The research conducted in this publication was funded by the Irish Research Council within the Euronews Project. In 2020, the Euronews project’s team, of which the author is part, conducted a sub-project entitled “1600 Experiment,” whose results are forthcoming. The author thanks Professor Brendan Dooley, Dr. Carlotta Paltrinieri, Dr. Davide Boerio, Lorenzo Allori and Wouter Kreuze for their work in transcribing the newsletters selected for this collaborative sub-project and creating the corpus, on which the author conducted the quantitative analysis subject of this publication. Finally, the author thanks the anonymous reviewers for their careful reading of the manuscript and constructive remarks.

1On Sunday, January 17, in the Viennese district of Leopoldstadt, an atrocity occurred: a butcher, his wife and their thirteen-year-old son were found dead in bed, having been brutally murdered and robbed of their belongings. Rigorous searches were carried out to discover the perpetrators of the crime. Such a tragic episode might be found in today’s printed or online newspapers, but instead this was reported in a handwritten newsletter from Vienna, dated January 23, 1700 (“Vienna 23 Gennaio 1700”). About two months later, in a newsletter dated March 13, the story was taken up again to announce the imprisonment of the Leopoldstadt family’s assassin (“Vienna 13 Marzo 1700”). The following week’s newsletter recounts the epilogue : the 24-year-old Georg Wagner confessed to having committed the triple murder. Although he only stole two florins and a few rings, robbery was the motive, and on the 16th, he was executed (“Vienna 20 Marzo 1700”). To satisfy the curiosity of a broad audience, Wagner’s death sentence was printed in a six-page pamphlet, giving the most macabre details of the murders (Urtheil So dem Hanß Georg Wagner), as was often the case for sensational crimes, acts of violence and executions in the early modern period. The interest aroused by this crime is also proven by its repeated mention in the handwritten newsletters, which were intended to circulate far beyond Vienna. As proof of their wide circulation, one does not need to consult Viennese documents or newspapers to know about this case but can just read the contemporary newsletters written in Italian that reached Florence at the time and are today kept at the Florentine State Archive.

2This case draws attention to the aspects of continuity and change across time in event narration, considering that the necessity to know and communicate the news has basically remained the same, while the media have changed radically. This contribution will explore the development and production of one news medium, the handwritten newsletters, also called avvisi in Italian, between the beginning and the end of the seventeenth century. This century witnessed the spread of the first printed newspapers, first in Strasbourg and Germany (Schröder 123-27), then in the following decades in the rest of Europe (Pettegree 182-90). However, the handwritten newsletters continued to circulate in parallel with the printed sheets until the late eighteenth century. In their early days, the newspapers seemed like nothing more than printed copies of the handwritten newsletters, following both the layout and the content of the latter (Pettegree 183). But in a short time, the printed genre developed special characteristics, due to the greater circulation possibilities and the increasing demand for political information, particularly from the Thirty Years’ War onwards (Weber, “The Early German Newspaper” 72-75). This article aims to investigate how the handwritten newsletter genre adapted and changed over time between the beginning and the end of the seventeenth century, applying a quantitate approach and digital methods to the study of seventeenth-century handwritten newsletters. The benefits of applying digital methods lie in the possibility of studying these primary textual sources in new ways, not only close-reading them but extracting and visualizing data contained in them. This approach falls within the Digital Humanities field, which aims to formulate new research questions, which are only answerable through computational tools and methods (Hughes et al. 152-53).

3This investigation takes the form of a case study of two corpora of avvisi, selected from the Medici grand ducal archive, housed today at the Archivio di Stato in Florence (Archivio Mediceo del Principato V-XXXIII ; Bellinazzi and Lamioni LI-LXXV). The Medici collection of avvisi, consisting of more than two hundred volumes, is remarkable for its temporal and geographical breadth since it covers the period between the mid-sixteenth and the mid-eighteenth century. It gathers newsletters from all the most important European hubs of the time, including Rome, Venice, Milan, Vienna, Hamburg, Cologne, The Hague, Brussels, Paris, London, and Madrid. It is the result of the information network created by Cosimo I de’ Medici and continued by his successors, made up of diplomats, secretaries, official and unofficial agents, spies, postmasters, and occasional informers (Barker 722-27). The literature on early modern diplomacy and diplomatic practice has emphasized the enormous importance of information, news, and newsletters for a state’s situation at home and abroad. News from foreign courts was the object of a political trade, and, especially for small states such as the Grand Duchy of Tuscany, remaining up-to-date was the key to playing a leading role in Europe (Frigo 133-34). The study was carried out on this collection because of its vast extent, which allows comparison between avvisi from different periods and places in order to identify the aspects of continuity and change.

