopotamia. Such papyrus documents were valuable and useful for their protagonists, and accepted in local courts, but with no sense of the special value and weight that Romans attached to wooden tablets in their own courts. Papyrus legal documents, coming as they did from a non-Roman tradition and hardly influenced by Roman substantive law before AD 212,³⁶ seem to have carried little weight in Roman courts, but over time other documents on papyrus, like personal letters, gained in value – depending on who had written them. Cicero was fairly contemptuous about litterae (‘letters’ – so ephemeral!) unless they clearly supported his case, but Quintilian was rather more circumspect: holograph letters came to be seen as reflecting the fides (good or bad) of the author, and everyone was carefully appreciative and admiring where letters of the emperors (and eventually imperial officials) were concerned.³⁷ This changing attitude towards personal documents on papyrus was not a negative comment on the authoritative value of wooden documents but an argument made in addition to it, and it represented a potential expansion of the arsenal of courtroom weaponry. As the deployment of evidence in Apuleius’s defence of himself on a charge of magic before the provincial governor in the second century shows, letters and such, depending on their source, could be valuable, but only as a supplement to the – already acknowledged – preponderant weight of tablets, which he used to make his final and most important points.³⁸

Long before late antiquity, then, and before Roman jurists started to interest themselves seriously in matters of documentation, proof, and whether writing was a crucial component of a legal act or not, writing on a wooden tablet had established itself as being of superlative value and importance. It was a significant and necessary component of a formal legal act, had auctoritas, could record first-person legal acts as well as personal testimony and turn them into established facts, and was recognized as a supreme form of proof in a courtroom setting. Over time, the physical form of the wooden tablet could (literally) expand to incorporate and also convey the fides of those who sealed it; and in its perfected form in the second century AD it was, as Apuleius’s case shows, nearly invincible in demonstrating what had happened, what was true, and which people should not be offended by impertinent challenges to what they had attested and sealed shut. Other types of writing – at least those on papyrus – offered mild competition to writing on wood but only because they too could demonstrate the fides and standing of the document’s author: great, in the case of the emperor; lesser in the case of everyone else. So writing was important, and increasingly so over time, but it was writing of a certain sort, in documents constructed in a certain way and of a certain shape, that was for centuries most important, for reasons that went far back into Rome’s religious and legal past.

Wooden tablets were also used in Roman legal procedure. Their deployment initially parallels that of the tablets used as templates for prayers, the tablet fixing a set text for proper reading aloud when extreme verbal correctness was crucial (a necessary obsession of late-Republican jurisconsults, who were mocked for it by Cicero³⁹). Under the formulary procedural system (the second of the three classical systems),⁴⁰ most likely the formula (given by the praetor to guide the judge) – in carefully accurate language – and the accusation (nomen deferre), in a criminal case, were written on tablets, as were the later inscriptio and libellus (of accusation).⁴¹The Campanian finds reveal many more types of procedural tablets, including vadimonia (promises to appear), attestations that one had appeared (tabulae sistendi), the setting of days for a hearing, the formal passing to the giving of the judgment (intertium), and the judgment itself; there also survive interrogations, declarations, and the performing of oaths, all of which, like witness-testimony, seized and finalized, in an authoritative way, otherwise transitory experiences.⁴² Many of these documents came as a surprise to scholars, since most other information about legal procedure had stressed its oral qualities before the late-antique period.⁴³ These wooden documents are not a way of making a record of an entire trial but instead fix important but individual contributions to the procedure of a trial and mark successive stages of a trial as they were completed. But the concept of the tabula for a perspective of the entire trial is important too, for Roman magistrates kept some records of their actions in office, records called publicae tabulae, and when these actions included hearing court cases (as city-magistrates or governors), information about the trial (plaintiff and defendant, advocates for and against, verdict) was entered into them, and their quality thought to reflect on the magistrate’s character and probity. This initially tight focus on the magistrate’s activities gradually expanded, after AD 284, to include more and more information about the trial, and more and more verbatim information from such trials came to be used in subsequent trials. The ‘tablet’ here was fixing and adding authority to the record of an event, and – as a metaphor for the magistrate’s entire archive, when papyrus rather than wood was later used – became another locus through which individual legal acts could achieve finality and validity. ‘Reading into the publicae tabulae’ or ‘entering into the acta’ (as this process was also known) made written legal acts presumptively and authoritatively true.

Only in the later second century AD and after did legal writers like Gaius – jurists and the trained staffs of the emperors – start to construct rules for a clear system of proof, and in handling questions and problems try to assess the role of writing, especially in the formal legal acts in which writing had for centuries been embedded. The way they tackled problems, which often arose because some formal ceremonial element had been omitted from the performance of these acts, shows that they recognized writing as one of the formalities of an act, along with gestures and formulaic language: they explored where the essence of a multi-component formal act might lie, sometimes alighting on an abstract quality (such as obligatio verbis, ‘obligation in words’ or voluntas, ‘intent’); they made compensatory arguments when one element had been mistakenly omitted or was flawed in execution, thus acknowledging that elements like writing and speech were complementary rather than primary and secondary (so Ulpian could say ‘more was announced and less written’ when there was a problem with a will); or, in (especially) the fifth and sixth centuries, they (finally) deemed formal elements – physical materials, special words in a set order, gestures – unnecessary, and identified writing as the all-encompassing ceremonial quality that made an act valid.⁴⁴Justinian was notably thorough in his own legislation in imposing common requirements on written documents of all sorts, while also systematizing and granting particular strength to the ‘public document’ drawn up with the assistance of a public notary.⁴⁵ Changes of this sort reflect the gathering strength of the emperor as both the actual and the symbolic font of the Roman law of the Empire,⁴⁶ but also reflect the long traditions and beliefs about the embedded quality of writing that inspired respect and interest in petitioners, jurists, and the emperor’s legal advisors and writers.

Because jurists and emperors weighed in on these issues so late, their opinions are where this story ends rather than where it begins. What the written wooden documents of Roman law meant to those who used them, and then to the orators who confirmed but also grappled with their weight and importance in court, was established long before the law’s intellectuals turned their razor-sharp gaze on them. Because of their close association with the emperor, Severan and late-antique jurists could write in his name and with his powers and gradually adjust what the role of writing was to be; even so, traces of what writing once meant are clearly perceptible in the answers they give and the opinions they propose. Physical form, embedded writing, proper ceremonial vouched for by witnesses, and sealing by the same imparted an antique strength to wooden documents that was appreciated for centuries.