History by Algorithms AI and the Future of Historical Research (Zvi Lotker) (Z-Library)

📄 File Format: PDF

💾 File Size: 37.2 MB

📖 Read Online ⬇️ Download

Registered users can read the full content for free

Register as a Gaohf Library member to read the complete e-book online for free and enjoy a better reading experience.

📄 Page 1

Lotker History by Algorithm s History by Algorithms AI and the Future of Historical Research Zvi Lotker

📄 Page 2

History by Algorithms

📄 Page 3

Zvi Lotker History by Algorithms AI and the Future of Historical Research

📄 Page 4

Zvi Lotker Faculty of Engineering Bar-Ilan University Ramat-Gan, Israel ISBN 978-3-031-93626-5 ISBN 978-3-031-93627-2 (eBook) https://doi.org/10.1007/978-3-031-93627-2 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2025 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland If disposing of this product, please recycle the paper.

📄 Page 5

Preface Herodotus of Halicarnassus, the Father of History, first recorded history “so that the actions of people will not fade with time, including other things and especially the cause for which they went to war with one another.” In our current age of big data, we record history as it happens and soon move on, overwhelmed by unprecedented modern documentation. History is a complex, evolving field that is rapidly expanding both in resources and accessibility. Our stories are buriedwhile they are beingwritten (Fig. 1). Thus, the need for intelligent machines to understand our personal and cultural narratives becomes more imperative than ever. Having machines understand what our history is and why it happened would allow a dialogue that makes use of man’s directing and machine’s endless memory [1]. The first step to getting machines to understand narrativewas taken inmy book,AnalyzingNarratives in Social Networks. Fig. 1 Cartoon about a dialogue between man and machine about history v

📄 Page 6

vi Preface [2] We now propose to continue this journey into the different structure of never- ending stories: Historical narratives. Overview The book is divided into three main parts. In Part I, we discuss the mathematical language of history; Part II uses simple models to analyze historical law written in mathematical language; and the Part III discusses the implications of general Large Language Models (LLMs) for the study of history. Synopsis For machines to participate in the study of history, it is necessary to develop mathe- matical tools capable of describing historical narratives. In the first part of the book, we take the first steps toward achieving this goal. Tools such as big-O notation and the relative to history are discussed in Chap. 1. We continue to develop mathematical tools needed to describe historical narrative in Chap. 2, where we discuss the implementation of Category Theory to the study of history and suggest that the definition of historical concepts should be defined using function diagrams and natural language definition simultaneously. For example, we use a function diagram to define reading against the grain. This approach reveals that reading against the grain actually compares past and present archives In our function diagram, we are forced to discuss the present as the element of comparison of the present to the past. This exact definition suggests that there is a reversed reading against the grain where the past archives try to read the future archives. In Chap. 3, we develop the mathematical tools for describing battles, a central theme of history. We use the Lanchester differential equation in two-dimensional space-time and extend it for the case where one army surrounds another. This is the central element in the study of history and a typical narrative of great battles. Analyzing the Battle of Cannae led by Hannibal, we were able to study it in greater detail than ever before. Calculations were made using the extended Lanchester equa- tion. The mathematical tools used in this chapter are differential equations, which are basic tools in exact science, but are rarely used in studying history. In Chap. 4, we extend the concept of the counters into clocks. In many digital history research books, the counters are heavily used. The standard methodology is to define subsets of words and then count their appearance in the documents. From that, the researchers tend to conclude some facts about the document. However, there is no methodology for comparing these counters, and what is their meaning from a mathematical point of view? We identify those counters with clocks and develop the theory of many clocks with respect to historical narratives. We show how to find the time when the clocks deviate from each other. This method allows machines to

