Skip to content

Commit

Permalink
copyedits for Anna's two pieces
Browse files Browse the repository at this point in the history
  • Loading branch information
gwijthoff committed Sep 19, 2023
1 parent f427f8f commit a0fd245
Show file tree
Hide file tree
Showing 2 changed files with 10 additions and 10 deletions.
8 changes: 4 additions & 4 deletions content/issues/4/_index.md
Original file line number Diff line number Diff line change
Expand Up @@ -24,10 +24,10 @@ contributors:
- Camey Van Sant
---

This issue of *Startwords* explores palimpsests of sound, script, and place in the language and history of medieval China. The two projects featured here work through layers of time with the goal of reconstructing now-lost literary soundscapes and urban landscapes. As a scholar of medieval Chinese literary and cultural history whose work is grounded in traditional methods, (with some limited experience in quantitative methods), I read these papers as an outsider to computational work in the humanities, but with great enthusiasm for their potential to produce new knowledge. Both projects build on recent, ground-breaking scholarship---in philology, phonology, and archaeology---and design new tools for old questions. What did a millennium of phonological change erase from the ways that ancient Chinese texts produced meaning through script and sound? How did people in the medieval Chinese capital of Chang'an experience its spatial, cultural, and linguistic complexity? As you will see, the answers to these questions can be discovered if we allow scholarly and scribal practices of the ancient past to inspire the unconventional use of computational tools in the present.
This issue of *Startwords* explores palimpsests of sound, script, and place in the language and history of medieval China. The two projects featured here work through layers of time with the goal of reconstructing now-lost literary soundscapes and urban landscapes. As a scholar of medieval Chinese literary and cultural history whose work is grounded in traditional methods (with some limited experience in quantitative methods), I read these papers as an outsider to computational work in the humanities, but with great enthusiasm for the papers' potential to produce new knowledge. Both projects build on recent, groundbreaking scholarship --- in philology, phonology, and archaeology --- and design new tools for old questions: What did a millennium of phonological change erase from the ways that ancient Chinese texts produced meaning through script and sound? How did people in the medieval Chinese capital of Chang'an experience its spatial, cultural, and linguistic complexity? As you will see, the answers to these questions can be discovered if we allow scholarly and scribal practices of the ancient past to inspire the unconventional use of computational tools in the present.

"Of Sonorous Medieval Chinese Texts and NLP Model Training: Reading the *Jingdian Shiwen* 經典釋文 as Semi-structured Data," by Nick Budak and Gian Rominger, presents the design of a machine learning algorithm to parse the data of a monumental sixth-century dictionary, Lu Deming's *Explanation of Words in the Classics and Canons* (*Jingdian shiwen* 經典釋文). Important new work in Chinese historical linguistics has reconstructed the phonology of Old and Middle Chinese, just as recent decades of manuscript discoveries from tombs have revolutionized our knowledge of graphic variation in the Chinese script in the ancient through medieval eras. Budak and Rominger use this scholarship in their innovative design of a natural language processing (NLP) model to parse the semi-structured data of the *Jingdian shiwen*, a text that captures features of the early Chinese language that slowly disappeared from the phonology of spoken Chinese. The *Jingdian shiwen* is a highly productive text for their purposes: as a dictionary, it encodes both phonological and semantic data, provides textual excerpts and examples to gloss its definitions, and attempts to stabilize reading and pronunciation practices in the context of historical change. The text is, as the authors put it, "effectively a machine-readable dataset millennia before such machines would exist." After mining the phonological data of *Jingdian shiwen*, the authors will transform it into first Middle and then Old Chinese, to create a fully machine-readable dataset. The ultimate aim of Budak's and Rominger's work is to recreate the complex soundscapes of early Chinese texts, which depended heavily on rhyme, wordplay, punning, and other forms of graphic and phonic interplay, and to deepen our understanding of their rich polyphony and polysemy.
"Of Sonorous Medieval Chinese Texts and NLP Model Training: Reading the *Jingdian Shiwen* 經典釋文 as Semi-structured Data," by Nick Budak and Gian Rominger, presents the design of a machine learning algorithm to parse the data of a monumental sixth-century dictionary, Lu Deming's *Explanation of Words in the Classics and Canons* (*Jingdian shiwen* 經典釋文). Important new work in Chinese historical linguistics has reconstructed the phonology of Old and Middle Chinese, just as recent decades of manuscript discoveries from tombs have revolutionized our knowledge of graphic variation in the Chinese script in the ancient through medieval eras. Budak and Rominger use this scholarship in their innovative design of a natural language processing (NLP) model to parse the semi-structured data of the *Jingdian shiwen*, a text that captures features of the early Chinese language that slowly disappeared from the phonology of spoken Chinese. The *Jingdian shiwen* is a highly productive text for their purposes: as a dictionary, it encodes both phonological and semantic data, provides textual excerpts and examples to gloss its definitions, and attempts to stabilize reading and pronunciation practices in the context of historical change. The text is, as the authors put it, "effectively a machine-readable dataset millennia before such machines would exist." After mining the phonological data of *Jingdian shiwen*, the authors will transform it into first Middle and then Old Chinese to create a fully machine-readable dataset. The ultimate aim of Budak and Rominger's work is to recreate the complex soundscapes of early Chinese texts, which depended heavily on rhyme, wordplay, punning, and other forms of graphic and phonic interplay, and to deepen our understanding of their rich polyphony and polysemy.

