A few historical examples

3Calculations show that for the city of Bologna in 1288, each year 1800 civil servants in various categories had to be “elected”; the task was delegated to a Council composed of 2000 members at the time (which was later expanded to include 4000 members), who were themselves selected annually by the Consiglio generale and the Consiglio del Popolo, each of which had 800 members. [5] These two councils were in turn elected on an annual basis. In addition, knighthood societies, merchant associations, the Popolo as autonomous legal entity, the guilds (the Arti), milites and other societates likewise elected their leaders, their board of directors, and their steering council. [6] In terms of understanding electoral procedures, all of these different kinds of elections must naturally be considered as a whole. Therefore, the gara degli uffici (the race for positions) that sought to influence elections and was criticised by contemporary historians did not only concern the consulate or, in Florence, the Priorato (priorate). [7]

4Unlike the elections of popes, kings, bishops and abbots, communal elections in Italy did not demonstrate a progression towards a universally applicable procedure that was refined at every stage. The Italian communal Constitution was instead characterised by frequent changes in organisation and electoral procedures which, when they were not exclusively defined for a specific, upcoming vote, were nonetheless always applicable for only a very short period of time. At the end of the twelfth century, the consuls of Piacenza were required to solemnly swear that, at the end of their mandate, they would accept the designation of their successors in accordance with whatever procedure the Council had adopted, assisted if need be by the guild leaders, either unanimously or by majority. At a time when most cities were ruled by foreign podestàs, the Council, according to its bylaws, was required to decide, shortly before the end of its mandate, if the city was to be governed again by a podestà, or by another form of government, in which case an ad hoc electoral procedure was needed. [8] In Florence, the Council of the Fourteen, which ruled between 1280 and 1283, was elected each month following a procedure that had to be redesigned each time. [9] The Ordinances of Justice (Ordinamenti della Giustizia) [10] stipulated that the “captain of the people” (capitano del popolo), working with the guilds as well as a number of notable figures, was responsible for determining how voting would be carried out the day before the end of the sitting priorate’s mandate. Since in Florence the consulte – the records of the motions proposed to the Council of the One Hundred [11] and the subsequent votes – were preserved for years, it is possible to follow the procedural debates closely regarding the election of Signoria members. [12] For example, we can see that on 24 November 1292, no fewer than 23 different motions were presented for debate. The new form of governance, in conformity with the length of the planned mandate, was to remain in effect for six months; 29 members of the Council had voted to this effect. The motion to limit its validity – and thus the mandate’s – to two months appears to have received 58 votes in favour, but likely failed to pass because it lacked the necessary quorum. John M. Najemy has analysed with great precision the frequency and radical nature of transformations in electoral procedures in Florence’s Signoria between 1280 and 1400, as well as how the political interests of specific groups were expressed in these restructuring processes. [13] What was true for Florence was already no longer the case in Siena, however, and Najemy neither posed nor answered the question of how such transformations concerning elections in Florentine communal government interacted with the procedures adopted for other elections. [14]

5The medieval electoral procedure used in Italian city-states that has produced the greatest number of historical studies is the complex, ten-round protocol used to select the doges of Venice. Introduced in the thirteenth century and employed until the eighteenth century, this protocol was unique not just because Venice was alone in having a doge, or because the city’s Constitution had been shaped very early on by the Serrata del Maggior Consiglio, [15] during which an ordinance concerning the doge’s election was introduced. The variability of electoral procedures, as well as the possibility of only being able to define them for a very short period of time in the lead-up to an election, appears to have been an element of what we might call the constitutional principles of the Italian communes: already entrenched in the procedure, these conditions are meaningful with regard to understanding the nature of elections. Regulations set for a longer term were primarily adopted when it was vital to ensure the dominance of a selected ruling group, for example through a kind of elective monarchy. In this light, such long-term arrangements would represent one aspect of the transition from the medieval commune to the city-state (or region-state) of the early Italian Renaissance. [16]

6Anyone attempting to study the electoral procedures of the Italian communes is thus immediately faced with a disconcerting variety of facts and situations that are not only very difficult to put into some sort of order, but which are moreover ill suited to systematic research. Many communal statutes tell us nothing about electoral procedures, with the result that this problem cannot be simply clarified by recourse to normative legal sources. Treatises on how to establish the best form of government largely avoided the topic of elections. Marsilius of Padua, whose political theory attached crucial importance to collective decision-making, made only the following desultory statement regarding electoral procedures: “The means of fashioning agreement for the aforementioned institution or election varies greatly from province to province.” [17] In many cases, important information is provided in contingent reports linked to an event, the writings of chroniclers, partisan criticisms of a procedure itself accused of being partisan, and even under the guise of assembly decisions and council proceedings, such as the aforementioned Florentine consulte.

Social logic, normative orientation and procedural conflicts

7In the aforementioned example from 1292, which cites one of countless procedural debates from the period, it becomes apparent that it is impossible to truly understand the motions proposed without taking into account the political backdrop: tensions between the Arti Maggiori (the major guilds) and the other, minor, guilds; [18] the threat posed by the Popolo to the magnates (magnati), who would be excluded from holding political office the following year with the passing of the Ordinances of Justice; as well as scissions within the Guelph party. [19] It was from this perspective that John M. Najemy was able to describe the evolution of electoral systems in Florence and convincingly highlight the struggles for a communal government initially based entirely on the Arti, as a principle of socio-political organisation, followed by the ultimate victory of a regime that sought to establish greater consensus among the leading families and defend against the guilds’ ambitions. [20]