📄 Page 7

Preface vii point to the critical event in the historical narrative. This concludes Part I of the book, which develops the mathematical tools for historical narratives. In Chap. 5, we introduce the concept of Networks and Macrohistory. We start with the definition of a network and provide several examples of graphs and their basic data structures. Then, we move to discuss some basic computational geometry with application to history. We end up this chapter with a causal graph. In Chap. 6, we introduce Microhistory. We use social networks as a mathematical model that describes society. We develop several rules of thumb on social networks, which allow us to resurrect ancient networks only from the seizure of societies. This allows us to compare ancient social networks and compute the meaning of the destruction of old cities such as Carthage by the Romans. A social network is a mathematical model that describes the fabric of society and is therefore useful when one tries to model historical events. Chapter 7 is the introduction to Part II of the book, where we discuss the impli- cations of the historical narrative on mathematical language. The following chapters will provide several examples of historical processes that can be modeled using simple mathematical models and provide windows to the mathematical laws of history. Chapter 8 discusses the historical question: Is the history of the core similar or different from the history of the periphery? This question is central to the philosophy of history since most of our data related to historical revolution/process is of the core, and we are missing the data of the periphery. By modeling the relationship between the core and periphery through social networks, we are able to categorize social revolutions and provide critical conditions on when the history of the core is relevant to the history of the periphery. Chapter 9 provides an example of historical law that focuses on the author instead of the historical event. The law says that any tremendous historical figure needs an antipodal figure that will be compared to him and will be worth it. This law comes not from history but from the historical narrative. Therefore, the author of history tends to pick evil characters in history after the witness disappears. In the context of philosophy, this uses the fact that according to Kant, when we analyze reality, we always have to go through the perception of reality. In science, there is always reality and perception of reality. This chapter also discusses the connection between history and distributed computing. Itmainly analyzes the ability of society to reach an agreement after the witnesses die through game theory and shows that it is necessary for collective memory to form the historical canonical text. Without the existence of a canonical historical text, the narrative will evaporate. In Chap. 10, we analyze the conversion of the averaging process in the collective memory formation game and show that if the nodes have a correct categorization of their neighbors’ opinions, the average process converges to the Nash equilibrium and can be used as a separate scheme that defines the different historical entities. A simple example is gun control in the USA. If you know a person’s opinion about gun control, you can glean their political opinion. In Chap. 11, we analyze the collective memory formation game in the case where some nodes have the wrong categorization of their neighbors’ opinions. In such

📄 Page 8

viii Preface instances, the average process converges to zero opinion, and the historical event becomes unimportant. The different parties cannot disagree on the historical event using just the averaging process as the separation process. However, if we wish to use the historical events as a separation event, we need to move the averaging process away from zero. This can be done by amplifying the probabilities in each step to amount to 1. This suggests that propaganda is necessary to use such events as separation. In Chap. 12, we model a stochastic terrorism network and explain the mechanism that generates random violence events using queuing theory. Chapter 13 defines the general historical machine and sets the scene for generating history by machines. In the chapter, we define the relationship between historical entities, machines that write history, and the historical narrative. The central idea in this chapter is universal consciousness, which is the subject of historical narratives such as society, state, country, and more. This ends the second part of the book. In Chap. 14, we begin the third part of the book, which explains how history transforms from text-based science to science based on information. In Chap. 15, we provide a general introduction to machine learning and the way AI can be used in the study of digital history. The chapter is divided into two parts. The first part is a general discussion of AI and the way historians may use it. The second part discusses LLM and prompt engineering in general. Chapter 16 deals with the new ways in which historical narrative can be repre- sented. We move from textual-based to video-based historical narrative. The chapter provides several algorithmic methods to take any text and transform it into a video. Chapter 17 explains how touseLLMas an expert system inhistorical text analyses. We provide several examples of expert systems, such as psychiatric experts, political science experts, and emotion analysis experts. We develop a method that takes any historical document, breaks it into paragraphs, transforms it into numerical values, aggregates those numbers into functions, and runs standard analysis tools on those functions. To demonstrate this method, we provide an analysis of Hitler’s Mein Kampf [3] and Churchill’s My Early Life [4]. In Chap. 18, we develop a methodology to analyze historical videos together with machine learning tools to study history. Lastly, Chap. 19 discusses Fake History from the perspective of the philosophy of science. This chapter provides several rules of thumb for Fake History and its evolution. Download Materials Code and data from the book are available at https://github.com/zvilo/History-by- Algorithm. Any future updates to the materials will be made directly in the same GitHub repository.

