Numărul 4 / 2010
IN MEMORIAM TUDOR DRĂGANU
ORALITY IN THE LAW OF CONTRACT: MANIFESTATION AND PROOF OF INTENT IN ANCIENT AFRICAN AND ROMAN SOCIETIES
Gardiol J. van Niekerk*
Abstract: There are many intersections in ancient Roman and African law and a superficial analysis of orality in the Roman law of contract provides insights that one could usefully employ in understanding the African law of contract. In Roman law, contracts could be constituted by words only but to incur liability the words had to be concretised by an "outward ritual". In African law purely verbal agreements did not give rise to legal liability. Transfer of property, related or unrelated to performance in terms of the contract gave rise to liability and in that sense contracts in African law differed from real contracts in Roman law. The prevalence and endurance of the stipulatio in early Roman law have been ascribed to the importance of the Roman virtue of fides. It was the non-specialised nature of African law rather than the lack of fides that explains why purely verbal contracts were unknown in African law.
Keywords: ancient African law; customary law; ancient Roman law; sources of law; preliterate; jurisprudential framework; contract; orality; language; writing; verbal contracts; stipulatio; concretisation of words; outward ritual; formalism; fides; non-specialised nature; manifestation and communication of intent; ritualistic declarations
Before commencing with this discussion it is important to consider the sources available on ancient African law. By tradition, ancient African culture may be characterised as preliterate. That means that those early societies had no written language. In consequence, no ancient written sources exist authored by indigenous Africans themselves. Non-legal materials are the earliest written sources of knowledge of pre-contact or ancient African law. These include texts of early European travellers dated from the fifteenth century onwards, and anthropological and ethnological writings on African culture, which started appearing in the mid-nineteenth century.
It is not an easy task to extricate information on African customary law from these sources since their main focus was not substantive law. Importantly, though, these works contain information on the traditional African law which has not been adulterated by the interpretations of jurists with preconceived European ideas. Moreover, they prove that as early as at least the fifteenth century, southern African societies were socially and politically well-organised and had established legal orders with reasoned laws and institutions, albeit different to the familiar European model.
Other sources of law are colonial reports of commissions of enquiry and parliamentary committees. Although there was initially little interest in the laws and institutions of the indigenous populations in Southern Africa, the colonial administrators soon realised that African customary law was there to stay and that they needed formal instruments regulating the application of that law. In addition they had insufficient knowledge of the substantive law. From the 1830s onwards, then, reports were compiled on African customary law and the regulation of its application. These yield more information on law since some of the reports deal exclusively with the substantive law. The first book in South Africa on African customary law appeared in 1858 but it took more than half a century before further works on that law were published. A Code of Zulu was promulgated in 1878.
Scientific analyses of African law are generally moulded in what is accepted as universally, or rather "European" intelligible legal language. In view of the fact that there are superficial likenesses in the legal phenomena of most legal systems in the primitive and the early stages of their development, the Roman jurisprudential framework, which is the most widely known, is generally preferred to explain African law. This analytical method unfortunately often resulted in the imposition of Roman legal rules and fundamental postulates on the traditional African law and consequently the attribution of dubious characteristic features to that law. Further, as a result of the reliance on a "non-African" framework to explain and interpret African customary law its continued existence came under threat.
In this article the goal is to rethink existing interpretations of selected aspects of ancient African customary law of contract by way of a comparative analysis with ancient Roman law and with reference to some of the earliest sources of African law. One of the shared characteristic features in these two ancient societies is a general emphasis on orality which was manifest in legal, political and economic life.
The pre-eminence of orality in African culture is understandable, given its preliterate tradition. In Roman law, though, this feature is rather unexpected, bearing in mind that it was from early on a literate culture. It is well-known that the Twelve Tables were inscribed on tablets in the fifth century BC, but there are conflicting views on when exactly writing was first introduced in private legal acts. By the late Republic, though, legal documentation was firmly established in legal practice.
But be that as it may, what is important for this analysis is that purely verbal contracts existed. The focus here will be limited to the best-known verbal contract in Roman law, the stipulatio, and the way in which the intention of the parties was communicated.
2 Words and intent
The relationship between the objective verbal utterances and the subjective intention of the parties to verbal contracts and the extent to which the Roman-law principles may cast a light on ancient African contracts will be considered first.
2.1 Roman law
In Roman law, form and the spoken word were inextricably linked. In order to have legal significance attached to a verbal contract, the words used to reach agreement had to be in a specific form and liability was incurred only if the parties conformed to the necessary ritualistic formal requirements.