8Studies of this type naturally lead to the question of whether it is judicious to categorise electoral procedures from a formal point of view – for example, as plurinominal contests, consensual processes, sortition, majority decisions, etc. – without contextualising them in relation to the political situation in which they were applied. Even when new procedures were set or earlier ones were modified to achieve new political objectives or to promote the interests of specific groups, however, such decisions had to be deemed legitimate in terms of “justice”. Here we touch upon a central point: the perception that the political community had of itself. First of all, the community was careful to exclude any acts of personal favouritism and to condemn the promotion of personal interests. During the thirteenth century, many cities openly decreed that only staunch supporters of the ruling party could be elected to office; this provision was also sometimes included in city statutes. Perhaps even more troubling to our modern sensibilities, certain social groups were periodically excluded from access to political office, including the magnates, the milites, and the representatives of various occupations and professional groups. In many places, employees, sailors, and other members of the working class were excluded from citizen assemblies. To justify such measures, the political community invoked the defence of justice, peace, and the common good. The process and outcomes of elections were seen as a touchstone by which to judge the success – or failure – of such objectives. Nonetheless, debate regarding the equitable distribution of offices was constant, even when the ruling party’s dominant share was not challenged. To that extent, electoral procedures could not be arbitrarily manipulated, but they had to be measured against the demand for justice that represented the foundation of domestic peace: the shape that they took was prescribed by a certain number of fundamental ideas about the commune. [21] From this perspective, it becomes both possible and pertinent to examine electoral systems independently of their political background.

9Nonetheless, even when such an attempt is made, one must still determine how elections for communal offices in government and administration that were endowed with such brief mandates (in most cases, just a few months) can realistically be compared with the elections of kings or bishops. It seems that the best way to develop a comparative political perspective is to follow, throughout the twelfth and thirteenth centuries, the development of certain electoral systems, and to establish their relationship with the community’s self-image and the concept of voting that formed the epistemological backdrop.

Elections “by compromise”

11Unfortunately, we only possess concrete information regarding consular electoral procedures from 1150, or several decades after the emergence of the consulate as an institution. In Genoa, Pisa and Pistoia, the election was limited to a small number of electors who had to make their decision by mutual agreement – in concordia. In Pistoia, it seems that during certain periods, there were three (or perhaps five) electors in charge of electing a committee of five consuls. There are likewise early accounts of this system in Venice. In 1172, eleven electors selected the doge, this number increasing to forty in 1178. At the end of the twelfth and thirteenth centuries, this form of indirect election by electoral commission had spread universally throughout the Italian communes. The podestà, the consuls of justice, the elders, the leaders of guilds and societates armorum (politico-military organisations), statute drafters and other agents in the commune’s self-government were all selected in this manner. The Council (Consilium) was likewise composed via indirect election. It should be noted that the decision made by the electors was final, which meant that they made a legitimate choice in the community’s name, and not merely one that another body, the Popolo or the Council for example, could subsequently accept or reject. The development of the communal electoral system therefore went hand-in-hand with the emergence of “elections by compromise” at the beginning of the twelfth century. [23] As a consequence, electors had to come to a decision unanimously or through consensus. The committees were partly appointed en bloc and in some cases electors had to ensure the equitable treatment of the states, guilds, and districts. Starting in the twelfth century, these electoral commissions also had recourse to majority decisions, which the community had to accept as binding. This transformation likewise occurred alongside the general development of elections “by compromise” and arbitration proceedings. [24]

12One of the particularities of the Italian communal electoral system was that elections were a multi-level procedure, in the sense that electors were themselves appointed by grand electors. The oldest surviving accounts attest to the fact that the electores consulum were selected by the electores electorum consulum. Similarly, in Venice in 1178, the forty electors of the doge had themselves been selected by four electors, who had in turn been chosen by People’s Assembly. This two-round procedure was subsequently confirmed for the electors of the podestà and those responsible for selecting the individuals in charge of amending statutes. With regard to first-stage electors, it is unclear whether they were required to reach a consensus on an electoral commission – like the commissions of consular electors – or if each one of them designated an elector consulum or an emendatorum constituti. This two-step procedure sometimes grew to include additional steps, as seen in the extreme case of the election of the doges in Venice. It no doubt also encouraged the trend of including sortition in electoral procedures, as was increasingly the case throughout the thirteenth century.

Relationships between political and ecclesiastical elections, and the specific contribution of Italian communes

13One look at how bishops were elected in Italian cities where communal Constitutions were simultaneously being drafted reveals that the evolution of the political electoral system was closely linked to the evolution of ecclesiastical elections. [25] Documents attesting to the appointment of bishops by electors (in the case of Genoa, the process was referred to by the technical term “per compromissum”) are older and even more numerous than those mentioning the procedures implemented to elect consuls. The formal call to open proceedings for episcopal elections was often emitted by the city’s governing bodies.

14According to the oldest existing documents, the process of establishing the College of Grand Electors could apparently unfold in a rather informal fashion. In Milan in 1111, when heated controversy surrounded Archbishop Grossolan, one notable secular figure took the floor before the People’s Assembly and proposed entrusting the decision to an arbitral committee composed of nine clergymen from each of the two parties, whose names he then proceeded to read out. If these eighteen men were to reach the conclusion that Grossolan did not legitimately fulfil his duties, they would have to agree upon a new archbishop. The entire city would then have to approve this decision. A certain number of individuals from the People’s Assembly, also named by this initiative-taker, were to swear, on behalf of the whole, that they accepted the committee’s decision. In Bergamo in 1133, the consuls asked the archpriest to initiate re-election proceedings. Addressing the primicerius, [26] the archpriest said: “Stand up, and appoint other grand electors that shall stand next to you”. Other examples from Bergamo, Genoa, Lucca, Como, Piacenza and Pavia all attest to the fact that the use of electors to appoint bishops had become a widespread practice. In accordance with the “per compromissum” principle, those electing the bishop had to reach a decision unanimously (majority decisions would only appear in the thirteenth century). During the same period, multi-round procedures were likewise documented. In Lucca, in 1225, sortition was used to elect a canon, who in turn had to nominate three electors. Subsequently, Pope Honorius III pronounced a landmark decree stipulating that sortition (or lotteries) were not sufficiently trustworthy to be used by electoral commissions. [27] In 1256, the cathedral chapter in Lucca had its two treasurers and two sacristans appoint three electors. In 1269, it delegated per viam compromissi the choice of the two filii populi electors to the treasurer and two sacristans; these “sons of the People” – who were also canons – selected the three episcopal electors from among their midst. The fact that only late accounts exist of multi-round episcopal election procedures and that attempts were made to introduce sortition into these procedures clearly illustrates that communal practices were transposed onto the ecclesiastical realm.