📄 Page 9

Preface ix Why Should You Read This Book? Most of the books written about digital history synthesize the philosophy of history with technical details from computer science. However, digital history borrowsmany philosophical aspects from computer science. According toMcLuhan [5], themedium is themessage. If that is the case, in digital history the message is the machine, since the machine is the medium. Therefore, to understand digital history, one needs to understand the philosophy of the machine. This book tries to demonstrate how the philosophy of themachine, as developed by computer scientists, influences the study of digital history. Historians face technical difficulties when encountering the language of computer science for the first time. However, we believe that historians who wish to work in the area of digital history should be familiar not onlywith the philosophy of history but alsowith the philosophy of computer science, since digital history is the fusion of both disciplines. To bridge the language barrier between historians and other readers with a human- ities background, we provide several appendices filled with prompts that can be used to converse with modern AI tools.We believe that these prompts can ease the process of bridging the language gap, making the whole process possible. Those with a background in exact sciences are much more familiar with the mathematical language used in the book. For them, reading the book should be easier. We hope that it will prompt them to develop the exact science of digital history. One of the advantages of the book is that it exposes the reader to a variety of toolboxes from exact science that can be used during research in digital humanities. Most of the books in digital history study history using computer science program- ming languages. For example, many works are based on counting the frequency of specific words in historical texts. For now, let’s call them “counters”. However, none of them ask what the mathematical properties of these counters are and how we can compare them. The purpose of this book is not to analyze a historical text, but rather to develop the mathematical theory needed for doing digital history. The main difference between exact science and the humanities is that while exact science tries to predict, the humanities try to interpret. This is the major obstacle when performing digital humanities. The problem is that once we transform the historical document from words to numbers, we may forget that we are still talking about interpretation, not prediction. Numbers tend to have a unique interpretation, and therefore the research tends to lose the freedom of interpretation. Researchers in digital humanities need to remember that in order to save the freedom of interpre- tation, we are forbidden to fully define how to transform the text into numbers. We should never forget that when we cross the bridge from text to numbers, we perform interpretation, and different bridges will generate different numbers, and therefore, we will have different interpretations. By understanding that transforming the text to numbers is an interpretation, we maintain the desired freedom of interpretation.

📄 Page 10

x Preface The book can be used as a textbook for a single-semester advanced course on digital history in the Faculty of Engineering. Zvi Lotker Faculty of Engineering Bar-Ilan University Ramat-Gan, Israel References 1. Herodotus and Alfred Denis Godley. Herodotus: With an English Translation by AD Godley. W. Heinemann, 1961. 2. LotkerZvi.AnalyzingNarratives in SocialNetworksTakingTuring to theArtsAuthors. Springer, 2021. 3. Adolf Hitler.Mein kampf , volume 1. Motilal Banarsidass, 2014. 4. Winston Churchill.My early life. DigiCat, 2022. 5. Marshall McLuhan. The medium is the message. In Communication theory, pages 390–402. Routledge, 2017.

📄 Page 11

Acknowledgements Many people contributed to the making of this book. First and foremost, I would like to thank Dr. Roman Fedorenko for the tremendous amount of time, thought, and effort he devoted to this project. His support, critical insights, patience, and hard work were invaluable—this book would not exist without him. I am also grateful to Rotem Alter for carefully proofreading the manuscript and correcting numerous language mistakes, to Maya Sharon for proofreading, and to Gaya Golan, who helped me with the early version of the book. For the beautiful illustrations that open each chapter, I owe special thanks to Raffael Blumenberg. Our conversations during the creative process of these figures also led to valuable insights about the book itself. Finally, Iwant to thankmypartner,Hamutal Tsamir, for her thoughtful discussions on history, and my beloved children, Hillel and Daniel Lotker-Hassine, for their patience and understanding throughout this journey. Competing Interests The author has no competing interests to declare that are relevant to the content of this manuscript. xi