Until the post-classical era, Romans did not distinguish between an external and internal component in the verbal contract, that is, between the external declarations (verba) of the parties and their internal intentions (voluntas).
The formal declarations were not intended as communication to the outside world and did not serve as publication of the real intention of the parties to outsiders, but manifested intent only between the parties themselves. Watson accordingly states that the ritualistic declarations were merely to show to the parties themselves "that they intended to make a contract". There was thus not necessarily a correlation between the formal words uttered and the subjective will of the parties and the words could thus neither be regarded as a manifestation nor provided proof of their intent.
It appears that it was rather the formality, the external ritual, which served as judicial proof that the contract had been concluded. However, this formality did not exactly function as evidence of the real intention of the parties, the object of the obligation, or whether one or both of the parties were obligated to perform in future. The law gave consequence to and enforced the actual wording of the stipulatio without looking into the surrounding circumstances of the promise or undertaking, and did so even if the words did not reflect the intention of the promissor and had no connection with actual reality.
The introduction of the exceptio doli and exceptio metus causa by the end of the Republic enabled the judge to start taking cognisance, to a limited extent, of the surrounding circumstances in which the contract had been concluded and so to contextualise it. That made the reason for its conclusion and the intention of the parties more relevant. Nevertheless, it still did not change the fact that as a rule the law still gave consequence to the formal words rather than the will of the parties. It was only in post-classical law that the emphasis shifted to the protection of the parties and that formalities were aimed at securing proof in case of litigation.
2.2 African law
While it is true that in accordance with Africa's oral tradition, agreements were always entered into verbally and that, unlike Roman law, the words did not have to be moulded in a specific form, words alone were nevertheless not sufficient to create liability or per se to convey the intention of the parties. This is illustrated by the maxims that "yesterday's word does not slaughter an ox" or "nobody buys the footprints of a bullock".
The parties' verbal communication had to be transformed into a concrete experience which occurred through the exchange of gifts, or performance in terms of the contract, or the physical pointing out of the object of the contract together with a specific verbal description.
The parties' intention was accordingly manifested in the words together with some physical activity. Confirmation of this may be found in the description of the cattle trade with the Khoi, the original inhabitants of the Cape of Good Hope by Grevenbroek, a seventeenth-century traveller. He observed that they did not consider themselves contractually bound unless an exchange of gifts had taken place.
In contrast to Roman law, agreement was reached by a protracted process of deliberation between the parties and between members of their respective families. Further, since a special etiquette was attached to all social and legal relationships, there were specific conventions as regards behaviour and taboo that had to be upheld in order to create an obligation. The various language taboos illustrate the close affinity between language, magic and religion in African culture. Cattle, for example, were regarded as an important form of legal tender and a complex cattle terminology existed. Terms referring to cattle differed depending on whether they were used in a legal, religious or kinship context. Animals were very specifically identified in transactions, using established, refined and complex linguistic colour-pattern terminology as well as personal names.
Also kinship terminology was complex and differed from one ethnic group to another. Among the Nguni tribes, for example, parents-in-law and children-in-law had to avoid using words phonetically resembling each other's names.
3 Concretisation of words
Although the verbal agreements had to be concretised in both Roman and African law, the way in which this occurred differed.
3.1 Roman law
The emphasis on orality in Roman law did not reduce the significance of formalism. In fact, the most distinguishable feature of early Roman law was what Schulz referred to as "actional formalism", meaning that all legal acts had a specific form. Thus, in the stipulatio,not the actual words but the form or external ritual created the legal bond; liability did not flow from the agreement of the parties but rather from the exchange of the prescribed formal phrases.
The requirements for the "outward ritual" of the stipulatio were that the communication had to be in the form of oral  questions and answers, inter praesentes;that there had to be exact correspondence between question and answer (the verb in the question and answer had to be the same); and that the answer had to follow immediately after the question (unitas actus).
It was ritualistic formalism, then, not writing, that served as the concretisation of the spoken word in the stipulatio. The predominant opinion among scholars is that although in classical law it was usual to reduce the oral agreement to writing, this was not a requirement for the validity of a contract but was purely for evidentiary purposes. During the pre-classical period of Roman law, written documentation was not regarded as a legal formality. The most obvious reasons for this phenomenon were that the physical presence of the parties was regarded as a guarantee against misunderstanding and enhanced clarity; and general illiteracy, the available methods of writing and time restraints. It was only in post-classical law that the oral stipulatio was superseded by a written contract.