5The 1600 corpus will be compared to a corpus comprising all the avvisi of the year 1700, in order to understand how the circulation of news, the form, and the content of the handwritten newsletters had changed. The time span of a century has been considered long enough for any new phenomena or processes that developed into the avvisi genre to be detected. The seventeenth century also witnessed the advent and success of weekly printed newspapers, a new medium in the existing system of information. In the Italian peninsula, newspapers began to be produced much later than in the rest of Europe : the first surviving copy was issued in Genoa in 1639 ; subsequently, newspapers began in other Italian cities, including Rome in 1646 and Venice in 1661 (Bellocchi 3 : 24-34). Therefore, 1700 was identified as a good term of comparison to highlight elements of continuity as well as the new trends that may have arisen in handwritten newsletters over the century.

6Although this study concentrates on handwritten newsletters, it is important to note that the Medici court also received printed sheets of news. In the seventeenth century, handwritten newsletters and printed newspapers were complementary and equally integrated into the information system (Scarborough King 434-35). This can be seen in a letter sent to the Grand Duke of Tuscany by Giannettino Giustiniani, a patrician of Genoa and informer of the Medici:

• 2 “Your Illustrious Lordship will read almost all the news that arrives here in the sheets of avvisi (...)

Già la Vostra Signoria Illustrissima dalli fogli d’avvisi che si sono introdotti di mandare alla stampa in questa nostra città la deve intendere quasi tutto ciò che vi arriva di nuovo. Anderò io per l’avenire più ristretto e conciso in annoiarla per quanto non sii solito di vedere detti fogli, ma essendo la prattica delle nuove assai qui da noi generalmente communicabile mi persuado che poco lascino adietro, quando la passione non gli facci alterare li fatti.2 (Morelli 308-9 ; Giustiniani 23 September 1639)

The letter was sent from Genoa in September 1639, just two months after the oldest example of Italian printed gazettes, dated July 29 (Neri 300-1). The periodical newspapers may have begun earlier since this is the first surviving copy : as Bellettini states, only a national census and in-depth archival research will shed light on the first journalistic activity in seventeenth-century Italy (Bellettini 466). However, Giustiniani’s letter attests that the Medici were early aware of the printed newspapers and received them soon, since the informer invited the Grand Duke to read the news from Genoa in the printed sheets. It is also notable that he defined them as “sheets of avvisi that have been sent to the press,” making it clear that the formal continuity between handwritten and printed newsletters was also perceived by contemporaries, even at the level of content. The means of production were changed, but both were referred to as sheets of avvisi.

9Many works on Italian handwritten newsletters have focused on the birth of the genre and its development during the sixteenth century and early seventeenth century (see, for instance, among others, Infelise, “From merchants’ letters to handwritten political avvisi ;” De Vivo 4-8 ; Molino, “Connected News”), while little attention has been given to how avvisi adapted and evolved during the seventeenth century. Moreover, handwritten newsletters have often been analyzed either at the level of production and circulation (Infelise, “Prima dei giornali”) or for the individual pieces of information they contain (Lenci), but few have been the attempts to adopt a quantitative approach to analyze whole collections of newsletters (Schobesberger, “Mapping the Fuggerzeitungen ;” Molino, “Beyond the Language Divide”). This contribution aims to address these gaps, moving the analysis into the seventeenth century and applying digital methods to compare two years of newsletters and point out the elements of continuity and novelty.

10This study focuses mainly on three elements: first, the hubs, where the news items were collected and which in the typical layout of the avvisi were written as centred headers, followed by the date; second, the places where the news items came from and where the reported events took place; third, the topics of the news items. Considering these three elements, a quantitative approach was employed to analyze the 1600 and 1700 corpora, followed by a qualitative analysis to gain a detailed understanding of some characteristics.