15It is likely that the development of established electoral procedures in the Italian communes was closely linked to an evolution in how the Church viewed elections, both before and after 1100, and to the consequences of this perception. However, this also signifies that the integrity of voting – in other words, its independence with regard to material interests, human tendencies and partisan bias – was from the beginning one of the main problems that electoral techniques were designed to solve. Likewise, elections by compromise in general, as well as other electoral systems unique to the Italian communes, are scarcely conceivable without analysing the notion of universitas [28] and the question of the individual representation of a collective community, issues that were highly pertinent at the time in the West. Thinking on these issues was informed by new directions in legal thought. [29] I suggest that the specific contribution of the Italian communes lies in the creation of multi-round voting procedures, as well as in the introduction of sortition. These techniques were the only ones that managed to institute fixed voting systems in cases of a very large and heterogeneous universitas, membership of which took multiple forms, without immediately establishing the dominance of a small group. Ruffini suggests that indirect elections always gave power to an influential minority; he thus describes the “oligarchical nature of elections in the first communal magistracies”. [30] Although that was perhaps the case – Najemy echoes this hypothesis – this conclusion does not help us to understand the forms of representation that led to the establishment of electoral procedures.

The election of council members in Italian communes (twelfth and thirteenth century)

17Prior to the middle of the twelfth century, a Council had been established in many Italian communes alongside the consular regime, to which numerous duties were conferred that had hitherto been the responsibility of the People’s Assembly. [31] As far as I know, with regard to the twelfth century, it is impossible to ascertain how the Council’s composition was determined. In the thirteenth century, in almost every case its composition was determined by grand electors. A large variety of forms can be observed here. Two later procedures give us a more precise idea of the phenomenon’s scope. When the rule of the Margrave of Este was toppled in Modena in 1306, the People’s Assembly first decided to instate a new Council of the Four Hundred, one hundred for each of the neighbourhoods associated with each gated entrance to the city. The capitano del popolo appointed two priests from each religious order: the Dominicans, Franciscans and Augustinians, who were each to name ten electors per neighbourhood; these electors were in turn responsible for electing the Council of the Four Hundred. According to Lucca’s statutes in 1308, in the city’s four neighbourhoods (as well as in the village of Borgo San Frediano which was subject to the same statute), a hundred and ten men were to be selected at random from among the inhabitants (vicini), each of whom would in turn individually name a member of the Council of the Five Hundred and Fifty from his own neighbourhood. All men who had been living with their families in the neighbourhood in question for at least eight days were eligible for the lottery; the sole requirement for each of the consiliatores being elected was that during the last estimo, their assets had been worth at least twenty-five lire (a very modest sum). These two fourteenthcentury procedures distinctly show that the electoral system was designed to prevent the formation of political “parties” and partisan political will. Modena illustrates the general rule, which consisted of having the relatively numerous members of a Council be elected by a relatively small number of electors. In Vicenza, for example, according to the statutes from 1264, an eight-man committee and the podestà determined who was elected to the Council of Forty, while a twenty-man committee and the podestà decided who was elected to the Council of the Four Hundred. On each occasion, these electoral committees were again appointed by the podestà and the twelve elders of the Popolo.

18Using this example, it would perhaps be helpful to illustrate the complexity of the electoral system as a whole. The twelve elders (anziani populi) who appointed the Council electors were themselves selected according to a complex procedure. Each of the eight guilds proposed four candidates for the Council of Forty; out of those thirty-two men, eight were ultimately elected to the office – one per guild. These eight were those who had obtained a majority throughout multiple rounds of voting. For its part, the Council of the Forty used a random draw to elect an eight-member electoral committee, which in turn proposed sixteen other candidates, from which the Council, following the same voting procedure, was to select four new members for the Council of the Elders. Once appointed, the elders, in collaboration with the foreign podestà – who was himself elected following a complex procedure – selected the grand electors who would determine the composition of the Council of the Forty and the Council of the Four Hundred. The two Councils and the electoral commission that selected the elders were recruited in equal proportion from each of the city’s neighbourhoods; the composition of the electoral commission in charge of electing the podestà, whose configuration was determined by the Grand Council, likewise took the city’s composition into account.

19Taking the city’s internal composition into consideration did not just mean highlighting the participation of the whole. On the contrary, it seemed designed to prevent the formation of political groups. This is illustrated by one measure whose implementation was required in Bologna in 1250 on the occasion of the Council’s election. A committee of forty experts – ten per neighbourhood, appointed by the podestà, the consuls representing the merchants and moneychangers, and the anziani populi – was responsible for subdividing each neighbourhood into four sub-neighbourhoods (contratae). Like the city’s neighbourhoods, the contratae were districts called upon to select the eight hundred members of the Consilium credentiae and the eight hundred members of the Consilium generale. In both cases, the procedure started by randomly selecting five electors per neighbourhood; once this committee of twenty grand electors was created, it met in conclave, attended by the podestà and a notary, to determine the future composition of the Council. The procedure was as follows: five members of the Council were first elected per neighbourhood, then five per contratae. After each round, one hundred members of the Council had thus been appointed, and the total committee was created after eight rounds. No further details concerning the voting process have survived: it seems that in each case, the electoral committee had to agree upon five individuals, instead of neighbourhood representatives being allowed to nominate their own men. In parallel to these kinds of procedures, during the thirteenth century, certain cities also turned directly to lotteries. For example, in Novara in 1287, two hundred members of the Consilium were randomly selected via lottery, after the podestà had first appointed fifty-one members on its own initiative. The lottery had to respect a clause which stated that out of the two hundred and fifty-one elected individuals, one third came from the guilds and another third from the milites; the final third was composed of those “qui non sunt nec de paraticis nec de militibus”. Naturally, such a procedure can only be correctly analysed if we know who exactly was allowed to participate in the lottery system.