However, there are also other views. Meyer, for instance, has shown that writing on tabulae was part of the stipulatio from very early on and she argues that although Roman jurists never explicitly mentioned writing as a prerequisite for or a part of the stipulatio, one should nevertheless bear in mind that "their analytical world left much out": Accordingly, Gaius never mentioned that there must be a continuus or unitas actus for a valid stipulation. Moreover, she maintains that Gaius never claimed his assessment of the stipulatio to be a complete and comprehensive description of that contract and that he intended it rather to be a description of its "core nature"; that his description lends itself to speculation and that it certainly does not prove conclusively that the stipulatio was not written but oral.
3.2 African law
Although there were no formalities as regards the actual communication of the intentions of the parties and although no form or ceremony was required for a contract to be regarded as valid, contracts in African law were nonetheless not completely without form. There were certain ritualistic behavioural requirements that had to be observed, some of which correspond with the outward ritual of Roman law. Thus, both parties had to be present and actual words had to be spoken.
The transfer of property, related or unrelated to performance in terms of the contract, was crucial in legal transactions and served as a manifestation and proof of the intention of the parties.
There are many examples of the transfer of property independent of the performance due in terms of the agreement. An agreement for the transfer of marriage goods (similar to dos in Roman law), for instance, was confirmed in a physical way when an animal was slaughtered. This animal did not form part of the marriage goods. It was a manifestation of the intention of the parties and served as proof of such. This, then, is where African contracts differ from real contracts in Roman law.
Confirmation of this practice may be found in two letters, written in the late nineteenth century in Tswana (an African language) to a Tswana News Paper. In view of the dearth of early written sources in any African vernacular, by indigenous Africans, this is indeed rare. What contributes to the value of the letters is that they do not contain second-hand interpretations of cultural information by non-Africans, but narratives of Africans themselves.
The prevalence and endurance of the stipulatio in early Roman law have been ascribed to the importance of the Roman virtue of fides. This immediately begs the question whether the absence of purely verbal contracts in African customary law was related to the possible absence of fides in that culture.
4.1 Roman law
Abundant literature on the topic affirms that fides which in Roman culture implied trust and trustworthiness formed a fundamental postulate of Roman religious, socio-political and legal life. For Romans it was the most sacred thing in life. Itformed the foundation of the binding effect of obligations, not only in the ius gentium, but also in the ius civile. Everybody, irrespective of nationality, was expected to observe the duty to keep his or her word. 
The centrality of fides in Roman life was not an attribute conceived of by modern Romanists. Fides is a recurring theme in the works of Cicero, and is referred to by Seneca, Cornelius Nepos, and others. The Greeks, too, commented on the exceptional trustworthiness of the Romans. Polybius notes that bound by their pledge of fides,even Roman officials, who were ex officio most exposed to the temptation, were restrained from skimming public funds.
Cicero saw fides as "fit quod dicitur" and associated it with justice (iustitia) and other related virtues. In fact, according to him fides formed the very foundation of justice. Within the context of the stipulatio, his proposition that fides derived from a promise made good, is certainly appropriate and underwrites its importance in legal conduct. 
Interestingly, in contrast to the opinion of modern Roman-law scholars, Seneca and Cornelius Nepos saw the formality of the stipulatio as evidence of the absence of fides. In their view informal agreements should suffice between friends and fides should eliminate the necessity for formal contracts.
4.2 African law
The first logical reason that comes to mind for the absence of purely verbal contracts in African law is that fides did not play a central role in that culture. Nevertheless, although a purely verbal agreement did not give rise to legal liability it was regarded as morally reprehensible to break a promise.
But trust between people, specifically neighbours, was certainly important in ancient Africa. Gluckman, who did extensive empirical research among the Barotse of Zambia, observed that all legal relationships in their law were based on "generosity and the utmost good faith".  The pre-eminence of trust in social and legal relationships was likewise reported for the Birwa of Botswana. Among these people, neighbours within the same settlements formed neighbourhood sets that interacted frequently with each other. These sets of people were not jurally defined units like households, but were informally dependent upon each other's co-operation for their welfare and safety. Special relationships of trust existed between such neighbours who assisted each others in various activities like reciprocal labour exchanges, and who contracted with each other, among others, to acquire livestock. 
It is indeed not the lack of fides that explains why purely verbal contracts were unknown in African law, but rather the fact that African law and culture were characteristically non-specialised. This feature manifests itself in the lack of separation, differentiation, classification, and delimitation of, amongst others, knowledge, concepts, ideas, duties and interests; and hence in a concomitant lack of abstraction.
In African law then, there was an emphasis on the concrete and legal reasoning was founded in sensory observation which was also a natural corollary of the general pre-literate condition of ancient Africa and abstract principles had to be concretised as explained earlier.