11The choice to compare only two years is a limitation of this article, such that any conclusions will need to be verified on a more varied sample of handwritten newsletters before being generalized. The archival nature of the sources constitutes another constraint. Although the Medici archival collection is well preserved, it has a peculiar history. The “Mediceo del Principato” collection was not originally created as an archive, but consisted of the documentary material kept at court, the result of the daily correspondence and work of the Grand Dukes and the grand-ducal family ; moreover, since the end of the seventeenth century, it suffered a fire as well as several relocations and rearrangements (Archivio Mediceo del Principato V-XXVIII). In consequence, it is not possible to assert with certainty that the 1600 and 1700 corpora comprise all the newsletters that arrived at the court : some may have been lost or preserved in volumes other than those indicated in the inventory. Symptomatically, the avvisi having Rome and Venice as hubs in the 1600 corpus are few, although these were two of the most prolific and important early modern news hubs because of their political role (Burke 389-390 ; Infelise, “Roman Avvisi” 212-14) and because of their positions as crucial nodes in the European postal network (Behringer 349-50 ; Schobesberger et al., “European Postal Networks” 28-30). From the mid-sixteenth century, residents and agents of the Medici sent weekly newsletters from these cities (Barker 723-24), and Roman and Venetian avvisi continued to be sent to the Medici court up to the fall of the ruling dynasty. However, in the 1600 dataset, only one comes from Rome, and two from Venice. This lack of documents, probably due to the vicissitudes of the archive, must be taken into consideration when evaluating these cities’ relative role in the European news network.

131600 and 1700 newsletters all share the same layout features, but they differ in length. The average length of the 1600 avvisi is two pages, and a single news item consists of 80 words on average. The number of words is similar in 1700, with 78 words per piece of news, although the newsletters are much longer, consisting in most cases of four pages. Furthermore, a standardization of hubs emerges as a typical feature of the newsletters written in 1700. In 1600, except in a few cases, a newsletter is composed of news coming only from one hub. In 1700, on the contrary, only half of the avvisi have a single hub, notably those from Rome, Venice, Paris, Madrid and Hamburg. The rest gather in the same document news from several hubs, which always appear in a standardized order. News items from Vienna and Warsaw are always combined in the same newsletter : Vienna appears as the first hub, with a date always ten days earlier than Warsaw. The same occurs with a series gathering news from London, The Hague, Hamburg, Brussels and Cologne : except for Hamburg, these cities appear exclusively in avvisi that have multiple hubs.

14In addition, it is worth noting that, limited to the years analyzed, these two multi-hub series are the only ones present in two copies within the Medici archive. The newsletters with Vienna and Warsaw as hubs are preserved both in a volume gathering newsletters from the Imperial Court (Mediceo del Principato, vol. 4579) and in a volume with letters and newsletters sent from Venice by the postmaster Matteo del Teglia (Mediceo del Principato, vol. 3049). Copies of the newsletters having London, The Hague, Hamburg, Brussels and Cologne as hubs are instead found in the volume from Venice just mentioned (Mediceo del Principato, vol. 3049) and in a volume labelled Avvisi di Londra (Mediceo del Principato, vol. 4191a). Although it is not always possible to identify who wrote and sent the newsletters to the Medici court, it is clear that the copies have two different origins. The 1700 dataset testifies that the European system of handwritten newsletters was still operating at the end of the seventeenth century and the beginning of the eighteenth century. Paola Molino describes it, in regard to the second half of the sixteenth century, as “periodical editions that circulated in several different versions and languages across geographical and political boundaries but that look very similar in form and content in the various existing collections” (“Connected News” 267). Not much has changed one century and a half later and the copies of the two series found in the 1700 dataset confirm the longevity of the European newsletters phenomenon : an articulated system of production, distribution and circulation of handwritten newsletters, that is comparable to periodical editions and printed newspapers because of their periodicity and constant structure from week to week. Copies of avvisi also show how they were no longer written by individual agents for their masters but had become commodities produced and distributed regularly to subscribers (Barbarics-Hermanik 166).

Table 1: distribution of news hubs in the 1600 dataset.

Graz	27 (23,5 %)		Venezia	2 (1,7 %)
Genova	18 (15,7 %)		Roma	1 (0,9 %)
Köln	18 (15,7 %)		Paris	1 (0,9 %)
Antwerpen	17 (14,8 %)		Amsterdam	1 (0,9 %)
Milano	14 (12,2 %)		Torino	1 (0,9 %)
Madrid	7 (6,1 %)		Praha	1 (0,9 %)
Bruxelles	3 (2,6 %)		Lyon	1 (0,9 %)
Wien	2 (1,7 %)		Oostende	1 (0,9 %)