"Toward a Deep Map of Chang'an," by Xin Wen, also builds on recent, revolutionary discoveries as well as traditional historical sources to conceptualize a layered, dynamic map of, Chang'an 長安 (located on the site of present-day Xi'an 西安), one of the largest and most diverse cities of the global medieval world. The explosive growth in the past few decades of archaeology on the site of Chang'an, the former capital of the Han (202 BCE-220 CE) and Tang (618-907) dynasties, gives Wen access to precise new GIS and material data for sites as varied as palaces, tombs, streets, and waterways in the city. Wen will combine this data with historical two-dimensional maps, historical accounts of Chang'an, and the most recent scholarship on urban centers in China to visualize and animate the city's social, cultural, and political history. This work has the potential to illuminate the experience of navigating the city---captured so vividly by people such as Tang poets and storytellers, and by Japanese monks who travelled to Chang'an---in multiple dimensions and across time.
"Toward a Deep Map of Chang'an," by Xin Wen, also builds on recent, revolutionary discoveries as well as traditional historical sources to conceptualize a layered, dynamic map of Chang'an 長安 (located on the site of present-day Xi'an 西安), one of the largest and most diverse cities of the global medieval world. The explosive growth in the past few decades of archaeology on the site of Chang'an, the former capital of the Han (202 BCE--220 CE) and Tang (618--907) dynasties, gives Wen access to precise new GIS and material data for sites as varied as palaces, tombs, streets, and waterways in the city. Wen will combine this data with historical two-dimensional maps, historical accounts of Chang'an, and the most recent scholarship on urban centers in China to visualize and animate the city's social, cultural, and political history. This work has the potential to illuminate the experience of navigating the city --- captured so vividly by people such as Tang poets and storytellers, and by Japanese monks who traveled to Chang'an --- in multiple dimensions and across time.

Budak and Rominger's language model and Wen's deep map are both works in progress. Much like Lu Deming's seventh-century commentary and Lü Dafang's eleventh-century map --- cumulative product of countless collaborations and influences that took years to complete --- the work of these digital humanists is a contribution to a set of scholarly inquiries that spans centuries. What we are treated to here is a portrait not only of what these projects might become, but also the ways that a sustained engagement with ancient forms of knowledge making can challenge us to rethink the predominant modes of understanding text and space in fields like data science and digital humanities.
Budak and Rominger's language model and Wen's deep map are both works in progress. Much like Lu Deming's seventh-century commentary and Lü Dafang's eleventh-century map --- cumulative products of countless collaborations and influences that took years to complete --- the work of these digital humanists is a contribution to a set of scholarly inquiries that spans centuries. What we are treated to here is a portrait not only of what these projects might become, but also of the ways that a sustained engagement with ancient forms of knowledge making can challenge us to rethink the predominant modes of understanding text and space in fields like data science and digital humanities.
12 changes: 6 additions & 6 deletions content/issues/4/institutional-landscape/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -10,13 +10,13 @@ date: 2022-08-01
images: ["issues/3/llm-limit-case/images/llm-limit-case-social.png"]
---