There are many intersections in ancient Roman and African contracts and a superficial analysis of orality in the Roman law of contract provides insights that one could usefully employ in understanding the African law of contract.
Roman society found legal certainty in formalism, which is the extreme adherence to form, because well-defined form made legal acts memorable. In the stipulatio form was manifested in specific ritualistic words.
In African society the emphasis on the concrete and the lack of abstraction excluded words as formality. However, specific behavioural conventions, which were not limited to delivery in terms of the contract, conferred form and created legal certainty, gave rise to liability, and were regarded as manifestations of the intention of the parties. Contrary to the traditional view, performance or part performance in terms of the contract was not the only way in which the verbal agreement could be confirmed and in that sense contracts in African law differed from real contracts in Roman law.
Further, unlike the Roman stipulatio, where abstract form completely overshadowed the subjective expectations of the parties, in African law, contracts were posited in reality. The parties retained their specific identities defined by their membership of specific family groups. The object of the contract never became a "colorless commodity" and the intention of the parties remained of paramount importance
 This is a paper read at the 64th Session of the Société Internationale "Fernand de Visscher" pour l'Histoire des Droits de l'Antiquité on "Communication et Publicité dans l'Antiquité: Profiles, Juridiques, Sociaux, économiques", held 28 September to 2 October 2010 in Barcelona, Spain.
 The following works deal specifically with Southern Africa: Olfert Dapper, Kaffrarie, of Lant der Hottentots (1668), Willem ten Rhyne Schediasma de Promontorio Bonae Spei (1686) and Johannes Gulielmus de Grevenbroek Gentis Africanae circa Promontorium Capitis Bona Spei Vulgo Hottentotten Nuncupatae Descriptio (1695). These works, with translations and annotations, were published in 1933 by The Van Riebeeck Society, Cape Town, as The Early Cape Hottentots edited by I. Schapera & B. Farrington. Other works from further afield are, e.g., those of Ioannis Leonis Africani De Totius Africae Descriptione, Libri IX(1556) and, at the end of the nineteenth century, John Mensah Sarbah's Fanti Customary Laws (1897), published in London by William Clowes & Sons Ltd. This work deals with the customary laws of the Fanti and Akan tribes of the Gold Coast and is based on the works of travellers of the fifteenth, sixteenth and seventeenth centuries. It also contains judicial decisions on customary laws.
 "European" is not a precise term but is used here as referring to what F. Wieacker "Foundations of European Legal Culture" 1990 (38) American J. of Comparative Law 1 at5, calls the Atlantic-European legal orders which include not only Western Europe, but also European settlements in North America, parts of Central and South America, Australia, New Zeeland and Southern Africa.
 It was Colonel Maclean's Compendium of Kafir Laws and Customs which was a compilation of various sources on the laws of the amaXhosa. Colonel Maclean was Chief Commissioner in British Kaffraria. His Compendium consisted, among others, of papers of a certain Reverend Dugmore, initially published in 1846 and 1847 in The Christian Watchman; a letter from Maclean; notes of Warner, Tambookie agent in British Kaffraria in 1856; and notes of Brownlee, Commissioner of the Gaika People.
 W.M. Seymour's Native Law and Custom was published by Juta & Co in Cape Town in 1911. During the 1950s the School of Oriental and African Studies embarked upon the first comprehensive project on the restatement of indigenous African private law in the former British colonies.
 W. Menski Comparative Law in a Global Context Cambridge University Press, Cambridge (2006) 466 observes that there is a common perception that unlike other legal systems, African law has never progressed beyond its primitive stage. However, since early Roman law, too, had many features which scholars have regarded as relics of the primitive stage of its development, such as the extreme formalism, ritual, symbolism and the ubiquitous magic it was only natural to use its jurisprudential framework to explain African law. With regard to relics of primitivism in Roman law see generally GY. Diósdi Contract in Roman Law. From the Twelve Tables to the Glossators Akadémiai Kiadó, Budapest, (1981) 42-43; G. MacCormack "Formalism, Symbolism and Magic in Early Roman Law" 1969 (37) Tijdschrift voor Rechtgeschiedenis 439; C. Wasserstein Fassberg "Form and Formalism: A Case Study" 1983 (31) American. J. of Comparative Law 627-630; K. Tuori "The Magic of Mancipatio" 2008 RIDA 499ff.
 In South Africa, today, only fractions of the true traditional African law form part of the official State law. Some sixty years ago already, Anthony Allott, warned that the traditional African law was rapidly being displaced by the imposed colonial law: See A.N. Allott "Introduction" in J.O. Ibik Malawi II: The Law of Land, Succession, Movable Property, Agreements and Civil Wrongs London, Sweet & Maxwell, London (1971) v.
