Rachel Midura, Ph.D.

Assistant Professor of Digital History at Virginia Tech

Italian Administrative Hands

21 Jul 2020 » tool

Italian Administrative Hands, 1550-1700 Handwritten Text Recognition (HTR) Model for Transkribus


Introduction:

The following documentation provides information for a model developed for use with the Transkribus transcription tool developed by the READ project and maintained by the Transkribus team at the University of Innsbruck. Registration and download of the platform are free, as is the use of the Handwritten Text Recognition model as of 7/21/2020.* Based on transcription of over 200 pages of Italian early modern handwritten texts, we have trained a neural network to provide a machine-generated transcription for use with similar documents. The following documentation provides background on training materials as well as model conventions, especially for common abbreviations. Bios of the collaborators and an example reference for use of the model can be found at the bottom of the page.

*a per-page pricing model is expected in the near future.

Model Description:

The Italian Administrative Hands model features a variety of Italian-language documents from state archives in Milan, Venice, Florence, Pisa, and Genoa. The training set represents a spectrum of humanistic, italic and cursive hands characteristic of administrative records, employed by secretaries and newswriters. The model has been trained to perform well with a mix of quantitative and qualitative information as well as many common proper nouns for the period, such as locations in Europe and contemporary rulers. Administrative documents often employ common superscript abbreviations, which the accompanying documentation treats in greater detail. The model can also be used with Latin, Spanish and French documents to some extent. The model represents a collaboration between Jake Dyble (Exeter/Pisa), Antonio Iodice (Exeter/Genoa), Sara Mansutti (Cork), and Rachel Midura (Virginia Tech).

Training Set:

Archivio di Stato di Milano (ASMi) Documents Description:

Documents from the state archive in Milan concentrate from 1580-1620, and derive primarily from the Carteggio delle cancellerie dello Stato, Atti di Governo, and Registri delle cancellerie dello Stato collections. The majority are correspondence, with a minority of edicts and financial records. Documents were collected in the course of “Masters of the Post,” a dissertation on the development of Europe’s communications networks in northern Italy. Many relate to the Spanish post office of Milan. Additional themes include jurisdiction, finance, plague, and policing.

Archivio di Stato di Venezia (ASVe) Documents Description:

Documents from the state archive in Venice concentrate from 1580-1620, and derive primarily from the Compagnia dei corrieri, Senato, and Inquisitori di Stato collections. The majority are correspondence, with a minority of edicts and council minutes. Documents were collected in the course of “Masters of the Post,” a dissertation on the development of Europe’s communications networks in northern Italy. Many relate to the Venetian company of couriers and their competitors. Additional themes include jurisdiction, espionage, plague, news-gathering and reporting.

Archivio di Stato Firenze (ASF) /Euronews Documents Description:

Documents from the state archive in Florence range from 1550 to 1650. They belong to the fond Mediceo del Principato, series Relazioni con stati italiani ed esteri. The majority are handwritten newsletters (called avvisi in Italian) sent by Florentine ambassadors and agents to the Medici Court, with a minority of correspondence. The documents are collected as a part of the Irish Research Council project Euronews, which explores the information connectedness of premodern European society by examining the regular circulation of news in a vast range of pre-newspaper networks.

Archivio di Stato Pisa (ASP) Documents Description:

The documents from the Pisan state archive are legal and commercial papers deriving from the period 1600-80 and taken from the atti civili sub-series of the archive of the Consoli del Mare di Pisa. The court of the Consuls was an ancient Tuscan tribunal with jurisdiction over various maritime and commercial affairs. The documents in question are the records of maritime average procedures, cost-sharing exercises in which expenses and damages were shared between interested parties. As such the corpus includes both first-hand narrative accounts, citations, witness examinations, lists of interrogatories, sentences, and a large number of accounting documents (calcoli). It thus provides a wide range of hands, non-standard layouts, proper names, and numbers. These documents were examined as part of the ERC-funded project AveTransRisk.

ASG Genoa Documents Description:

The documents from the Genoese state archive are legal and commercial papers deriving from the period 1590-1700 and taken from the Notai Giudiziari and the Conservatori del Mare collections. The court of the Conservatori del Mare was an ancient Genoese tribunal with jurisdiction over various maritime and commercial affairs, while some of the Notai Giudiziari worked for the office of the calculators, the main magistracy involved in a General Average procedure. The documents in question are the records of maritime average procedures, cost-sharing exercises in which expenses and damages were shared between interested parties. As such the corpus includes both first-hand narrative accounts, citations, witness examinations, lists of interrogatories, sentences, and a large number of accounting documents (calcoli). It thus provides a wide range of hands, non-standard layouts, proper names, and numbers. These documents were examined as part of the ERC-funded project AveTransRisk.

Conventions:

The following conventions are intended to provide examples of how collaborators trained the model to handle abbreviation, contractions, and other common stylistic variations.