📄 Page 12

Contents Part I The Mathematical Language of Digital History 1 Asymptotics in Digital Humanities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.3 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.3.1 Big-O Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.3.2 Big-Omega Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.3.3 Big-Theta Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.4 The Problem with Big-O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.5 Elites in Social Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.6 Machine Learning and Asymptotics . . . . . . . . . . . . . . . . . . . . . . . . 10 1.7 Exponential Growth in Non-Linear History . . . . . . . . . . . . . . . . . 11 1.8 Examples Throughout History . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1.8.1 Rome: Republic to Empire . . . . . . . . . . . . . . . . . . . . . . . 12 1.8.2 The American Revolution . . . . . . . . . . . . . . . . . . . . . . . . 14 1.8.3 France: Viva La Revolución . . . . . . . . . . . . . . . . . . . . . . 15 1.8.4 Russia: The 1917 Revolution . . . . . . . . . . . . . . . . . . . . . 16 1.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 1.10 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 2 Function Diagrams in Digital Humanities . . . . . . . . . . . . . . . . . . . . . . . 21 2.1 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.2 What Is History? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.2.1 Place . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.2.2 Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.2.3 Meaning of Historical Event . . . . . . . . . . . . . . . . . . . . . 24 2.2.4 Historical Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 2.2.5 Example World War II . . . . . . . . . . . . . . . . . . . . . . . . . . 25 xiii

📄 Page 13

xiv Contents 2.3 Description of Function Diagrams . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.3.1 Category Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.3.2 Analyzing Narratives in Social Networks . . . . . . . . . . . 27 2.3.3 From Narratives to Historical Narratives . . . . . . . . . . . 28 2.4 One-Dimensional Historical Narrative . . . . . . . . . . . . . . . . . . . . . . 28 2.4.1 The Process of Building an Archive . . . . . . . . . . . . . . . 29 2.4.2 Machine Learning and Anomaly Detection . . . . . . . . . 31 2.5 Comparative History or D-Dimensional Historical Narrative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.5.1 The Process of Building Archives in Comparative History . . . . . . . . . . . . . . . . . . . . . . . . . . 33 2.5.2 Explaining the Comparative Function Diagram . . . . . 33 2.6 History with Social Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2.6.1 Constructing Archives in Historical Social Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 2.6.2 Explaining the History Social Network Diagram . . . . 35 2.7 Reading Against the Grain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 2.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 2.9 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 3 History of War . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.3 Lanchester’s Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 3.4 Adapting Lanchester’s Laws for Flanking . . . . . . . . . . . . . . . . . . . 46 3.5 Incorporate Geography for the Lanchester’s Model . . . . . . . . . . . 49 3.6 Battle of Cannae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.6.1 The Forces Before the Battle . . . . . . . . . . . . . . . . . . . . . 50 3.6.2 Modeling the Battle of Cannae . . . . . . . . . . . . . . . . . . . 51 3.6.3 The First Cavalry Battle . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.6.4 Counterfactual of the Second Cavalry Battle . . . . . . . . 53 3.6.5 The First Infantry Battle . . . . . . . . . . . . . . . . . . . . . . . . . 55 3.6.6 The Final Battle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 3.6.7 Computing the Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 3.7 General Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 3.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 3.9 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 4 History Through Clocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4.1 Syuzhet and Fabula Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4.2 Clocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 4.3 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 4.4 Comparing Two Clocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4.5 Programming Tips for Clocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

📄 Page 14

Contents xv 4.5.1 Generating Subjective Clock from a List . . . . . . . . . . . 73 4.5.2 Generating a Subjective Clock from Historical Document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 4.6 The History of Two Clocks: Examples . . . . . . . . . . . . . . . . . . . . . 75 4.6.1 From List to Clocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4.6.2 Greenwich Clock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4.6.3 Uniform Clock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4.6.4 Knowledge Clock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 4.6.5 Wikipedia Pageviews . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 4.6.6 Network Clock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 4.7 Time Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 4.7.1 The Current List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 4.8 Many Clocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 4.9 Many Clocks Synthetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 4.10 Combining the Physicists’ Clocks . . . . . . . . . . . . . . . . . . . . . . . . . 86 4.11 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 4.12 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 5 Networks and Macrohistory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 5.2 Macrohistory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 5.3 Graph Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 5.3.1 Types of Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 5.3.2 Functions of a Graph . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 5.3.3 Data Structures for Graph Representation . . . . . . . . . . 95 5.4 Interval Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 5.5 Spacetime Intersection Graph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 5.5.1 Convex Hull . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 5.5.2 Delaunay Graph and Minimum Spanning Tree . . . . . . 99 5.6 Knowledge Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 5.7 Dynamic Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 5.8 Causality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 5.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 5.10 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 6 Networks and Microhistory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 6.2 Social Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 6.2.1 Number of Edges in a Social Network . . . . . . . . . . . . . 112 6.2.2 Connectivity Argument . . . . . . . . . . . . . . . . . . . . . . . . . . 113 6.2.3 Heavy-Tail of Degree Distribution . . . . . . . . . . . . . . . . 113 6.2.4 Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

