什么是数字人文？

Chapter 1: What Is Digital Humanities?

The Super Librarian's Magic

Imagine walking into a library that contains every book ever written in human history. Now imagine asking the librarian: "Which books from 500 years ago talked about dragons?" And within five seconds, the librarian hands you a list of 347 books from exactly that era, sorted by relevance, complete with the exact passages where dragons are mentioned.

Can a human librarian do this? No way. A human brain can hold maybe 10,000 books worth of information at best. But a computer? A computer can hold and search through *billions* of books simultaneously.

This magical librarian doesn't actually exist in physical form—but in the world of Digital Humanities, we build computer programs that can do exactly this kind of magical searching and analysis. We call this field Digital Humanities (DH), and it's one of the most exciting ways that computers are helping us understand human history, literature, art, and culture.

> Digital Humanities is the art of asking big questions about human culture—and using computers to find the answers. When we combine the power of computers with the curiosity of historians, writers, and artists, we can discover things that neither humans nor machines could find alone.

Let me tell you about Maria, a digital humanist who studies ancient Greek poetry. One day, she wondered: "Did Greek poets from different cities use different words to describe love?" This is a question that would take a human researcher decades to answer—reading through thousands of ancient poems, making notes, comparing patterns. But Maria wrote a computer program that read all 12,000 surviving ancient Greek poems in just three hours. The computer found patterns that human scholars had missed for hundreds of years: poets from Athens used words related to beauty and harmony, while poets from Sparta used words related to courage and sacrifice. These patterns revealed something profound about the different values of these two ancient cities.

This is the magic of Digital Humanities. It's not about replacing human understanding—it's about giving humans superpowers to explore our cultural heritage in ways that were simply impossible before.

---

A Brief History: From the First Computers to AI

The Year 1949: Everything Began with Counting Words

The story of Digital Humanities starts in 1949, when a researcher named Father Roberto Busa connected a computer to a massive project: analyzing every word in the works of Thomas Aquinas, a famous philosopher from the 1200s. Father Busa wasn't trying to replace human scholarship—he was trying to do something humans physically couldn't. The Saint's complete works contained over 11 million words in Latin, and understanding all the connections between these words required computational help.

This was the very first example of what we now call Digital Humanities. The computer couldn't understand philosophy or Latin poetry—it could only count and sort. But by counting, it revealed patterns that human eyes might have missed.

The 1960s-1980s: Building the Digital Libraries

As computers became more powerful, scholars began creating digital versions of important texts. Instead of physical books that could only be read by one person at a time, digital texts could be copied infinitely and analyzed by computers around the world.

During this period, important projects emerged:

• **The Oxford English Dictionary** began using computers to track how English words changed meaning over 1,000 years
• **The Perseus Digital Library** at Tufts University made ancient Greek and Latin texts searchable by computer
• Historians started encoding historical documents so computers could find connections between events, people, and places

The 1990s: The Internet Changes Everything

When the World Wide Web exploded in popularity during the 1990s, Digital Humanities exploded too. Suddenly, scholars could share digital texts and tools with anyone in the world. The text of thousands of books, millions of historical documents, and countless images became available online.

During this era, researchers developed new methods:

• **Corpus linguistics tools** that could analyze millions of sentences to understand how language actually works
• **Geographic Information Systems (GIS)** that could map historical events onto modern maps, revealing patterns in migration, trade, and warfare
• **Text encoding standards** that let scholars around the world mark up documents in consistent ways, making it easier for computers to understand the structure of texts

The 2000s-2010s: Big Data Comes to the Humanities

Google scanned millions of books. Archives digitized millions of photographs. Scientists created massive databases of genetic information, climate records, and archaeological finds. The humanities suddenly had access to "big data"—and with it, the ability to answer questions that required seeing the big picture, not just individual examples.

This period saw major breakthroughs:

• **Google Books Ngram Viewer** let anyone see how word usage changed over 500 years of published books
• **Network analysis** revealed hidden social connections in historical networks—finding out who was really influencing whom in ancient Rome or medieval Florence
• **Distant reading** emerged as a philosophy: instead of reading every book carefully (close reading), scholars could analyze millions of texts at once to find broad patterns in literature and culture

The 2020s: Artificial Intelligence Joins the Team

Today, the newest Digital Humanities projects use artificial intelligence and machine learning. AI systems can:

• Translate ancient languages that humans have forgotten how to read
• Identify which artist painted an unidentified artwork by analyzing brush strokes
• Reconstruct damaged ancient documents from just fragments
• Generate hypotheses about historical connections that human scholars can then investigate

But here's the crucial point: AI doesn't replace human judgment. A computer might suggest 50 possible translations for a damaged ancient tablet, but a human historian must decide which translation makes the most sense historically and culturally. Digital Humanities is always a partnership between human curiosity and computer power.

---

The Three Superpowers: Volume, Pattern, Visualization

What exactly can computers do that humans cannot? Digital Humanities tools typically give us three superpowers: the ability to handle massive Volume, the ability to find hidden Patterns, and the ability to make information Visual.

Superpower 1: Volume — Reading What No Human Can Read

Human readers can process perhaps 200-300 words per minute. A fast reader might get through one novel every few days. Over an entire lifetime, a dedicated scholar might read 10,000 books. But there are over 130 million books in existence, and that's not counting the billions of articles, letters, diaries, and documents in archives around the world.

Computers can read all of these—well, at least the ones that have been digitized. Google's database of scanned books alone contains over 40 million books. A computer can search through all of them for a specific phrase in less time than it takes you to blink.

This ability to handle massive volume matters because some of the most important questions in humanities require seeing everything, not just a sample:

• **Which words appeared in English literature for the first time in 1850?** We need to check millions of texts to be sure.
• **Were there any women who wrote about science in the 1700s?** We need to search archives that have never been fully catalogued.
• **Did medieval Europeans have words for emotions that we don't have today?** We need to examine texts from dozens of languages across hundreds of years.

Volume isn't just about quantity—it's about completeness. When computers analyze everything, we can be confident that ou