Appointing the podestà during the thirteenth and fourteenth centuries

20From the perspective of the techniques implemented, we shall briefly examine the election procedure for podestàs, which thirteenth-century Italian communes frequently designated as neutral bodies with executive and judicial powers, established for relatively short periods of time (six months to a year). [32] There as well, the use of grand electors was widespread. It is possible that originally grand electors had appointed future podestàs directly, who then ruled over the community as a whole, as did the consuls. This theory is supported by a passage in Treviso’s statutes from 1207. The recording of the proceedings for the podestà election in Genoa in 1237 likewise supports this theory. This latter election provoked considerable controversy, as one of the six randomly selected members of the electoral commission had refused to accept the nomination made by the other five. In Piacenza in 1223, documents attest to the existence of an external arbitrating judge to broker agreement between the aristocracy and the Popolo on their mutual appointment of a foreign podestà. First, the arbitrating judge selected two hundred men from each of the two parties; from these, party leaders subsequently had to choose four delegates to form an electoral committee. But the third-day conclave during which the delegates were locked up without food or drink only proved that they could not reach an agreement. [33] The judge then had each party designate sixty representatives, from which three electors for each party were randomly drawn; the electors in turn had to elect one candidate during a conclave that lasted from Sunday to Friday. From that point on, the existing documents do not mention any other procedures (though a few modifications were observed). Although in every case, grand electors were used to appoint the future podestà, they were now little more than a sort of selection commission, which did not directly choose an official but merely proposed a list of possible candidates. The Council decided in what order these candidates were to be presented; in some cases, it could even reject the list. In Siena in 1250, three electors chosen by the grand electors were required to present the Council with a list of four candidates, all of whom had to have been previously approved by the majority, before they proceeded to a vote on the order in which the candidates would be presented. In Vicenza, the procedure for electing the podestà first stipulated that twenty men from the Grand Council were to be randomly selected, five from each neighbourhood; out of these twenty men, the Grand Council then had to appoint the eight electores who were to propose three candidates for the podestà to the Council of the Four Hundred (which was a different body). The Council of the Four Hundred then voted again to determine the order of presentation.

21In sources from the end of the thirteenth century, and especially those from the fourteenth century, the specifics become increasingly clear. [34] An especially detailed description can be found in Treviso’s statutes from 1313, which specify the procedure to be followed for the election of the podestà: they are complemented by the publication of an account of the 1313 election that explicitly states that the procedure took place in conformity with the statutes. According to that account, three months before the end of his mandate, the podestà was required to summon the Council of the Three Hundred, which was in charge of selecting eight of its men, four from the nobility and four from the Popolo, all aged 30 or older. Isolated in the city’s Communal Palace, the eight men were to designate twelve electors, six for each of the two statutory bodies. Still respecting the same parity, these twelve men in turn had to select the true electors of the podestà, which the existing podestà then led to an appropriate location for the conclave. Only at that time could the meeting of the Council adjourn. The electors were to receive only one meal per day, and no one was allowed to speak to them, with the exception of the servants bringing in the meals. When the electoral committee had agreed upon a list of three candidates that was then registered by the podestà’s notary, the podestà was to summon the Council again. Each member of the Council was to deposit his ballot in one of the three urns carried by the podestà’s servants. The podestà had to list the candidates aloud and point out the urn that corresponded to each name. During the voting process, each Council member had to show his bulletin in his open hand, fingers spread wide, before tossing it into the urn of his choice. The candidate receiving the least number of votes was placed in third position on the list; the same procedure was then followed in another round to determine the first and second places.

22One account from 1315 describes how the podestà, originally from Gubbio, convoked the Council’s session. The meeting began with the reading of the statutes concerning the election of the podestà, which were then explained and translated from Latin into the vernacular (volgare). Finally, the podestà had his notary choose the first eight electors, whose names, here as in the subsequent rounds of the election, were recorded in the official proceedings. The eight men first had to take the electoral oath, after which they were placed outside of the Council, in the Communal Palace’s chapel where they elected the twelve electors for the next round “unanimously and by consensus, without dissent”. The latter then swore an oath before the Council and, once again in the chapel, designated the eight electors of the podestà; according to the official record, this occurred “unanimously and by consensus” (“unanimes et concordes”), with the exception of the sole dissenting voice of Sir Uguçono de Pagano. The electors swore on the Holy Gospel that they would proceed to a legitimate election in conformity with Treviso’s statutes and, still in the same chapel, rapidly reached agreement without dissent on a list of three names, recorded by the notary in a brief. This explains how voting on the order of the names could also occur during the same Council session. The podestà designated the white urn as corresponding to the first name, in alphabetical order, the red urn as corresponding to the second name, and the blue urn as corresponding to the third and final name. With 86 votes, a certain Monaldus de Castelduranto came in third. Petrus de la Brancha from Gubbio and Poncinus Ponzonus from Cremona obtained 91 and 109 votes, respectively, in the first round; Petrus de la Brancha ultimately won with 153 votes, compared to Poncinus Ponzonus’ 133. The future podestà belonged to the same family as the sitting podestà. The result was immediately proclaimed before the people’s general assembly, which had been summoned by the ringing of the church bells. Then the Council convened a second time, and its 286 members unanimously selected Petrus de Valle, a notary, as their proxy in their dealings with the future podestà.

The emergence of majority voting

23In the context of this article, it is not necessary to analyse other variations on these kinds of selection procedures. With regard to the election of the podestà, it would be interesting to examine if direct decision-making by a select electoral commission actually evolved into a two-step procedure, with the commission elaborating a list of names and the Council then determining the order of these names. A more in-depth study should likewise determine if the Council was bound by the electoral commission’s proposal, only able to modify the order of the names, or if, as was the case in Siena, the Council could reject certain propositions and impose the nomination of other candidates. Naturally, there were still other possibilities, as shown by the example of the procedure adopted in Brescia in 1281, where the podestà’s election was entrusted to a committee of fifty electors that had to obtain at least a two-thirds majority. For the doge’s election in Venice – which was the process most similar to the election of the podestà – majority decisions became the norm at the beginning of the thirteenth century. In 1229, however, an electoral impasse required the use of sortition; for subsequent elections of the doge the number of electors was thus increased to forty-one. It was only later, with the closing of the Maggior Consiglio, that qualified majority decisions would become a requirement.