 It is not surprising then that agreements were always entered into verbally and that oral communication of intent in juristic acts prevailed in both African and Roman law. However, the function and consequences of words in legal acts differed. This feature was not limited to these two ancient societies. See generally M. Kaser Das römische Privatrecht Vol. I (1971) 39ff., 230ff.; Vol. II (1975) 73ff., on orality and formalism in early Roman law.
 See, e.g., Kaser I (n. 8) 230-231; F. Schulz History of Roman Legal Science University Press, Oxford (1953) 25; E.Meyer Legitimacy and Law in the Roman World. Tabulae in Roman Belief and Practice Cambridge University Press, Cambridge, (2004) 36ff.
Evidence exists that legal acts may have been recorded in writing as early as the sixth century BC. However, the fact that orality dominated in private legal acts and that legal documents were not an essential component of the legal process in early Roman law are confirmed by the Twelve Tables, Tab. VI.1 which states: Cum nexum faciet mancipiumque, uti lingua nuncupassit, ita ius esto (When a person makes bond and conveyance, according as he specified with his tongue so shall be the law): Translation by the Yale Law School's Avalon Project at http://avalon.law.yale.edu/ancient/twelve_tables.asp (accessed 31 August 2010).
 Cicero clearly distinguishes between written and unwritten law, and specifically refers to legal documentation in mancipatio and stipulatio. See, e.g., Cicero Ep. Att. 12.17: [T]amen velim des operam, ut investiges ex consponsorum tabulis, sitne ita ... (I should be glad if you would verify the truth of that statement from the account books of other surities ...); Ep. Att. 16.11.7: Etsi nondum stipulationes legerem ... (Though I have not yet read the agreements ...), de Orat. 2.100: At vero in foro, tabulae, testimonia, pacta conventa, stipulationes ... tota cognoscenda est (But in the law courts, documents, evidence, informal agreements, formal contracts ... must all be examined). Nevertheless, Roman scholars debate the significance of the documentation of legal acts: cf. the discussion under 3.1 below.
 Anecdotal evidence of this contract abounds in the literature: see, eg., Plautus Bacch. 880-883; Cicero Rhet. Her. 2.13-14 and Ep. Att. 16.11, De Or. 2 100, De Leg. 1 14. On verbal contracts, see generally Ulp. Inst. 3.15; D. 45.1; Kaser I (n. 8) 538-543.
 In fact, the intention of the parties in verbal contracts was not of primary concern: R. Zimmermann The Law of Obligations. Roman Foundations of the Civilian Tradition Juta & Co., Cape Town 1990) 563-565, 622, 626.
 Watson "Artificiality, Reality and Roman Contract Law" 1989 (57) Tijdschrift voor Rechtsgeschiedenis 151.
 S. Amos The History and Principles of the Civil Law of Rome Kegan Paul, London (1883) 203 refers to it as an "outward ritual".
 Amos (n. 14) 202-4, 215, 219.
 Amos (n. 14) 202; F. De Zulueta The Institutes of Gaius Part II Commentary Clarendon Press, Oxford (1953) 151-152; Zimmermann (n.12) 70 83-84; Wasserstein Fassberg (n. 6) 627.
 Watson "Artificiality" (n. 13) 155-156; Amos (n. 14) 202. Validity of the legal act came from its form not from consensus: J.A. Harrill "The Influence of Roman Contract Law on Early Baptismal Formulae" Papers Presented at the Thirteenth International Conference on Patristic Studies, Oxford 1999 Belgium 2001 (35) Studia Patristica276ff. Cf. D.G. Kleyn "The Reality of Real Contracts" 1995 (58) Tydskrif vir Hedendaagse Romeins-Hollandse Reg 16 at 16-17.
 See, e.g., Gaius 4.116a: Thus, if I have taken a stipulary promise from you of a sum of money, on the understanding that I will advance you the amount on loan, and then I do not advance it, it is undeniable that an action lies against you for the money; for you are legally liable to pay it, being bound by the stipulation; but, because it is inequitable that you should be condemned on this account, it is settled that you must be protected by an exceptio doli mali. Cf. De Zulueta II (n. 16) 147-148, 151; Nicholas An Introduction to Roman Law Oxford University Press, London (1962) 22 164; J.A.C. Thomas The Institutes of Justinian Juta, Wynberg (1975) p. 209.