16Concerning the hubs, the 1700 dataset shows the emergence of new information centres, such as Hamburg, London, Warsaw, The Hague, and the disappearance of others (see table 2). Comparing the hubs’ geographical distribution between the two years shows how the centres have moved towards Northern Europe (see fig. 1 and 2). In 1700 Italy played a peripheral role : the only two hubs in Italy remained Rome and Venice, while Genoa and Milan, quite important in 1600, completely lost their position. The same happened to Graz, which was replaced by the capital of the Habsburg Empire, Vienna. The 1700 dataset attests to the growth of Paris and Brussels, which took over the position of Antwerp as a city of power and news from Spanish Netherlands, while Cologne and Madrid turned out to be relevant centres both in 1600 and 1700. Finally, the seventeenth century witnesses the surge of Hamburg as an economic and information centre. Much attention has been devoted to the success of Hamburg in the German newspaper market from the first half of the seventeenth century onwards (Hillgärtner 38-39, 62-63). However, as the avvisi of the Medici collection attests, also the production of handwritten newsletters vividly grew. In 1600 no news came from or concerned Hamburg ; and also the Fuggerzeitungen collection, ranging between 1569 and 1605, does not list Hamburg among the most frequent places of dispatch (Schobesberger, “Mapping the Fuggerzeitungen” 222-23). After a century, however, Hamburg has become the most prolific hub of the dataset, appearing 132 times (see table 2) and confirming the coexistence of different news media in its early modern urban space.

Table 2: distribution of news hubs in the 1700 dataset

Hamburg	132 (16,8 %)		Venezia	35 (4,5 %)
Wien	86 (11 %)		Madrid	33 (4,2 %)
London	86 (11 %)		Paris	33 (4,2 %)
Bruxelles	85 (10,8 %)		Lyon	1 (0,1 %)
Warsaw	85 (10,8 %)		Zurich	1 (0,1 %)
‘s‑Gravenhage	85 (10,8 %)		Lisboa	1 (0,1 %)
Köln	82 (10,5 %)		Luzern	1 (0,1 %)
Roma	38 (4,9 %)

17The shift of the news hubs to the north denotes the emergence of new centres of newsletter production and a shift in the European balance. In 1600 many news items concerned the Holy Roman Empire and the vicissitudes of the Low Countries, but above all, the attention was on Italy. By contrast, in 1700, the great events and consequently the greatest flow of news came from Northern Europe. Even in a small Italian state, such as the Grand Duchy of Tuscany, much attention was devoted to what happened between the powers that were deciding the destiny of Europe and Italy, especially after the death of Charles II of Spain, which occurred in 1700.

18The article attempts now to analyze the newsletters at the level of single news items. For each news item, it was recorded where the event took place, which presumably corresponds to where the piece of news came from. However, the newsletters do not always indicate the precise city or place of the event, and in the absence of further details, only the region or country could be recorded. This explains the combination of cities, regions and countries at the centre of the following analysis, although efforts have been made to identify the precise location of events whenever possible. To visualize and explore the links between the hubs and the places whence the pieces of news arrive, the two datasets have been submitted to network analysis. This allows us to visualize the relationships (“edges”) between some entities (“nodes”) ; different indices, then, help to understand the centrality of a node within the network and in relation to other nodes (Weingart). As stated by Ruth Ahnert, these methods have not yet been applied to their full potential to study the news network, although many historians have used the word network to define the system of early modern communication (Ahnert 130-31).

19Whereas the early modern news system has often been represented geographically on maps (see, for instance, Schobesberger, “Mapping the Fuggerzeitungen” 223-26), network analysis provides a different and alternative view, based on the number of news items travelling between places, regardless of geographical distances. Ahnert stresses the difference between maps and networks, stating that “the construction and measurement of communication networks provide an alternative way of understanding the connectivity of places and people in ways that may help us to redefine what we think of as centres and peripheries” (132). Network analysis, therefore, disregards the spatial aspect of the map to focus on the relationships between nodes, which may be people, places or other entities. Two nodes may be very distant geographically but represented as being close in the network visualization because they have a strong connection with each other. In the force-directed graphs utilized in this article, the most connected nodes are in the centre while the less connected ones tend to be at the network’s periphery.