24For concrete decisions, Italian communes generally began to adopt the majority principle during the second third of the twelfth century. In the urban leagues, often also ruled collegially, and in other forms of inter-communal organisations, the majority principle was often accompanied by a host of additional safeguards, ranging from blocking minorities to the creation of veto power; many began to reflect upon the limits of the principle’s application. [35] But even though the pope’s election by qualified majority voting was recognised as early as 1178, [36] adapting the majority principle to communal electoral procedures seems to have been an uneven process. The election of the elders of Bologna’s guilds and societates armorum provides a counterpoint to the examples above: starting in the middle of the thirteenth century, the guilds and societies would pool together all of their members and randomly draw nine electors, at least two thirds of whom had to agree upon the final nominee. From the latter half of the thirteenth century and especially during the fourteenth century, majority decisions subsequently played a particularly important role in two situations. First, they were used when an electoral committee’s proposal had to be presented in a specific order (as was the case for the podestà’s election); or when electors had to give the Council twice the number of candidates required, as was the case for many minor offices. When the list contained only four or six names, two or three of which had to be selected, electors then turned to multiround voting (as was the case for determining the order of the list for the podestà), in which the candidates who received the fewest votes were eliminated round by round until only the required number of candidates remained. Second, majority voting was also applied for establishing lists of possible candidates for certain specific offices, the final selection occurring via sortition. The most famous examples of this process were found in Florence and Siena during the first half of the fourteenth century. Committees were in charge of drawing up lists of eligible citizens. The committee then voted on each of the names; only those which obtained a qualified majority were then included in the lottery. During these procedures, the list of names could be whittled down to one sixth or even one tenth of its original size. [37]

25One notable point is that even when some kind of majority vote occurred, even if this was confined to a small committee of electors, the majority principle does not ever appear to have been based on the principle of the “wisest part” (“sanior pars”) making decisions in electoral commissions. [38] When majority voting was combined with the random selection of a small number of electors from a vast assembly (as was the case in Bologna, described above), [39] the debate could certainly rage on for quite some time before it was determined who constituted the sanius concilium. This thus also marks a notable difference with the majority decisions of a Council composed of those who were deemed the best and the wisest. [40] Nonetheless, we should not overlook the reason proffered by Marsilius of Padua to justify the imposition of the will of the valentior pars on the community, which he defined as a sort of “strong majority”: [41] in fact, there would always be a few men who, whether out of spite or stupidity, would refuse to support the consensus. [42]

Sortition and minor offices (thirteenth and fourteenth centuries)

27“Elections by lottery” appear initially to have been used for positions of minor importance, and which had previously been allocated directly by the consuls or the podestà. [43] While the method’s use can be traced throughout the thirteenth century, the provisions that stipulated the use of sortition specifically for minor offices rarely mention pre-selection processes, nor what form the latter might have taken. For Florence and Siena, we are familiar with the committees that subsequently, over the course of the fourteenth century, also determined who could be elected to the commune’s highest offices. Nonetheless, regulations of this sort can be found as early as the beginning of the thirteenth century. Whenever the random selection of officeholders was not limited to the members of an elected Council, lists of qualified and admissible citizens were established.

28It is possible that this evolution began within a group which, due to its formation and occupation, had access to equivalent qualifications, and for whose members being appointed to a communal office represented a material advantage over other guild members, especially the notaries. Treviso’s statutes, revised between 1231 and 1233 but evidently expanding upon older regulations designed prior to 1224, stipulated that all the notaries and statute drafters employed by the commune (traditionally for short-term mandates) were to be elected per rodulum et per sortem [44] so that the commune’s notary positions could fall to everyone, “individually and as a whole”. A list existed of all the notaries who did not reject a communal position of their own volition, and of those who were not excluded by a committee formed to that end. Using an unknown procedure, two judges, two knights, and two notaries were to be nominated to this committee, who had to swear in good faith, putting aside their likes and dislikes as well as all matters of personal interest, that they would impartially examine the candidates for communal posts. A judge from the podestà and a grammar-school teacher were to advise the committee; no candidate could be approved as having been “examined and qualified” without the committee’s vote. The provisions of the “election by lottery” method stated that the entire list had to be exhausted before a notary could be allowed to participate in the lottery a second time. With regard to procedural points, the communal statutes from 1263 refer back to the statutes of the notary guild, which highlights the fact that internal agreements within the guild perhaps also played a role. During the lottery, first the total number of notaries needed was drawn from the rodulus, then the institution for which they would work was randomly selected. However, here as well qualification criteria were quickly added to the procedure. In 1222-1223, podestà Marino Dandolo decreed that the podestà could choose – from the 31 notaries randomly selected for the given period – the four men that he believed were the most qualified to work with him as archivists (tabularius, pl. tabularii), before the other positions were filled by lottery. According to the existing documentation, this method appears to have an older source, as the only new element commented upon is the fact that two judges from the podestà had the right to choose the two best notaries from the group randomly selected.

29This provision regarding notaries seems to have served as the inspiration for podestà Odolricus de Beseno’s efforts at improving the organisation of public offices in 1224. He invoked the help of God for such a sensitive task, and employed the advice of the two judges and knight from his familia. He sorted all the “laymen” qualified for the positions into three groups of different rank (gradus): the term “laymen” designated all those who were neither judges nor notaries on the register. The highest-ranking group was composed of candidates for the positions of consul, procurator, property wealth surveyor, and head archivist. The second group included the consuls in charge of the rural agglomerations under communal rule, the two administrators of the commune’s heritage, the jurors and those responsible for estimating the cost of the urban cavalry (militia), overseeing and maintaining public assets, roads, squares and pipelines, and preventing damage from private use. The lowest group included commune representatives in the villages, those in charge of the scales used to weigh foodstuffs and of the weights and measures necessary to do this. Selection occurred by reducing the total number of individuals on each list to the number of officeholders required for six months (i.e., 12 for the first group, 14 for the second and 31 for the third). This then created three sub-groups that corresponded to the hierarchy of the different offices to be filled. Within these groups of 12, 14 and 31 randomly selected candidates, the distribution of mandates once again required a complex process of sortition. There again, the modalities were subsequently changed: before the other offices were filled, the podestà had to select the tabularius and the treasurer from the first group, the two masarii from the second group, and the three scale supervisors for meat and cheese from the third.