 The Romans were surprisingly indifferent to problems of evidence which were regarded as the domain of the judge, not the jurist. The contracting parties relied principally on fides as security: Zimmermann (n.12) 70, 89, 624-625. Kaser I (n. 8) 39f., II (n. 8) 73f., indicates that in certain instances something additional was required to formal words to bring about public knowledge and to serve as evidence of the intention of the parties.
 It is generally accepted that in African law liability ensued only where one party had in fact partially or totally fulfilled his obligation. M. Gluckman The Ideas in Barotse Jurisprudence Manchester University Press, Manchester (1972) 181 points out that it was not enough to merely point out the goods, but that delivery had to take place before an obligation could be created. But constructive delivery was sufficient. See generally idem 177-182; see, too, M.W. Prinsloo & L.P. Vorster "Elements" in A.C. Myburgh Indigenous Contract in Bophuthatswana Van Schaik, Pretoria (1990) 6‑7 10‑11; F.P. van R. Whelpton Inheemse Kontraktereg, unpublished LL.D. Thesis, University of South Africa, (1991) 81‑83.
 "Lentswe la maabane ga le thlabe kgomo".
 "Obi nto nantwi anmon": Sarbah (n. 2) 93.
 This ancient custom still persisted in the early twentieth century in South Africa. See, e.g., Xapa v Ntsoko 1919 EDL 177 esp. at 181.
 See, e.g., Grevenbroek (n 2) 136 137. See also Sarbah (n. 2) 93ff., who observed that "trama" or earnest money was required to "bind the contract".
 In Tswana law this process was described as "go tshitsinya", which literally means "to introduce": Whelpton (n. 20) 81‑83
 A.C. Myburgh "Language" in in AC Myburgh (ed) Anthropology for Southern Africa Van Schaik, Pretoria (1981) Myburgh140-144.
 Gluckman Barotse Jurisprudence (n. 20) 183; M. Poland, D. Hammond-Tooke & L.Voight The Abundanta Herds. A Celebration of the Cattle of the Zulu People Fernwood PressSimon's Town (2003) 34.
 Poland, Hammond-Tooke & Voight (n. 27) 36-37; they point out that this intricate naming and classification has significant alliterative and lyrical qualities.
 The restrictions on the use of certain words reminds of the limitation in Roman law of the use of spondeo to Roman citizens with its possible religious origins in the oath before the Roman gods: Gaius 3.92; Harrill (n. 17) 277; De Zulueta II (n. 16) 153: Gaius 3.92, points out that the sponsio was restricted to Roman citizens but that other forms of stipulation were also available to foreigners and that other languages could even be used, as long as the parties could understand each other.
 Schulz Roman Legal Science (n. 9) 25-26.
 According to Harrill (n. 17) 279, the the longest surviving non-legal description of the stipulatio may be found in Varro Rust. 2.2.5-6.
 A phrase coined by Amos (n. 14) 202; for a general discussion of this ritual see Zimmermann (n.12) 72-75; Kaser I (n. 8) 539ff.; Meyer (n. 9) 116-117; W.W. Buckland A Textbook of Roman Law from Augustus to Justinian (1966) 434-435.
 Gaius 3.105: That a dumb man can neither stipulate or promise is obvious. The same is accepted also in the case of a deaf man, because it is necessary both that the stipulator should hear the words of the promissor and that the promissor should hear those of the stipulator.
 Gaius 3.92: A verbal contract is formed by question and answer, thus: 'Dost thou solemnly promise that a thing shall be conveyed to me?' 'I do solemnly promise.' 'Wilt thou convey?' 'I will convey.' 'Dost thou pledge thy credit?' 'I pledge my credit.' 'Dost thou bid me trust thee as guarantor?' 'I bid thee trust me as guarantor.' 'Wilt thou perform?' 'I will perform.' The requirement of specific words - originally limited to Latin, spondere, and for Roman citizens - was relaxed in classical times when any words could be used: Buckland Textbook (n. 32) 434-435 contra B. Nicholas "The Form of the Stipulation in Roman law" 1953 (69) The Law Quarterly Review 63 at 65ff. There is however recent documentary evidence that slaves and foreigners could make use of the sponsio: J. Urbanik "Sponsio Servi" 1998 (28) Journal of Juristic Papyrology 185-201, quoted in Harrill (n. 17) 278 n. 2.
 Gaius 3.136: "[A] verbal obligation cannot be formed between parties at a distance"; see also Gaius 3.138. Cf. Meyer (n. 9) 255.
 Gaius 3.102.