📄 Page 15

xvi Contents 6.3 Structures of a Social Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 6.3.1 Erdős–Rényi Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 6.3.2 Core-Periphery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 6.4 Counting Motifs in Social Network . . . . . . . . . . . . . . . . . . . . . . . . 123 6.4.1 Combinatorial Argument . . . . . . . . . . . . . . . . . . . . . . . . 124 6.4.2 Core-Periphery and the Number of Triangles . . . . . . . 125 6.4.3 Counting the Number of Triangles: Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 6.5 Network Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 6.6 Application to History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 6.6.1 The First Punic War . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 6.6.2 Refugees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 6.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 6.8 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Part II Metahistory: Computational Models for History 7 What Is a Computational Model for History . . . . . . . . . . . . . . . . . . . . . 141 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 8 History Through the Core . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 8.2 The Homogeneous Core-Periphery Revolutionary Fluid Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 8.2.1 The Revolution Starts at the Periphery Before the Core . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 8.2.2 The Revolution Starts at the Core Before the Periphery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 8.2.3 Categories of Revolution Using a Homogeneous Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 8.3 General Core-Periphery Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 8.3.1 The Base of the Induction . . . . . . . . . . . . . . . . . . . . . . . . 152 8.3.2 General Case of Induction . . . . . . . . . . . . . . . . . . . . . . . 153 8.3.3 Categories of Revolution Using the Non-homogeneous Model . . . . . . . . . . . . . . . . . . . . 154 8.3.4 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 8.4 Accessing the Core Through Wikipedia . . . . . . . . . . . . . . . . . . . . 155 8.5 COVID-19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 8.5.1 Estimating the Number of Deaths at the Core . . . . . . . 157 8.5.2 Estimating the Number of Deaths Periphery . . . . . . . . 157 8.5.3 Numerical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 8.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 8.7 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

📄 Page 16

Contents xvii 9 History After the Death of the Witnesses . . . . . . . . . . . . . . . . . . . . . . . . 161 9.1 Related Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 9.1.1 Related Works on Consensus . . . . . . . . . . . . . . . . . . . . . 163 9.1.2 Related Works on Collective Memory . . . . . . . . . . . . . 164 9.2 Narrative History and Consensus in Distributive Computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164 9.3 Consensus in Distributed Computing . . . . . . . . . . . . . . . . . . . . . . . 165 9.3.1 Faulty Processor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166 9.4 Distributed Computing and Digital History: A Short Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 9.4.1 Similarities Between Distributed Computing and Digital History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 9.4.2 Differences Between Distributed Computing and Digital History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 9.5 The Model of Collective Memory Formation Game . . . . . . . . . . 169 9.5.1 Transforming Definitions from Consensus . . . . . . . . . . 170 9.5.2 Deviations from Consensus . . . . . . . . . . . . . . . . . . . . . . 171 9.5.3 Life Span . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 9.6 Historical Consensus After the Death of the Witnesses . . . . . . . . 174 9.6.1 No Canonical Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174 9.6.2 Existing Canonical Text . . . . . . . . . . . . . . . . . . . . . . . . . 174 9.6.3 Death of Witnesses and the Narrative of History . . . . . 174 9.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 9.8 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 10 History While the Witnesses Are Still Alive . . . . . . . . . . . . . . . . . . . . . . 179 10.1 Conversion to the Collective Memory . . . . . . . . . . . . . . . . . . . . . . 180 10.2 Example of a Collective Memory Formation Game . . . . . . . . . . . 180 10.2.1 A Single Historical Event . . . . . . . . . . . . . . . . . . . . . . . . 180 10.2.2 Red and Blue Collective Entities . . . . . . . . . . . . . . . . . . 181 10.2.3 Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 10.2.4 Communication Graph . . . . . . . . . . . . . . . . . . . . . . . . . . 181 10.2.5 Pure Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 10.2.6 Utility Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 10.2.7 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 10.2.8 Nash Equilibrium of a Collective Memory Formation Game in Pure Strategy . . . . . . . . . . . . . . . . . 183 10.3 Conversion to the Nash Equilibrium . . . . . . . . . . . . . . . . . . . . . . . 186 10.3.1 Algorithmic Approach for Solving Collective Memory Formation Game . . . . . . . . . . . . . . . . . . . . . . . 186 10.3.2 Markov Process with One Opinion . . . . . . . . . . . . . . . . 187