20Anhert focuses exclusively on social networks: the nodes are people, and the edges are the relationships between them (132). The case is different for the avvisi, whose network consists of the relationships between the hubs and the places where the events took place and the news came from. In this case, nodes are places and edges represent the transmission of news from the locations where the events occur, to the hub, i.e., the gathering place of news. A small case study by Renate Pieper adopts network analysis on avvisi to reconstruct the network through which news concerning the New World was transmitted via handwritten newsletters. It concentrates on the news route, and the network is built by joining two different aspects: the places mentioned in the newsletters and the circulation of the newsletters as documents between the places of compilation and the commissioners. Edges connect the place where the news was reported, the place where the newsletter was written, and the place where it was received (Pieper 506-11). The network visualizations produced in this article, on the other hand, are based only on the places mentioned in the avvisi, i.e., the hubs and locations of the news, without taking into account the routes taken by the documents after their compilation. This methodological choice is primarily determined by the intention to focus on the individual places from which the news items originate, an aspect not yet analyzed in-depth in regard to early modern communication. Secondly, the datasets are based on a single collection and the place where the newsletters were received is always Florence, thus the network would have been biased in this aspect.

22The main difference between the two graphs lies in their shape. The 1600 network does not present a single defined centre, but shows clusters of nodes grouped around the main nodes, which correspond to the hubs of the avvisi (fig. 3). One information cluster gathers around Graz and Prague ; the second one comprises Genoa and Milan, while the third cluster comprises Antwerp, Brussels and Cologne. Each cluster is quite distant from the other ones, having its own sphere of influence from where its hubs gather news, rarely shared with other clusters. The network is different in 1700 since the centre is more populated by nodes (fig. 4). Nodes at the periphery convey the news to a single hub ; many of those towards the centre of the graph, on the other hand, are shared between two or more hubs. It appears that more hubs collect news from the same places, a more limited phenomenon in 1600. For instance, news from India and the East Indies is found in newsletters from London, The Hague and Paris ; news from Holstein is reported in newsletters from Hamburg, The Hague, Paris and Vienna ; news from Istanbul is collected in avvisi of Vienna, Warsaw, Venice and Paris. Such examples show how a place no longer reports news only to the nearest hub, as happened in 1600, but to several hubs.

24Moreover, the interest of the 1700 newsletters and their readers goes beyond the European borders reaching America, Africa and the East Indies. Fifteen news items come from the Gulf of Darien, in Spanish Central America, reporting the last events of the Scottish attempt to settle a colony there, threatening the Spaniards’ supremacy. The items appear in newsletters from both London and Madrid, as one would expect, but also in avvisi from Paris. Christopher Storrs reflects on how the Scots’ incursion has only been partially analyzed by Scottish and English historians, who have ignored the role played by Spain, and has been entirely neglected by Spanish historians, who have concentrated only on Spanish decline (Storrs 5-8). In this case, handwritten newsletters seem to be a useful source for restoring the importance assigned to an event by contemporary media, since the Scottish unsuccessful attempt was widely discussed and reported at that time, both in the kingdoms directly involved and in others such as France. This is only one example of the many 1700 news items reporting about trade, colonies, clashes and piracy, from Jamaica, Puerto Rico, Mexico, the Virgin Islands, Colombia, Perù, Brazil, Morocco, Algeria, Tunisia, Madagascar, India, Indonesia and China. In 1600 there is only one news item mentioning a ship arriving from Brazil ; all the other news refers to the European sphere, including the five pieces of news from Istanbul. In conclusion, the 1700 dataset demonstrates increasing attention to global news, which goes beyond the borders of the state and the neighbouring kingdoms.

25The role of the news places within the newsletters can be measured through the weighted out-degree, which measures the number of outgoing edges of a node, ponderated by the weight of the edges. It thus indicates how many news items flow from one place and makes it possible to understand which places are the most covered by the avvisi. In the 1600 dataset, the places with the highest weighted out-degree are : (1) Genova, (2) Graz, (3) Milano, (4) Praha, (5) Bruxelles, (6) Espana, (7) Magyarorszag, (8) Wien, (9) Savoie, (10) Roma, (11) Flanders, (12) Transilvania, (13) Nagykanizsa, (14) Köln, (15) Madrid, (16) Antwerpen, (17) Piemonte, (18) Saluzzo, (19) Napoli, (20) Gent.