Common Abbreviations:

Common Abbreviations 1 Common Abbreviations 2


Superscript and Subscript Abbreviations:

Superscript and Subscript Abbreviations


Contractions (End):

End Contractions 1 End Contractions 2


Contractions (Middle):

Middle Contractions


Et/It:

Et and It


Ii/Jj:

Ii and Jj


Ss:

Long S


Uu/Vv:

Uu and Vv


Numbers and Symbols:

Numbers and Symbols 1 Numbers and Symbols 2


Font:

Font

Citation Prompt:

When using the model, please use the following citation information, provided here in Chicago format for your convenience:

Footnote:

Jake Dyble, Antonio Iodice, Sara Mansutti and Rachel Midura (Primary Investigators), “Italian Administrative Hands, 1550-1700 Handwritten Text Recognition (HTR) Model for Transkribus,” Early Modern Digital Itineraries, July 2020, https://emdigit.org/tool/2020/07/21/italian-administrative-hands.html.

Bibliography:

Dyble, Jake, Antonio Iodice, Sara Mansutti and Rachel Midura (Primary Investigators). “Italian Administrative Hands, 1550-1700 Handwritten Text Recognition (HTR) Model for Transkribus.” Early Modern Digital Itineraries. July 2020. https://emdigit.org/tool/2020/07/21/italian-administrative-hands.html.

Contributors:

Jake Dyble is a third-year doctoral student based jointly at the University of Exeter and the Università di Pisa, researching maritime law in the ‘free port’ of Livorno (Tuscany) during the seventeenth century. His thesis focuses on the use and development of maritime averages, a transnational legal instrument operating across political, legal, and cultural boundaries. He is a member of the European Research Council project AveTransRisk and is supervised by Professor Maria Fusaro (Exeter) and Professor Andrea Addobbati (Pisa). Before beginning work on the PhD, Jake studied at the University of Cambridge where he completed a BA in History (first-class honours) and was awarded the Trevelyan History Prize. He then went on to complete an MPhil in Medieval History, also at Cambridge.

Antonio Iodice is a third-year PhD student within the ERC project AveTransRisk directed by Maria Fusaro. He is based jointly in the Exeter University and the University of Genoa, under the supervision of Luisa Piccinno. He studies General Averages in Genoa during the Early Modern period. Antonio has achieved his Master’s double degree at the University of Rome “La Sapienza” in partnership with the University of Grenoble “Pierre Mendès-France”. He has achieved his first PhD in Modern History at the University of Naples “Federico II” with a thesis called “The free port, spreading of an economic model: politics, actors, ideologies, myth. Two compared realities: Genoa and Marseille (1590-1817)”. He is a member of the project A global history of free ports, coordinated by Antonio Trampus (Ca’ Foscari University) and Koen Stapelbroek (Helsinki University). He is an editor in the online journal Giornale di Storia. Forthcoming publications: G_eneral Averages in Genoa: between rules and practice_, which will be published in the volume “Sharing Risk: General Average, 6th – 21st Centuries”, edited by Maria Fusaro, Luisa Piccinno and Andrea Addobbati; Managing shipping risk: General Average and marine insurance in Early modern Genoa, written together with Luisa Piccinno, which will be published in a volume in the series “Comparative Studies in the History of Insurance Law” published by Duncker & Humblot and edited by Guido Rossi and Phillip Hellwege; Napoli, un sogno di franchigia, 1617-1739, which will be published in a volume dedicated to Anna Maria Rao, edited by Pasquale Palmieri and others; Il porto franco di Marsiglia, Palladium de prosperité (1669-1794), which is being published in the acts of conference Les règles des lieux, held at the Ecole française de Rome in September 2016.

Sara Mansutti is a PhD student in Digital Humanities at the University College Cork and a doctoral fellow within the EURONEWS Project, funded by the Irish Research Council. Her PhD project focuses on the newsletters sent by Cosimo Bartoli, Florentine resident in Venice, to the Medici between 1562 and 1572. Her research aims to understand the role of the handwritten newsletters within the diplomacy of the Medici Court and to investigate the relationship between letters and avvisi. In 2017 she completed a master’s degree in Italian Philology and Literature at the University of Udine. Part of her MA thesis, involving the transcription and analysis of a mid Sixteenth-century manuscript inventory of books, has been published in the volume Printing R-Evolution and Society 1540-1500. Fifty Years that Changed Europe (2020).

Rachel Midura is an assistant professor of digital history at Virginia Polytechnic Institute and State University. She has recently finished her dissertation at Stanford University, “Masters of the Post: Northern Italy and European Communications Networks, 1530-1730,” on early modern surveillance, espionage, and postal systems. She has previously worked as a senior graduate research fellow at the Stanford Center for Spatial and Textual Analysis, and collaborated on the Grand Tour and Early Modern Mobility projects. Her work appears in collected volumes such as The Renaissance of Letters and Empires of Knowledge: Scientific Networks in the Early Modern World. Her article “Itinerizing Europe: Early Modern Spatial Networks in Printed Itineraries, 1545-1747” is forthcoming in the Journal of Social History.