Riding the wind of rebirth

"However, in this census we discovered that they have not disappeared. They are just hidden in the dialects of various places, waiting for us to discover them."

"In other words, scholars in the past have discovered quite a lot, but they could only record the phenomenon, or the 'recipe', but could not record the 'taste'. And those who later looked at the 'recipe' they left behind could only imagine it."

"My master's thesis is to solve this problem." Zhou Zhi continued, "Chinese characters are logographic characters and do not directly represent pronunciation. However, the vast majority of Chinese characters are phono-semantic characters, and characters with the same phonetic component often have relatively similar pronunciations. Although after more than two thousand years of evolution, it is entirely possible that they have become very different in Middle Chinese and modern dialects, but since the pronunciation of the phonetic component and the character are very similar when the character was created, it is not impossible to infer the information of the lost initial consonant based on this."

"This is something that previous scholars have studied very thoroughly, but they all overlooked 'sampling'."

"For example, Min, from our sampling, should have differentiated from the late ancient Chinese, because they contain levels that cannot be deduced from the rules of Qieyun. For example, the following dental initials can be divided into ten pronunciations in the original Min language: clear short, clear turn, clear charcoal, voiced straight, voiced long, voiced head, nasal south, nasal year, side come, side old. The differences in other modulation positions on stops and nasals are similar to the phenomenon in the table above."

"From this we can roughly infer that Ancient Chinese had two sets of nasal sounds, voiced and unvoiced, and only the voiced nasal sounds generated the Middle Chinese nasal initials, while in Proto-Min both sets of initials were completely preserved."

"However, this is not the crux of the problem. The crux of the problem is that the current reconstruction scholars do not understand the original Min language and do not know the expression methods of these two sets of ancient nasal initials. Therefore, they introduced the reconstruction method inferred by Western scholars into the reconstruction. And that set of inferences is based on Western phonetic writing, which is very different from Chinese phonetic writing and is fundamentally wrong."

"Even though we have not yet reached a consensus on the origin of ancient Chinese, and different scholars have given very different complex consonants, we can at least use the existing text materials, combined with the initials and finals samples in various dialects, to look for traces of ancient rhythms in these samples, find matching examples, and use them as materials to start building the phonetic system of Middle Chinese."

"This is the only way to get back to the original source, rather than reading the Book of Songs as if it were full of foreign language features such as tongue-twisting."

"At least we can be very sure of one thing, that is, the Chinese language has been a language different from foreign languages ​​since ancient times. At least when Chinese characters were formed, the pronunciation of this language was already very different from that of foreign languages. This set of characters is extremely concise compared to foreign phonetic characters, so it must be matched with a tone system to complete complex expressions."

"The two complement each other. It would be rogue to ignore one and talk about the other. We need to correct this phenomenon instead of following the crowd."

"Good! Good boy, your knowledge is improving again!" Qi Lao's voice sounded outside the door: "Seeing you messing around in Daliang Mountain, I thought you had abandoned all your knowledge!"

"How long have you been hiding outside and eavesdropping?" Old Master Wang looked up, then pointed at the pair of covers on the table and said to Old Master Qi with a smile, "Look, there are still some young people who can recognize this set of things."

"So, Elbow, your thesis is based on the reconstruction of Medieval phonetics?" Grandpa Qi asked Zhou Zhi as he sat down to admire the porcelain.

"No, my topic is "Engineering Construction of Digital Information Environment Based on the Simulation of Medieval Chinese Phonetics". It mainly talks about ideas, design, architecture, engineering construction, and construction management. It is mainly to let experts understand how to use science and engineering knowledge to deal with liberal arts projects."

"In fact, I think oracle bone inscription combination is also a very suitable research direction for you. And with your current achievements, it is more than enough to exchange for a doctorate degree from Peking University." Mr. Qi asked Mr. Wang: "I don't think I have heard of any doctor who has presided over the National Social Science Fund Project and the 863 Program Engineering Project? Elbow is the direct leader." "Old Gu and his school are like this, they like to hide things," Mr. Wang sighed, "Don't ask, if you ask, just reflect on your own lack of knowledge and the inability to write books. Elbow is already a top-notch show-off now."

"Even if it is not the topic of the defense, we will still continue to delve into the oracle bone splicing project." Zhou Zhi smiled and said, "As one of the prelude projects for the Xia, Shang and Zhou dynasties, we dare not delay it. By the way, it is a good opportunity to report our progress on this project to the two elders."

So far, millions of pieces of oracle bone inscriptions have been discovered and sorted out. Of course, these millions of pieces are not all oracle bones. Most of them exist in the form of rubbings and pictures.

Previously, the splicing was carried out using the graphic edge recognition technology developed by Clover, but the project progress only advanced 5%, because there are many limitations to splicing based solely on the edges. For example, the splicing is successful, but the text periods are different, the meaning is incoherent, etc., which accounts for a very high proportion.

In addition, due to various reasons such as different image acquisition methods, different sizes, different contrasts, and different shooting angles, a large number of oracle bones cannot be edge-matched.

After introducing Xiaozhi to this project, Mai Xiaomiao rebuilt the algorithm model and the efficiency was greatly improved.

Currently, the progress of the splicing project has reached 30%, which has reached the engineering limit.

The engineering limit means that the oracle bones that can be spliced ​​together have basically been spliced ​​together, and the remaining ones are scattered oracle bones that cannot be spliced ​​together. They only exist to study the meaning of the individual words on them, and the meaning of the text can no longer be connected.

But even so, the project has reached a height that was unattainable by predecessors. Currently, the oracle bones that have been approved for splicing and forming meaningful texts number more than 6,000 pieces, which has expanded the oracle bone inscriptions by more than three times!

With so much more text, many of the previously "new words" have become "familiar words" and can be deciphered.

In addition to interpreting new characters, the historical information recorded in these texts is even more important. It provides very important historical evidence for the Shang Dynasty chronicles, and also provides follow-up research materials for the Xia, Shang and Zhou dynasty dating project.

With Xiaozhi, we have sufficient computing power to add as many analysis dimensions as possible, such as the roots of oracle bone inscriptions, calligraphy features, character composition rules, corruption, simplification, the evolution of the six scripts, etc. The accuracy of the resulting splicing results is greatly improved.

Now oracle workers all over the country are very busy, so busy that they can be said to be in a mess. All this is brought about by the oracle splicing project.

So even Mr. Qi felt it was a pity not to include this project as part of Zhou Zhi's master's thesis defense.

But everyone seems to have overlooked that Zhou Zhi was preparing to get his master's degree one year early. (End of this chapter)