26The first five nodes are also hubs, meaning that most of the news reported in the newsletters came from the big centres where the newsletters were written. However, in addition to the big cities, newsletters gathered news from less central places, for example, peripheral regions where conflicts were taking place. The strong presence of news from Hungary, Transylvania, and Nagykanizsa, for instance, is linked to the Long Turkish War, fought in these years, and proves how much the rulers of Southern and Central Europe needed to be up-to-date about the Turkish threat. The top twenty nodes by the same measure in the 1700 dataset are : (1) Roma, (2) Wien, (3) Madrid, (4) Venezia, (5) London, (6) Paris, (7) Bruxelles, (8) ‘s‑Gravenhage, (9) Riga, (10) Warszawa, (11) Istanbul, (12) Liivimaa, (13) Calaisis, (14) Nederland, (15) France, (16) Neuf-Brisach, (17) Holstein, (18) Köln, (19) Tonning, (20) Scotland.

27As in 1600, in 1700 ten places out of twenty are not hubs, meaning that the balance between news coming from the centres and news from the peripheries remains relatively unchanged in the newsletters. The top three producers of news are Rome (weighted out-degree : 425), Vienna (weighted out-degree : 374) and Madrid (weighted out-degree : 300) : it shows that the seat of a court guarantees a substantial production of news.

28In both 1600 and 1700 some hubs remain outside the first twenty nodes by weighted out-degree, signifying that the newsletters produced in these centres report much more news from outside than about the city itself. The data concerning Venice are significant. In 1600 more news reached Venice (weighted in-degree : 26) than left the city (weighted out-degree : 4), although the sample is small because there are only two avvisi from this city. After a hundred years the situation has changed : in 1700 the newsletters from Venice gathered much news from outside (weighted in-degree : 420), but a considerable amount of news items also concerned Venice itself (weighted out-degree : 269). The data confirm Venice’s involution within the information trade at the end of the seventeenth century, shifting from being cosmopolitan to being provincial, becoming “a center of information about itself” (Burke 408). Something similar happened to Hamburg, which became a European centre for the dissemination of news from the beginning of the seventeenth century (Bellingradt 221). As mentioned above, Hamburg is the most prominent hub in the avvisi of the Medici collection in the year 1700. However, surprisingly, it is not in the top twenty places with the highest weighted out-degree. As a hub, it received mainly news items from abroad (weighted in-degree : 359), while few pieces of news about events happening in the city are reported (weighted out-degree : 27). Hamburg has a similar role to that of Venice in the year 1600, being more important as a centre of reception and dissemination of news coming from the Baltic Sea area than as a producer of news.

29Betweenness centrality is another relevant measure in network analysis. It tells how central a node is to the network’s organization, measuring how often a node appears on the shortest paths between any pairs of nodes in the network (R. Ahnert and S. Ahnert 12). The nodes with a high betweenness centrality score play a bridging role and are the most important in the flow between nodes. In 1600, the places with the highest betweenness are Milan, Genoa, Cologne, Graz, Brussels and Vienna ; their function as nodes of the early modern postal system certainly plays a role in their centrality (Schobesberger et al., “European Postal Networks” 28-30). It should also be noted that Rome, Venice and Madrid have a betweenness centrality score of zero. If no conclusions can be drawn for the first two cities, given the scarcity of data for the year 1600, Madrid turns out to play a marginal role within the network, more marginal than Brussels, Torino and Lyon. In the 1700 dataset, the most central nodes by betweenness centrality are Vienna, Paris, Madrid, Venice, Warsaw, Rome, The Hague, Hamburg and London : these centres hold the network together. Cologne and Brussels are the only two hubs without a high betweenness score, meaning that they reported the news but did not play a role in the flow of information : the absence of these two centres would not undermine the network, as would be the case without Vienna and Paris, which are crucial in connecting other places.

30Network analysis permits insight not only into the major information centres, but also into the large number of places from which the news items are collected. In the 1600 dataset, 92 nodes out of 177 occur only once, meaning that only one piece of news from these places is reported in the avvisi. In 1700, the nodes occurring once are 156 out of 441. In both years, a significant amount of news arrived from minor or peripheral places, uncovering a widespread network for the circulation of news. The data also reveal that there has been a slight decline in the number of places whence only one new item comes, from about half of places in 1600 to about a third in 1700. It suggests that with time the handwritten newsletters have become less focused on the peripheries and more focused on the centres, but only a comparison with other years can confirm this hypothesis.

