Tolstoy Everywhere · Princeton

![workshop logo cropped](images/workshop-logo-cropped.png)

### Tolstoy Everywhere
Unleashing the Information Hidden in the 90-Volume Collected Works

[Boris Orekhov](http://nevmenandr.net/bo.php) ¹  ·  [Frank Fischer](https://www.hse.ru/en/org/persons/182492735) ¹ ²

¹ Higher School of Economics, Moscow 
² DARIAH-EU

[Slavic DH Workshop](https://cdh.princeton.edu/events/2019/05/slavic-dh-workshop-russian-literary-studies-digital-age/)  ·  Princeton University  🇺🇸  ·  28 May 2019, 1:30–3:00 p.m.

### **Chapters**

1. The Tolstoy Project
2. The 91st Volume – An Exceptional Index
3. The Web App
4. Summary

---

# Chapter 1.

### The Tolstoy Project

### Digital Edition

- digital edition of collected works in 90 volumes, published between 1928 and 1958: novels, short stories, plays, letters, diaries, … (~ 46,000 pages)
- cooperation between [Tolstoy Museum](http://tolstoymuseum.ru/) and Higher School of Economics, Moscow

### Digitising Tolstoy

![movie](images/startup.jpg)

- *Startup* (2014), movie about the history of Yandex
- 0:32:46: “The complete works of Leo Tolstoy is now uploaded to the system.”
- in the 1990s when the movie takes place, Tolstoy was not yet fully digitised 😊

### Digitising Tolstoy

- it started with [„All of Tolstoy in One Click“](http://tolstoy.ru/projects/tolstoy-in-one-click/): all volumes scanned and OCRed (supported by ABBYY)
- corrections were crowdsourced
- abundant media coverage: [New Yorker](https://www.newyorker.com/books/page-turner/crowdsourcing-tolstoy), [The Guardian](https://www.theguardian.com/books/2013/oct/16/all-leo-tolstoy-one-click-project-digitisation), etc.
- texts available in PDF, EPUB, FB2, MOBI and XHTML (see http://tolstoy.ru/creativity/90-volume-collection-of-the-works/)

### Our Contribution

- texts converted to [basic TEI](https://github.com/tolstoydigital/TEI)
- [online search interface](http://search.effits.ru/)
- richer markup
- onthologies
- linked open data
- where do we start? index!

---

# Chapter 2.

### The 91st Volume – An Exceptional Index

### Index

- indexes for a long time and not by chance accompany academic editions
- book not as actual reading device, but as instrument (looking for references and quotes)
- indexes make this work efficient
- proper names in a fiction text have a special function

### Proper Names

- draw attention (including capitalisation)
- heavily loaded associatively
- often have a special spelling (foreign names)
- important factor of the overall structure of a text, especially if they appear with high density

### 91st Volume

![91st volume (cover)](images/91st-volume-cover.jpg)

- supplement volume containing indexes of works and proper names (fiction and non-fiction works like diaries, letters)
- 16,256 registered entries
- good starting point to create an onthology

### Onthology

| entity                     | ID   | wikidata  | comment               | geo                  |
|:--------------------------:|:----:|:---------:|:---------------------:|:--------------------:|
| Лао-Цзы (Лао-тзе)          | 7627 | Q24446595 |                       |                      |
| Алексеева Мария Васильевна | 272  |           | дочь В. И. Алексеева. |                      |
| Алексеевка Тульской губ.   | 274  |           |                       | 53.590620, 38.093302 |
| Алексеевка Самарской губ.  | 275  |           |                       | 52.580285, 51.275200 |

### Next Step: Web App

- web app: http://index.tolstoy.ru/
- target audience: specialists as well as enthusiasts
- general question: even with digitised full texts at hand, how can we benefit from the structuring and registration efforts of the past?

---

# Chapter 3.

### The Web App

![web app statistics](images/webapp-statistics.png)

brown: number of references in the text – beige: number of mentions in the comments

### Search for Entities

- original index functionality retained: mapping proper names to volumes and pages
- one-click direct jumps to pages
- added functionality:
 - flexible search: entering „ava“ will list results like „Poltava“, „Bavariâ“, „Abdulla-al’-Mamun Zuravardi“
 - word-cloud representation conveys a first idea about most frequent words in the corpus

### Flexible Search

![search](images/001.png)

### Case in Point: Dürnstein (Austria)

![search](images/duernstein.jpg)

### Search With Tag

![tag](images/003.png)

### Alphabetical List

![tag](images/002.png)

### Co-Occurrence Calculator

![calculator](images/008.png)

### Entities Network

![Victor Hugo co-occurrence network](images/screenshot-victor-hugo.png)

word-cloud representation for entry ‘Victor Hugo’ showing co-occurring named entities – URL: http://index.tolstoy.ru/person/4079/

### Individual Pages for All Entities

![card](images/009.png)

### Studying Life and Works of Leo Tolstoy by Means of Network Analysis (1/3)

![whole graph](images/general-graph.jpg)

co-occurence graph for all 90 volumes – network is downloadable and can be customised – URL: http://index.tolstoy.ru/general/graph/

### Studying Life and Works of Leo Tolstoy by Means of Network Analysis (2/3)

- studying co-occurrences of proper names in the same environment (in our case, on the same page) help to understand larger contexts
- example:
 - the Hindu scripture „Bhagavad Gita“ is found on 5 pages in Tolstoy’s Complete Works, see http://index.tolstoy.ru/person/2138/
 - it shares these five pages with a total of 43 other names mentioned

### More Examples

- the proximity of these mentionings is not accidental, in our case they form a kind of an „India cluster“ containing works like „Hitopadesha“, „Dhammapada“, „Vamana Purana“, or names like Sri Ramakrishna Paramahansa
- for Tolstoy, these titles and names are part of a set of carriers of philosophical knowledge, and are associated with names like Xenophon, Montaigne, Montesquieu, Pascal, Skovoroda, Socrates
- these networks provide great opportunities for understanding the whole range of Tolstoy’s interests and ideas

### Studying Life and Works of Leo Tolstoy by Means of Network Analysis (3/3)

Nodes ranked by weighted degree in the fictional network (vols. 1–45):*

1. Россия/Русь (2778)
2. Наполеон I Бонапарт (2712)
3. Москва (2689)
4. Александр I Павлович (1873)
5. Петербург (1830)
6. Кутузов Михаил Илларионович/Голенищев-Кутузов (1515)
7. Франция (1390)
8. Европа (1212)
9. Англия (938)
10. Германия (789)

* Many thanks to Daniil Skorinkin for providing this data.

### Exploring the Text Structure

![heatmap](images/007.png)

---

# Summary

- the work of the editors of Tolstoy’s complete works is exceptional
- it allows you to systematize the entities in the texts of Tolstoy and build an ontology without NLP
- entities allow you to reconstruct the way Tolstoy thinks
- entities allow you to see the whole picture of the text
- making indexes machine-readable is an additional way to research the Collected Works of single authors or similar printed text collections that come with rich and detailed indexes
- functionalities of traditional indexes are not only retained, but also enhanced (flexible searches, visualisations of co-occurrences, etc.)

Thanks.

– https://hum.hse.ru/digital/ –