30No information is available, at least from the city’s statutes, regarding the criteria that governed assignment to one of the three groups. Nonetheless, it is evident that such criteria must have existed, and that they stemmed primarily from social status – in particular social status as defined by membership of a certain class. Many cases illustrate that possession of taxable assets worth a certain minimum value was a necessary condition for eligibility for certain offices. We can likewise infer this requirement by the personal assets that such officeholders were obligated to pledge for the omissions or damages committed under their remit, or by the guarantor they had to nominate for such responsibilities. The controversy that raged during the fourteenth century in Florence and in Siena between the major and minor guilds regarding the possibility of eligibility for various positions is well documented. Similar debates were also waged in other Italian cities. The limits to eligibility have hitherto not been the subject of in-depth analysis, having only been mentioned in monographs on the internal controversies of a specific city. It is nevertheless evident that without a detailed analysis of the prerequisites for election to a given office, the logic of the political system on which the nomination process was based cannot be fully understood. From this point of view, the present article likewise remains incomplete.

Sortition for the highest offices (fourteenth century)

31For those familiar with the subject, it is easy to observe that rules such as those adopted in Treviso in 1224 led directly to the imborsazioni of the fourteenth century. [45] In Treviso itself, this kind of system is clearly attested to from the beginning of the fourteenth century; nonetheless, it remains impossible to identify an uninterrupted progression from the rules governing the appointment of individuals to minor offices to city governance, if only because a Signoria had been established in the meantime. [46] According to the 1313 statutes, the podestà and the curia (the consuls and the elders) selected eight citizens every six months – i.e., one neighbourhood representative from each of the higher status groups and a representative of the Popolo – who were responsible for establishing or revising the list of candidates eligible for sortition. At fixed times, all those who believed they were qualified could present themselves before the committee and ask to be placed on the list. In 1314, this practice, which had been rapidly treated in 1313 due to its familiarity, was modified and supplemented by with a number of detailed rules. A special 120 paragraph-long statute on the offices and their attribution replaced a single statement. The committee in charge of establishing the lists (adoptio rodulorum) would no longer be formed by the podestà and its curia. Instead, eight electors were randomly selected from the Council, who then met in conclave with the podestà and his representatives and elected the eight notaries (rodularii) chosen to hold a long-term mandate. The latter were entrusted with the distribution of ballots in the lottery purses, and charging with establishing copulae, or entire sections, that were randomly selected en bloc for certain positions. The procedure’s transformation from 1313 to 1314 can certainly be explained by the fact that, unlike the period that preceded the Signoria, the members of the podestà’s curia, and thus the members of the commune’s government, were now also selected by lottery. The system of indirect election prevented the sitting curia from directly appointing the individuals that would be in charge of choosing candidates for the following election.

32As the example of Treviso demonstrates, the introduction of sortition was evidently accompanied quite early on by the establishment of electoral lists, for which small committees remained the primary decision-makers. The procedure’s norms appear to have changed during the first half of the thirteenth century; Pisa’s history seems to indicate that this transformation started with the allocation of mandates for judges and notaries. The application of such “electoral procedures” to the highest government offices, as it appears to have emerged c.1300, can also be explained by a desire to resolve internal tensions, as well as by debate surrounding the fair attribution of public offices. In the case of Florence and Siena, such electoral procedures have been studied from this perspective.

33Without going into the details, the principle according to which government offices were attributed in the two aforementioned cities should at least be mentioned here. Committees, whose composition and selection were subject to highly detailed regulation, established the lists of eligible citizens, on which a special committee or a Council subsequently voted, candidate by candidate if necessary. A high quorum considerably reduced the number of possible candidates. In Siena, the scrutineers (accoppiatori) then established groups of nine men who were elected en bloc for the duration of the mandate, to lead the government alongside the podestà and the capitano del popolo. [47] Here as well, the trend of giving all qualified candidates an equal chance (to the extent that this was possible) prevailed; this could be seen in Florence, where it was only later that procedures were developed to allow existing officeholders who had proved their worth to participate in the lottery once again before all of the initial names had been drawn. The successors of former electors could sometimes be found in the ranks of the selection committees or the accoppiatori, but they no longer directly chose candidates that were then imposed on the community; they were instead presented as trustworthy representatives that guaranteed the qualification of candidates and the favourable composition of collective governing bodies. Here, undoubtedly, was a decisive shift in terms of how the political community was conceived, as well as the function and meaning of “elections” for the community.

34Of course, the regulations adopted in Florence and Siena were in response to the dominant trend, since the fourteenth century, of providing officeholders at the end of their mandates with a certain input regarding the choice of their successors. According to statutes from 1310, the Nove governatori e difenditori del Comune e del popolo di Siena were to be elected by a committee to which the nine sitting governors belonged, in addition to the podestà and the capitano del popolo, as well as at least three consuls from the merchants’ guild. During the scrutinium, [48] all candidates had to obtain at least ten out of fourteen votes. As early as 1318, however, candidacies had to be approved by a majority in the Legislative Council, which could reject the nominations provided and propose new candidates. The tendency to try to reach a broad consensus, at least within the dominant group, was clearly observable, even before it was expressed through electoral lists and the lottery system. In Florence during the same period, the Signoria was likewise granted full powers to designate its successors, and sometimes even the collegial bodies that would follow them for several mandates. In 1318, members of the Signoria were entrusted with choice of operating elections themselves, or delegating this to an electoral committee. [49] In 1328, new reforms put an end to the priors’ supremacy. The latter had quickly figured out how to pre-determine the new members of the Signoria for several years at a time. In this manner, they guaranteed the continued influence of their own group, but also a broader support base, since many men and families were promised a future communal position in this fashion.