32Between 1600 and 1700, the distribution of topics remains relatively stable, with Military, Politics and Diplomacy being the most prevalent topics in both years : these three topics together amount to 78 % of the news in 1600 and 69 % in 1700 (fig. 5). These percentages quantitatively convey the idea one gets from reading the handwritten newsletters and reflect their original function, which was to inform about all events affecting states. Whoever read the avvisi, in 1600 as well as in 1700, was mainly interested in knowing the political, military and diplomatic affairs occurring in Europe. This percentages correspond to what Petitjean states about a corpus of avvisi from Venice ranging between 1606 and 1615, 75/80 % of which cover political, administrative and military news (133-34). The focus on political and military matters is also found in early printed newspapers, with percentages very similar to those of the handwritten newsletters. Although Wilke uses slightly different categories in his thematic analysis of eighteenth-century German newspapers, he notes that 70 % of the news items concern military, political and diplomatic events, while 30 % are devoted to economic, cultural, religious events and natural catastrophes (68-69).

33With regard to the less prevalent topics, Government, which also includes crimes, rises from 7% in 1600 to 10% in 1700. From reading the 1700 avvisi, there appears to be a growing interest in small local news items, such as thefts and murders, which were reported in the newsletters but had no implications at the European level. In a newsletter from Brussels on April 16, 1700, for instance, a young man who appeared in Namur selling jewellery is physically described because it is believed that he could be the one who had committed a theft in a church in Cambrai a few days earlier (“Brusselles 16 detto”). On May 22, 1700, a newsletter from Venice reported the suicide of a woman in calle Frezzaria, who threw herself from the balcony into the canal, putting an end to her delirium (“Venetia 22 Maggio 1700”). An avviso from Paris dispatches the news that a 64-year-old woman was broken on the wheel in Sint-Niklaas, having been convicted of poisoning five husbands (“Parigi 5 Marzo 1700”). Although a characteristic of Italian avvisi, compared to German Zeitungen, is to indulge in reporting many details about crimes, accidents, and trials (Molino, “Connected News” 283), it happens more frequently in the 1700 dataset than in the 1600 one that brief stories of nameless characters notable only for their fortunes or misfortunes are narrated in the newsletters. Despite their local dimension, they circulated all over Europe and attracted readers’ curiosity, as the triple homicide news opening this article shows.

35Gossip is also included in the topic of “Culture”. News about the private life of certain personalities occurs mainly in newsletters from Rome, and the story may be updated from week to week. Four newsletters of June and July 1700, for instance, deal with the second marriage of the noblewoman Angela Mignanelli, widow of Giovanni Stefano Gavotti, with Carlo Butii. The event interested the Romans, since an avviso announced that many curious people came to the wedding, and the news-reading public outside Rome could consume ever new details, in four different editions of the newsletters (“Roma 27 Giugno 1700 ;” “Roma 29 Giugno 1700 ;” “Roma li 29 Giugno ;” “Roma 3 Luglio 1700”). The presence of local news regarding crime, entertainment, and gossip may partly be ascribed to a change whereby the genre was no longer intended exclusively for noble courts, secretariats of governments and ambassadors, and merchants, as in the early days, but for a wider audience of readers and even listeners among the general public.

36This article has compared two years, 1600 and 1700, of handwritten newsletters belonging to the Medici collection to evince the elements of continuity as well as change of the avvisi genre over time. A quantitative approach and digital methods, especially network analysis, have been applied to study how news circulated and compare the form and content of the 1600 and 1700 handwritten newsletters. It is our hope that this contribution may become a stepping stone for future comparative research since both its results and approach may be used to study new corpora of handwritten newsletters from other years, languages or collections. Three elements have been analyzed in detail : the hubs, the places where the news items came from and the topics of the news items. Across the seventeenth century and despite the appearance of printed gazettes, the avvisi did not change significantly in form or content but remained a very conservative genre. The layout, the main topics covered, and the balance between news from the centres and news from the peripheries remained unchanged, confirming the longevity of the European newsletters phenomenon. This article has, however, shown that handwritten newsletters became longer in 1700 and assumed a more standardized form, similar to printed periodical editions. The system of newsletters’ production became more articulated, and the information hubs moved toward Northern Europe. By 1700, two new opposing trends appeared. On the one hand, newsletters testify to an increased focus on global news. The news network becomes more interconnected, the areas of influence of individual hubs are reduced, and many places, both European and overseas, communicate news to various centres. If everyday life becomes more dependent on European and global issues, on the other hand, readers develop an interest in local news, which is enjoyable to read although not politically or commercially important. The combination of global and local news that emerges in the 1700 newsletters reveals how such a conservative genre was beginning to slowly embrace new elements of a changing society.