3 ms·

Classics should in the post data scientific age be moved into subfield of computational linguistics. Literature is both empirical (See recent work on statistic

Classics should in the post data scientific age be moved into subfield of computational linguistics. Literature is both empirical (See recent work on statistical approach to computational literature studues) and subset of linguistics proper. The same thing holds for classics as a proper subset of literature. Problem isn’t’ really disbanding the classics department but not having classics studies computationally based or not handing field over to computational linguists.

DiggyJohnson 5y ago
I’ll bite. Why would it be valuable to analyze literature or classics computationally? Are you suggesting that they ought to be considered a subset of linguistics? Linguistics is about language, not the content of language.
- qualudeheart 5y ago
  The meaning of texts and language as Derrida tells us has a graph like structure. Each word in this sentenc eis defined in relation to other words and the meaning of the whole sentence can therefor be modelled as a graph. Modern neural networks— like those used at Palantir Technologies for link analysis- can be applied to the structure of any text including this post right here. GPT-3 can also be used for summaries of literature and biblical exegesis and will soon be an indispensable part of any computational linguist’s toolkit.
- UncleMeat 5y ago
  There is value in this. The digital humanities has given us new methods to approach the study of literature (and a great many other fields), which has been dominated by the method of close reading for a long time. For example, can scale our analysis of the transfer of techniques between works and use rigorous mathematical analysis to model influence and networks in a new way. There is a lot of garbage work in the digital humanities (like all fields) that is basically just doing word choice analysis with computers, but it truly is a new method that is available to researchers to put in their toolchest and new methods are almost always good.

DiggyJohnson 5y ago

I’ll bite. Why would it be valuable to analyze literature or classics computationally? Are you suggesting that they ought to be considered a subset of linguistics? Linguistics is about language, not the content of language.

qualudeheart 5y ago
The meaning of texts and language as Derrida tells us has a graph like structure. Each word in this sentenc eis defined in relation to other words and the meaning of the whole sentence can therefor be modelled as a graph. Modern neural networks— like those used at Palantir Technologies for link analysis- can be applied to the structure of any text including this post right here. GPT-3 can also be used for summaries of literature and biblical exegesis and will soon be an indispensable part of any computational linguist’s toolkit.
UncleMeat 5y ago
There is value in this. The digital humanities has given us new methods to approach the study of literature (and a great many other fields), which has been dominated by the method of close reading for a long time. For example, can scale our analysis of the transfer of techniques between works and use rigorous mathematical analysis to model influence and networks in a new way. There is a lot of garbage work in the digital humanities (like all fields) that is basically just doing word choice analysis with computers, but it truly is a new method that is available to researchers to put in their toolchest and new methods are almost always good.