 Venuleius D 45.1.137pr.: continuus actus; Modestinus D 184.108.40.206; Ulpianus D 220.127.116.11; Gaius D.18.104.22.168; De Zulueta II (n. 16) 153; Nicholas "The Form of the Stipulation" (n. 34) 64-65; Buckland A Manual of Roman Private Law (1939) 264; Cf Meyer (n. 9) 116-117.
 See Zimmermann (n.12) 80-82 for a discussion of the conversion of the verbal contract into a written one.
 Zimmermann (n.12) 79; Kaser I (n. 8) 540ff., II (n. 8) 373; G. MacCormack "The Oral and Written Stipulation in the Institutes" in P.G. Stein & A.D.E. Lewis Studies in Justinians Institutes in Memory of J.A.C. Thomas Sweet & Maxwell, London (1983) 96ff; P.J. du Plessis "The Roman Concept of Lex Contractus" 2006 (3) Roman Legal Tradiotion 79-80. Nicholas "The Form of the Stipulation" (n. 34) 77ff., 233ff. De Zulueta II (n. 16) 155 observes that Cicero, as layman, was wrong to assume that stipulationes were among the res quae ex scripto aguntur: written stipulationes were valid only if orally confirmed and oral stipulationes were valid irrespective of whether they had been documented; see further 156 -157.
 M. Kaser Das römische Zivilprozessrecht (2nd ed. by Hackl (1996)) 10-11; E. Metzger "Roman judges, case law, and principles of procedure" 2004 (22-2) Law and History Review 264ff; Meyer (n. 9) 2 36-39.
 Schulz Roman Legal Science (n. 9) 25-26.
 E. Metzger "Roman Judges, Case Law, and Principles of Procedure" 2004 (22-2) Law and History Review 246 at 262ff.
 Harrill (n. 17) 275 at 276; Buckland Manual (n. 37) 263; Zimmermann (n. 12) 71 esp. n. 20. Nicholas Introduction (n. 18) at 196 points out that there is some controversy regarding exactly when the written stipulatio replaced the oral one, but some scholars are of the opinion that the time could be fixed at 472 with a rescript of Emperor Leo which removed the requirement of the formal words. Nicholas "The Form of the Stipulation" (n. 34) 62ff., argues, though, that the rescript of Leo abolished the use of specific formal words, but not the oral contract. See also Thomas (n. 18) 209; Kleyn (n. 17) 18-19.
 See Meyer (n. 9) at 117; see further 116-117 and the sources quoted in nn.102-106; 253-261. Based on Roman texts, scholars have debated the possibility also of various other actions being part of the stipulatio such as the pouring of libations, offering the right hand as a symbol of the fides, a combination of these two actions and holding and breaking a reed.
 Contra Nicholas "The Form of the Stipulation" (n. 34) 65ff, who argues that Gaius provided an exhaustive list of formal words. This would disprove the argument that Gaius described only the core nature of the contract.
 Diósdi (n. 6) at 52, too, is of the opinion that that the predominance of orality had not endured that long and that already during classical times the speaking of formal words was no longer necessary where there was a written document in place. He claims that the Roman jurists "were not interested in whether the parties had recited the stipulatio contained in the document" and that this polemic was merely an issue of modern scholarly debate: See generally idem 51ff.
 As in Roman law, the nod of a head was not an indication of the intention of a party.
 See Gluckman Barotse Jurisprudence (n. 20)176; contra L.P. Vorster "Independent service" in A.C. Myburgh Indigenous Contract in Bophuthatswana (n. 20) 52-53.
 E.g., among the Tswana, a verbal agreement with several people to perform a specific task was concretised in the slaughtering of an animal which was divided among those who agreed to do the work. The gift of the meat could not be regarded as payment for the services which would be rendered in terms of the contract.
 The custom of transferring marriage goods differed among the different ethnic groups: L.P. Vorster et al. Urbanites' Perceptions of Lobolo: Mamelodi and Atteridgeville Unisa Press, Pretoria (2000) 76. Betrothal or the agreement to transfer marital guardianship over a women to the family of the prospective husband was concretised when the boy's family offered a gift to the girl's family to ratify the verbal agreement: J. Church "Betrothal and Marriage: Contractual Aspects" in A.C. Myburgh Indigenous Contract in Bophuthatswana Van Schaick, Pretoria (1990) 84-86. In infant betrothals the family of the baby boy would give a goat and a cow to the family of the baby girl as part performance and to confirm liability to transfer the girl in marriage when she reaches a marriagable age.