In March of 2023, Harvard University hosted "[Tools of the Trade: The Way Forward](https://sites.harvard.edu/tools-of-the-trade/)," an exciting three-day conference surveying recent trends in East Asian digital humanities. With the support of a wide range of [funding sources from Harvard and abroad](https://sites.harvard.edu/tools-of-the-trade/sponsors/), over 150 participants attended from the United States, Europe, and East Asia. The papers and plenary sessions covered over 3,000 years of East Asian history, included multiple East Asian languages and scripts, and featured new databases, platforms, and methodologies in digital humanistic research.
In March of 2023, Harvard University hosted "[Tools of the Trade: The Way Forward](https://sites.harvard.edu/tools-of-the-trade/)," an exciting three-day conference surveying recent trends in East Asian digital humanities (DH). With the support of a wide range of [funding sources from Harvard and abroad](https://sites.harvard.edu/tools-of-the-trade/sponsors/), over 150 participants attended from the United States, Europe, and East Asia. The papers and plenary sessions covered over three thousand years of East Asian history, included multiple East Asian languages and scripts, and featured new databases, platforms, and methodologies in DH research.

The participants represented an equally wide range of national centers and East Asian, European, and U.S. institutions; unsurprisingly, junior scholars were especially active throughout the conference. As Alíz Horváth and Hilde De Weerdt noted in their recent introduction to a special issue of *International Journal of Digital Humanities* (2023) 4:1-4 on East Asian digital humanities (DH), the rapid expansion of computational and data-driven research in East Asian studies is everywhere in evidence: in new dissertations, journals, conferences, job opportunities, and workshops.[^1]
The participants represented an equally wide range of national centers and East Asian, European, and U.S. institutions; unsurprisingly, junior scholars were especially active throughout the conference. As Alíz Horváth and Hilde De Weerdt noted in their recent introduction to a special issue of *International Journal of Digital Humanities* on East Asian DH, the rapid expansion of computational and data-driven research in East Asian studies is everywhere in evidence: in new dissertations, journals, conferences, job opportunities, and workshops.[^1]

Interest among East Asianists in new tools and methods can be seen among scholars across the globe, but it is clear that East Asian nations are leading the way in terms of public funding for this new work at top universities, libraries, and national centers. The Harvard conference also, and perhaps unexpectedly, shed light on the uneven institutional landscape in North America for DH research in East Asian studies and prompted reflections on how institutions might collaborate to promote this work more effectively and systematically. In the current structure of graduate education in the U.S., many Ph.Ds are trained in area studies departments at institutions that may or may not have a DH center like those at Princeton, [Stanford University,](https://digitalhumanities.stanford.edu/) and the [University of Virginia](https://dh.library.virginia.edu/), for example. They may also have limited access to skills workshops aimed at humanists, such as those that have been sponsored at the annual meeting of the [Association of Asian Studies](https://www.asianstudies.org/), the field organization that has also launched [new DH fellowships](https://www.asianstudies.org/grants-awards/striving-for-diversity-equity-and-inclusion-in-asian-studies-humanities-grants-for-asian-studies-scholars/aas-fellowships/#aas-digital-humanities-fellowships). Senior advisors may be slow to recognize the potential of DH in East Asian studies and many junior scholars are necessarily self-taught in the tools, methods, and software that are changing their fields.
Interest among East Asianists in new tools and methods can be seen among scholars across the globe, but it is clear that East Asian nations are leading the way in terms of public funding for this new work at top universities, libraries, and national centers. The Harvard conference also, and perhaps unexpectedly, shed light on the uneven institutional landscape in North America for DH research in East Asian studies and prompted reflections on how institutions might collaborate to promote this work more effectively and systematically. In the current structure of graduate education in the U.S., many PhDs are trained in area studies departments at institutions that may or may not have DH centers like those at Princeton, [Stanford University,](https://digitalhumanities.stanford.edu/) and the [University of Virginia](https://dh.library.virginia.edu/), for example. They may also have limited access to skills workshops aimed at humanists, such as those that have been sponsored at the annual meeting of the [Association for Asian Studies](https://www.asianstudies.org/), the field organization that has also launched [new DH fellowships](https://www.asianstudies.org/grants-awards/striving-for-diversity-equity-and-inclusion-in-asian-studies-humanities-grants-for-asian-studies-scholars/aas-fellowships/#aas-digital-humanities-fellowships). Senior advisors may be slow to recognize the potential of DH in East Asian studies and many junior scholars are necessarily self-taught in the tools, methods, and software that are changing their fields.