35An analysis of these fourteenth century transformations would no doubt provide interesting context, but would unfortunately divert us from our main question regarding the evolution of electoral modalities. In summary, some final remarks will briefly highlight certain elements that might be misconstrued from a modern perspective, if the latter failed to take into account the objectives of different electoral systems, rooted as they were in specific conceptions of community.

The case of Florence

36Florentine debates regarding the various electoral procedures discussed above allow me to underscore a crucial point. On 24 November 1292, the constant debate regarding the election of priors continued to rage in the Council of the Guilds: [50] 23 different procedural proposals were made. The issue at stake was whether the twelve major guilds should on principle be seen as having equal rights during elections, or if – as had been the case until then – a limited number of corporations (the bankers, lawyers and principal merchants) should essentially be tasked with selecting the priors. From this perspective, the crucial question was to determine whether, during the nomination process, power lay primarily with the sitting priors or if, in some fashion or other, it should be delegated to the guild leaders. Independently of all political controversy, however, multi-level selection procedures were envisioned, wherein sortition and secret ballots were designed to complement each other as the need arose. In addition to representing the guilds, such elections were also supposed to represent equally the city’s six districts. The electoral system that was chosen after lengthy debate was very similar to that proposed by the chronicler Dino Compagni. The directorate of each guild was charged with designating a candidate for each of the six neighbourhoods, in such a way that the prior for each neighbourhood would always be chosen from the twelve guild candidates. For each of the six neighbourhoods, the election had its own round; after each round, the rule stipulating that two priors could not belong to the same guild reduced the number of candidates for the following round. The Council selected the representative from each neighbourhood by a majority vote, but the members of the Council belonging to the neighbourhood in question did not vote. The provision excluding members of the Council from voting in the election of a prior from their neighbourhood had itself been voted through by an overwhelming majority of 80 votes to 7. In the debates, almost all of the proposals had included this provision or a similar measure, such as having the representatives of one of the six neighbourhoods elected by Council members belonging to another neighbourhood. In proposals where the election of the priors was the responsibility of the guilds, a guild was not allowed to vote when it was time to elect a representative from its own members. There was a broad consensus across all political parties regarding the principle that members of a group should not elect their own representatives. Within the committee responsible for Florence’s administration, no one wished to see the mandated representative of a particular group or party, but rather a man who everyone – and especially the members of different groups – could accept. Nonetheless, all representatives ultimately had to belong to one of the guilds.

37Dino Compagni, frequently described as a staunch partisan of guild republicanism, explained the intentions that guided him at the time. [51] In 1282, when the Government of the Fourteen, which had been formed through a compromise between the Guelphs and the Ghibellines, threatened to crumble, a still-young Compagni had met up with five other members of the Popolo (collectively known as popolani) to demand that three priors and the guild leaders be added to the Government of the Fourteen, as a sort of permanent means of protecting the guild population. In the beginning, the new institution had been met with great enthusiasm in Florence: the Popolo was heartened, and new decrees were passed that were designed to ensure justice and freedom. But in just a few short years, such hopes were utterly dashed. Compagni writes:

38

“In fact, their legal mandates consisted of preserving the Commune’s assets and [ensuring that] the communal government protected the rights of everyone and prevented the poor from being oppressed by the rich and powerful. As long as these principles were respected, they were of great benefit to the Popolo; but the situation was quickly reversed, as the efforts of citizens occupying public offices were no longer directed towards enforcing these laws, but to corrupting them. If one of their friends or family members incurred a penalty, they worked things out with the communal government so that the crime went unnoticed and thus unpunished. And these citizens did not merely neglect to protect the city’s assets: they found better ways to pillage them. This is how they diverted huge sums of money from the Treasury under the pretext of repaying men who had provided services. Far from coming to the aid of the poor, the dignitaries consistently trampled their rights, as did the popolani grassi in office, who allied themselves with the nobles. Many among them were protected by their money from the communal sanctions that they incurred. This is why the honest citizens of the Popolo were discontented; this is why they despised the rule of the priors, because the dignitaries of the Guelph faction were in fact the masters of the city.” [52]

39This formed the backdrop for the electoral debates of 1292, in which the people tried to find the best way of electing a good priorate; i.e., a good government that protected the people from the arbitrary whims of the nobles and the city from the domination of a clique.

Sortition as a means of combating factional interests

40In Florence, the political goal was therefore to entrust city governance to guild representatives, which inevitably meant shifting the political equilibrium (an endeavour heartily supported by the guilds). However, as the minutes of various sessions illustrate, the proposals relied on the arguments that those who had hitherto governed the city had not fulfilled their mission in service to the community. The guilds hoped to find in their midst citizens who would not try to rise above the rest and would guarantee justice, including for the powerless. However, these hopes were likewise dashed. Expanding the participation of the Popolo in government was frequently touted as the solution that would finally bring disinterested men into government and administrative offices.

41When in Siena in 1368 the riformati recruited from the minor guilds and from amongst the salaried workers toppled the Government of the Twelve and instated a coalition government composed of the Party of the Nine, which had governed until 1355, the Party of the Twelve, and the reformers’ own group, [53] one of their own members described it thusly:

42

“Know that the riformati, in their field, are the best craftsmen in the world: they hadn’t come there to steal anything; out of 100 of them, 99 would die as beggars. They always showed that they were loyal to the Commune, and the enemies of the nobles and the violent. They would tolerate no arrogance in Siena; they were esteemed by the nobles and the Twelve with a mixture of fear and respect. Know that it was more honourable to say, ‘I am one of the reformers’ than ‘I am one of the lords’.” [54]