 The London Missionary Society published the newspaper Mahoco a Becwana during the years 1883-1896. T.M. Mgadla and S.C. Volz translated into English and edited a selection of the letters that Africans had written to the editor of the newspaper. These were published as Vol. 37 of the Second Series of the Van Riebeeck Society in Cape Town in 2006 as Words of Batswana. Letters to Mahoko a Becwana 1883-1896.
 See Baruni Makutle's letter in Magadla & Volz idem at 145-146 and the letter of Paramount Chief Montshiwa Tawana at 161.
 See among others A.R. Fromchuck The Concept of Fides in the Histories of Tacitus, unpublished Ph.D. Thesis, Bryn Mawr College, University of Michigan Ann Arbor (1972) 1ff; D.H. Van Zyl Justice and Equity in Cicero Van Schaik, Pretoria (1991) passim; Meyer (n. 9) 150ff; Zimmermann (n.12) 68-70; F. Schulz Principles of Roman Law Clarendon Press, Oxford,(1936) 223ff (especially 326-328 for its significance in law).
 See Cicero in Verr. 22.214.171.124: fidem sanctissimam in vita qui putat.
 See, e.g., Kaser I (n. 8) 27, 33, 35, 39, 87 esp. his references to the connection between fides and sacral law.
 See, e.g., de Rep. 4.7, ad Fam. 16.10.2, de Offic. 1.7.23. de Part. Orat. 22.78 Cicero observes: "That part of virtue displayed ... in matters of trust [is called] faith."
 Ben. 3.15.1-2.
 Att. 9.5.
 Hist. VI 56.14-15: "... whereas among the Romans those who as magistrates and legates are dealing with large sums of money maintain correct conduct just because they have pledged their faith by oath. 15. Whereas elsewhere it is a rare thing to find a man who keeps his hands off public money, and whose record is clean in this respect, among the Romans one rarely comes across a man who has been detected in such conduct ... .". Also Cicero iremarked on the role of fides in affairs or relationships of trust: de Part. Orat. 22.78: in creditis rebus fides; cf. Van Zyl (n. 53) 97-98.
 At 21 Meyer (n. 9) comments that "it was at first rare (although eventually better known) for suspicion of corruption to touch the Romans themselves".
 de Rep. 4.7: Fides enim nomen ipsum mihi videtur habere, cum fit, quod igitur. (Faith seems to me to get its very name from the fact that what is promised is performed).
 de Offic. 1.7.23: fundamentum autum est iustitiae fides, id est dictorum conventorumque constantia et veritas, ex quo, quamquam hoc videbitur fortasse cuipiam durius, tamen audeamus imitari Stoicos, qui studiose exquirunt, unde verba sint ducta, credamusque, quia fiat, quod dictum est, appelatam fidem ... . Cf. Van Zyl (n. 53)123.
 In his de Offic. 1.7.23 he wrote that undertakings and agreements (dictorum conventorumque)should be upheld and the resulting obligations be discharged. This applied in both public and private acts.
 Ben 3.15 1-2: 1 Utinam nulla stipulatio emptorum venditori obligaret nec pacta conventaque impressis signis custodirentur, fides potius illa servaret. 2 Sed necessaria optimis praetulerunt et cogere fidem quam expectare malunt ... .
 Att. 9.5: "[H]e came to the rescue and lent her the money without interest and without any contract, considering it the greatest profit to be known as mindful and grateful, and at the same time desiring to show that it was his way to be a friend to mankind and not to their fortunes ... ."
 I. Schapera in M. Gluckman (ed.) Ideas and Procedures in African Customary Law Oxford University Press, Oxford(1969) 327-328 reported for the Tswana of Botswana that executory contracts created legal liability. However, this is not a general view and Epstein & Gluckman note that he may have interpreted the law incorrectly: Gluckman Barotse Jurisprudence (n. 20) 180, 182-183; at 182-185, he explains the divergence of the centrality of good faith in all spheres of African life and the fact that a mere agreement did not incur liability as follows: Liability in contract is indeed only incurred when performance has taken place or property transferred, but when the obligation is created, it is governed by good faith (not in the modern ethical sense). Cf., further, N. Mahoney "Contract and Neighbourly Exchange among the Birwa of Botswana" 1977 (21) J of African Law 40 at59.
 Gluckman Barotse Jurisprudence (n. 20) 175 and generally 174-176.
 Mahoney (n. 66) 40, 49-53; cf. Gluckman Barotse Jurisprudence (n. 20)170ff.
 A.C. Myburgh Papers on Indigenous Law Van Schaik Pretoria (1985) 2ff.
 See C.T. Wonnell "The Abstract Character of Contract Law" 1990 (22) Connecticut L.R.. 437 at 438-441.