📄 Page 17

xviii Contents 10.3.3 Matrix Version of the Weighted Average Algorithm with Two Opinions . . . . . . . . . . . . . . . . . . . . 190 10.3.4 Reaching a Consensus . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 10.4 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 10.4.1 Reaching a Consensus on Disagreement with Complete Information . . . . . . . . . . . . . . . . . . . . . . 195 10.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 10.6 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 11 Propaganda in History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 11.1 The Price of Wrong Characterization . . . . . . . . . . . . . . . . . . . . . . . 202 11.2 Failing to Reach a Consensus on Disagreement . . . . . . . . . . . . . . 203 11.3 Publicity and Propaganda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 11.4 One-Sided Propaganda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 11.4.1 Propaganda of the Majority . . . . . . . . . . . . . . . . . . . . . . 206 11.4.2 Propaganda of the Minority . . . . . . . . . . . . . . . . . . . . . . 207 11.5 Historical Memory Formation Game . . . . . . . . . . . . . . . . . . . . . . . 208 11.5.1 The American Civil War . . . . . . . . . . . . . . . . . . . . . . . . . 208 11.5.2 2022 Russian Invasion of Ukraine . . . . . . . . . . . . . . . . . 210 11.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 11.7 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 12 Stochastic Terrorism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 12.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 12.3 The Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 12.4 Birth–Death Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 12.4.1 The Queue as Data Structure . . . . . . . . . . . . . . . . . . . . . 219 12.4.2 From Queues to Birth and Death Process . . . . . . . . . . . 219 12.5 Integer Birth–Death Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 12.5.1 The DQueue as Data Structure . . . . . . . . . . . . . . . . . . . . 221 12.5.2 From Dqueue to Integer Birth and Death Process . . . . 221 12.5.3 Stationary Distribution of the Integer Birth Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 12.5.4 Ornstein-Uhlenbeck . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 12.5.5 Jackson Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 12.5.6 Stochastic Terrorism Network . . . . . . . . . . . . . . . . . . . . 227 12.5.7 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 12.6 Fake News Generating an Echo Chamber . . . . . . . . . . . . . . . . . . . 230 12.6.1 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 12.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 12.8 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234

📄 Page 18

Contents xix 13 Historian Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 13.2 Related Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239 13.3 Objective Consciousness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 13.4 Machine Consciousness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 13.5 Consciousness and Historical Ententes . . . . . . . . . . . . . . . . . . . . . 242 13.6 Life Cycle of a Narrative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 13.6.1 The Author as a Conscious Machine . . . . . . . . . . . . . . . 245 13.6.2 The Translator as a Conscious Machine . . . . . . . . . . . . 245 13.6.3 The Reader as a Conscious Machine . . . . . . . . . . . . . . . 246 13.7 Universal Consciousness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 13.7.1 Difficulties in Universal Consciousness . . . . . . . . . . . . 248 13.7.2 Collective Universality . . . . . . . . . . . . . . . . . . . . . . . . . . 249 13.8 Historian Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 13.9 From Timeline to Historical Entities . . . . . . . . . . . . . . . . . . . . . . . 250 13.9.1 Interval Caterpillar Graph . . . . . . . . . . . . . . . . . . . . . . . . 251 13.10 Attention Schema of Interval Graph Caterpillar . . . . . . . . . . . . . . 252 13.11 Attention Schema Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 13.11.1 Number of Hyperlinks Attention Schema . . . . . . . . . . 254 13.11.2 Number of Words Attention Schema . . . . . . . . . . . . . . 254 13.11.3 Page Views Attention Schema . . . . . . . . . . . . . . . . . . . . 255 13.11.4 Centrality Attention Schema . . . . . . . . . . . . . . . . . . . . . 255 13.11.5 Causality Attention Schema . . . . . . . . . . . . . . . . . . . . . . 255 13.11.6 WWII Attention Schema Examples . . . . . . . . . . . . . . . 256 13.12 Historical Narrative of Two Timelines . . . . . . . . . . . . . . . . . . . . . . 256 13.12.1 War Between Timelines . . . . . . . . . . . . . . . . . . . . . . . . . 256 13.12.2 Cooperation Between Two Timelines . . . . . . . . . . . . . . 256 13.13 What is History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 13.14 Eliminating Negative Probabilities Using the Model of Historian Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 13.15 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 13.16 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262 Part III History Through AI 14 Information and History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267 15 Machine Learning and History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271 15.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271 15.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 15.3 Basic AI Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 15.4 Basic AI Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 15.4.1 Mathematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 15.4.2 Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274