National organizations like the [Mellon Foundation](https://www.mellon.org/search/digital%20humanities) and the [National Endowment for the Humanities](https://www.neh.gov/divisions/odh/resources-for-applicants-to-the-NEH-office-of-digital-humanities) have been generous, enthusiastic funders of digital humanistic research, including training programs at a range of smaller and medium-sized institutions; such programs are of course highly competitive and limited in number. But given the rapid growth and significance of computational research in East Asian studies, as well as some of the language- and script-specific challenges that East Asian data and corpora pose, now seems to be a good moment for East Asianists *across* institutions to strategize about sharing resources and creating new opportunities --- in order to increase access for more scholars and build out East Asian humanities more broadly.
National organizations like the [Mellon Foundation](https://www.mellon.org/search/digital%20humanities) and the [National Endowment for the Humanities](https://www.neh.gov/divisions/odh/resources-for-applicants-to-the-NEH-office-of-digital-humanities) have been generous, enthusiastic funders of DH research, including training programs at a range of smaller and medium-sized institutions; such programs are of course highly competitive and limited in number. But given the rapid growth and significance of computational research in East Asian studies, as well as some of the language- and script-specific challenges that East Asian data and corpora pose, now seems to be a good moment for East Asianists *across* institutions to strategize about sharing resources and creating new opportunities --- in order to increase access for more scholars and build out East Asian humanities more broadly.

Many possibilities spring to mind, some of which are already underway, such as:

Expand All @@ -26,6 +26,6 @@ Many possibilities spring to mind, some of which are already underway, such as:

- collaboration among libraries with strong East Asian digital resources to share (within necessary restrictions) materials and tools.

Whether or not scholars wish to incorporate data and computational methods in their research on East Asia after having access to these opportunities, such efforts would increase awareness of their potential, work to lessen the barriers to understanding new methodologies and tools, and increase digital humanistic literacy within and across national borders.
Whether or not scholars wish to incorporate data and computational methods in their research on East Asia after having access to these opportunities, such efforts would increase awareness of the potential of those methods, work to lessen the barriers to understanding new tools, and increase DH literacy within and across national borders.

[^1]: Alíz Horváth and Hilde De Weerdt, "Special Issue on Digital Humanities and East Asian Studies," *International Journal of Digital Humanities* (4, no. 1), February 1, 202, <https://doi.org/10.1007/s42803-022-00064-5>.
[^1]: Alíz Horváth and Hilde De Weerdt, introduction to “Special Issue on Digital Humanities and East Asian Studies,” ed. Alíz Horváth and Hilde De Weerdt, *International Journal of Digital Humanities* 4, nos. 1–3 (Feb. 2023): 1–4, <https://doi.org/10.1007/s42803-022-00064-5>.

0 comments on commit a0fd245

Please sign in to comment.