43However, such expansion in participation and other large-scale changes did not, in the long term, lead to a situation in which a group assigned its representatives, independently from the community, to the commune’s governing body, the Council, or other offices. In Florence, people wanted to elect guild representatives as priors because they wanted a better and fairer form of government, but corporate representatives in the modern sense were to be avoided at all cost – i.e., representatives that the guilds would have directly chosen. [55] Even if, in Bologna, elders were directly selected from within the guilds, and societates armorum to perform various functions for the Popolo, their election by grand electors who were randomly appointed by lottery at least guaranteed that the wishes of the guild’s leaders or majority were not expressed directly during the selection. By taking into account the electoral regulations of the city’s neighbourhoods, guilds and other constitutionally recognised societates, in addition to various combinations of the above, a wide selection of possible candidates would emerge, creating an election that was impossible to predict ahead of time and thus very difficult to manipulate; the way in which nominations were ordered likewise added new obstacles from this perspective. The candidate recruitment base and the nomination method were not designed to appoint representatives mandated by the ruling parties; on the contrary, they aimed to prevent special interests from being expressed in elections. Using one’s position (or having used it) to favour the interests of a specific group or family was the worst accusation that could be levelled against an officeholder. [56]

44Let us simply note in passing that the electoral systems described above, as well as their corresponding conception of elections and community, were entirely compatible with the growing rigidity of public government and did not, in principle, contradict the Renaissance’s shift towards Signoria [57] governments and principalities. Consequently, they make no sense in terms of our modern opposition between “democracy” and “dictatorship”. [58] In the ideological debates that raged between Florence and Milan on the subject of liberty and tyranny, and on the monarchical legal order and the arbitrariness of oligarchy or ochlocracy, [59] the issue of electing a government with a restricted mandate or of recognising a hereditary government had culminated in the question of the state and the form of domination in the Aristotelian sense. The electoral systems developed in the Italian communes ended up being entirely compatible with the conception of an impartial authority that was devoted to the community and the common good, once they were stripped of their original function of controlling “from the bottom up”. Under the enlightened despot Leopold II, Holy Roman Emperor, [60] during the allocation of positions, sortition still played an important role as an “electoral procedure”. It was only with the French Revolution that new conceptions arose of the political community and of the individual’s relationship to the whole, thanks to which electoral forms and the notion of voting also obtained a new status in the process of establishing the political community.

45***

46Given our current state of knowledge – that of research in general, and my own in particular – what I hope to highlight in this conclusion from a comparative perspective must necessarily be contingent on further analysis. Taking into account the frequency that characterised not only elections, but also the countless debates on voting methods and principles, there is no doubt that these political experiments were present in everything that was written or discussed in Italy at the time on the subject of voting and elections. The introduction of majority decisions, which were frequently applied even before 1150 in communal collegial bodies for votes on factual matters – even though the majority principle was in contradiction with the existing conceptions of voting and community – as well as the use of the conclave as an electoral forum, lead us to believe that a close relationship existed between these political experiments and the regulations of ecclesiastical law. However, I do not wish to present the communal origins of modern electoral methods as a means to contradict Léo Moulin’s argument that these roots were religious and ecclesiastical. [61] In fact, the reference to modernity that stems from the issue of the emergence (and principles) of majority voting only prevents us from truly understanding the plurality of medieval electoral systems and the role these played in the attempts to organise communal politics along peaceful lines. The “election” of an individual to an office or a Council was not to be guided by the representation of special interest groups or a certain party’s mandate. While the reality naturally sometimes deviated from this ideal, such departures were vehemently criticised, with charges equally levelled at those who betrayed their electoral mandates and those who were responsible for electing them. An element of contingency in the multi-step decision-making process, often due to the use of sortition, was gradually introduced to the delegation of electoral capacity. This was evidently how the community believed that it could best guarantee impartiality and overcome selfish private interests.

47Ultimately, medieval electoral systems did not transcend structures wherein a small group of plenipotentiaries voted for the community as a whole, or pre-determined voting results thanks to important prior decisions. These were essential characteristics that fundamentally differentiated the Popolo’s medieval forms of communal sovereignty from modern liberal democracies. First of all, expanding the base of the Popolo’s regime did not come about via granting new groups voting rights: the idea of universal suffrage never truly took shape, despite the fact that there was constant discussion regarding the notion of “the people’s vote”. On the contrary, the circle of those who voted remained shockingly small. A tiny number of individuals were in charge of voting on the community’s behalf; the same was true regarding decisions that left Councils with relatively little leeway for action. What has been analysed as the growing “democratisation” of commune life was the result of considerable expansion in the circle of potential electoral candidates. [62] Given the large number of offices and the significant scope of the Councils, the fact that officeholders of any mandate, no matter how short, had to wait one, two, three or even four years before being eligible for re-election meant that a significant percentage of citizens acquired the possibility of participating in the commune’s politics and administration. What the majority of mandates offered was not so much the possibility of wielding political influence – which was better obtained through election to one of the legislative councils – but rather duties, honours, the possibility of gains and benefits, but also obligations, and first and foremost the ability to try to prevent others from obtaining unjustified advantages.

48There is no sign of a transition towards “original elections” as the expression of the general will, even in its fundamental premises, and nowhere did the formalisation of electoral procedures entail linking voting results to the general will. The goal of elections, which consisted in reaching a decision for the benefit and utility of the community, was primarily dictated by internal criteria: the obligation of electing the best and most useful individual(s), selecting from only fair and impartial candidates. Legal procedures, both highly electorally diverse and constantly changing, were designed to exclude partisanship, the dominance of cliques, and overweening ambition – in other words, the fundamental ills that threatened the goal of communal harmony. Communities strove to perfect the selection procedures to nominate those who were to act in their best interest. However, the community was not viewed as a sum of individuals whose individual wishes were to be taken into account in the election process. Decisions were the purview of the universitas as such, which delegated true decision-making prerogatives to a small number of citizens, in a process that was often determined on a case-by-case basis and which frequently included a certain element of chance.

49There certainly remains room for debate regarding to what extent such electoral systems prevented dominance by various factions – which was rejected in theory – from ever occurring in practice in the Italian communes. Likewise, the question remains: to what extent did the existing social structures hinder the development of “more democratic” electoral systems? However, the reality of the era can only be understood once we determine both what the objectives were that governed the development of such highly complex electoral systems, and what limits were imposed on human attempts to organise their societies and communities by conceptions which, at the time, were seen as universally valid.