📄 Page 19

xx Contents 15.4.3 Physics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 15.5 Prompt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276 15.5.1 Precision Versus Innovation . . . . . . . . . . . . . . . . . . . . . . 276 15.5.2 Dialogue Versus Monologue . . . . . . . . . . . . . . . . . . . . . 277 15.5.3 Experts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278 15.6 How to Write a Good Historical Prompt . . . . . . . . . . . . . . . . . . . . 278 15.6.1 Evaluating a Sample Prompt . . . . . . . . . . . . . . . . . . . . . 279 15.7 AI and Two Clocks in the Gallic War . . . . . . . . . . . . . . . . . . . . . . . 279 15.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 15.9 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 16 Representing Historical Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 16.1 Word Clouds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 16.2 Evolving Graph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286 16.3 Generating a Slideshow and Movies of Historical Texts Using AI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 16.4 Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 16.4.1 Personal Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 16.4.2 Tree Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 16.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 16.6 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294 17 Expert System in History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 17.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 17.2 Function Diagram for Experts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 17.3 Applying AI as an Expert in Humanities . . . . . . . . . . . . . . . . . . . . 296 17.4 Estimating Violence in UN Summaries . . . . . . . . . . . . . . . . . . . . . 297 17.5 Analyzing Historical Documents with AI Expertise . . . . . . . . . . 299 17.5.1 Step One: Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 17.5.2 Step Two: Processing Historical Text . . . . . . . . . . . . . . 299 17.5.3 Step Three: Textual Analyses . . . . . . . . . . . . . . . . . . . . . 299 17.5.4 Step Four: Aggregate . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300 17.5.5 Step Five: Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300 17.6 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 17.6.1 Step Three: Psychiatric Expert . . . . . . . . . . . . . . . . . . . . 302 17.6.2 Step Three: Political Science Expert . . . . . . . . . . . . . . . 303 17.6.3 Step Three: Emotion Analysis Expert . . . . . . . . . . . . . . 304 17.7 Example: Adolf Hitler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304 17.7.1 Example: Step Two . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304 17.7.2 Example: Step Three . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 17.7.3 Example: Step Four . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 17.7.4 Example: Step Five . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

📄 Page 20

Contents xxi 17.8 Winston Churchill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 17.8.1 Example: Step Two . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 17.8.2 Example: Step Three . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 17.8.3 Example: Step Four . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 17.8.4 Correlation Between Experts . . . . . . . . . . . . . . . . . . . . . 311 17.9 Comparing Hitler to Churchill . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 17.10 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 17.11 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 18 Analyzing History with Video . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 18.1 Function Diagram for Analyzing Video . . . . . . . . . . . . . . . . . . . . . 321 18.2 Braking the Video Into Several Dimensions . . . . . . . . . . . . . . . . . 322 18.2.1 Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322 18.2.2 From Video to Frame . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 18.3 Machine Learning Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324 18.3.1 Recognizing the Candidates . . . . . . . . . . . . . . . . . . . . . . 324 18.3.2 Speak, Hands for Me . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 18.3.3 From Body Movements to Clocks . . . . . . . . . . . . . . . . . 327 18.4 Computing Body Language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 18.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328 18.6 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330 19 Fake History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331 19.1 Function Diagram of Fake History . . . . . . . . . . . . . . . . . . . . . . . . . 331 19.2 Copernicus Rule of Thumb for Fake History . . . . . . . . . . . . . . . . 332 19.3 Occam’s Razor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 19.4 Fakes Are Much Easier to Generate Than “Truth” . . . . . . . . . . . . 333 19.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334 19.6 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334 Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335 Appendix A: Mathematical Background for Computational History Using AI Prompts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349