One Republic of Learning

The humanities have embraced quantitative analysis and will continue to do so. But like others, I see this as a complement to interpretation, not a replacement of it.

True story. I'm interested in a word, in various forms. I can chart its arc of usage historically very easily (Google ngram). I can also find out how many times certain authors and groups of authors use it.

The problem with this word is that it means two quite distinct things. (From a humanist's point of view, it's not a problem--it's part of what makes the word interesting.) It would be delightful if an algorithm could tell which sense was being used in each case. But as things stand, you actually have to read the sentences in which the word appears to tell.

But that's not all. As you start noting the broad distinction between the two senses of the word, you start realizing there are subtler distinctions. Sometimes the word has a negative connotation, sometimes a neutral one. Sometimes it characterizes a process, sometimes a thing. And so on and on. And you have to read yet more to figure all this out: the paragraph, or page, or chapter, or the whole book.

So digital humanities has actually driven me back deeper into literary criticism. This is all great. Maybe not for somebody who wants to tell a triumphal tale of quantification replacing interpretation. But great for humanists, who usually realize that reality is more complicated than such tales declare.

Digitising information enables enormous amounts of material to be made more easily available to a very large public. This is marvellous. What many funders and administrators most love about digital humanities is that its focus on data ensures that scholars operate largely as archivists or librarians, bringing in more research money while producing less controversial results. Critical thinking, and challenging authority or even recognising how knowledge has been used to buttress authority, which have been central to the humanities since at least the time of Socrates, are no longer part of the equation. The analysis offered on websites of the treasure troves that digital humanities can make easily available is usually a significant dumbing down of even the good general books on the subject one can find in almost any public library, let alone be assigned in a university course on the subject, but increasing these last two are no longer being consulted by any but the most advance scholars because they do not turn up as easily on google.

I hope they don't Mr. Leroi. In the last 15 years humans have been deprived already enough of our humanities. Feelings, sense for beauty, and other reasons for disputes, without which humans won't stay human, should be left alone by mathematics. One enjoys the music of J.S.Bach even more without analyzing its maths, while listening to its beautiful sounds.

Digitization made my reaction to this presentation easily cut-and-pasteable:

Full of high sentence, but a bit obtuse;
At times, indeed, almost ridiculous—
Almost, at times, the Fool.

Leroi should be more careful at describing who the actors will be when the algorithms become very powerful. Algorithms will not be "extracting meaning from data," though they might indicate trends from which we humans will produce meaning. Meaning itself is the Bakhtinian chronotopic artifact of a human. It is the Foucaultian epistemic moment, not a statistic. After all, we have had fabulous numbers related to the humanities for quite a long time. Consider sales of books, advances paid to authors, salaries to professors, and speaking fees paid to critics. Those statistics may have produced indications about what has value in a consumer-driven world, but they do not give us insight into the meaning of the underlying humanistic production.

"The old order changeth, yielding place to new,
And God fulfills himself in many ways,
Lest one good custom should corrupt the world."

The creators of new art will use the new tools that information science has provided. It is those who merely criticize old art that lament the gaining of these new powers.

I recall reading once that Hunter S. Thomson retyped the works of F. Scott Fitzgerald just to figure out how he "did it". Tomorrow's original writers are probably sitting in front of their laptops figuring out how to visualize their models' stats. And a decade from now, the critics will be whining that no one cares about the old algorithms any longer...

"To create is divine, to reproduce is human." Man Ray (creator of the "Shakespearean Equations")

Chuckle. fadoodle. the article take the 2015 naive chuztpah award The bright graduate data boy will 'prove' whatever he likes, by his definition . When he has algorithms that tell me precisely how and why she walks in beauty like the night, or what makes the sea wine dark I will walk humbly away. Meantime no doubt he can tell me the frequency of adverbs in the complete works of Senator Saltonstall. I am not grateful.

The question whether Shakespeare wrote a particular passage is not the same as how to interpret that passage. The former is more amenable to quantitative methods, and the latter more susceptible to hermeneutics. Methodologies are naturally selective of their preferred questions.

We've had the mathematics of music since the time of Pythagoras (6th century BCE). It doesn't tell you what moves a listening audience.
The use of quantitative methods in the humanities may answer a number of questions, but "what is the beautiful?" is not among them.

Thought-provoking essay. What disturbs me, however, are the overtones. The insinuation that lovers of arts will come out to defend the humanities with false arguments, using “well-honed” vocabulary.

I think what’s more annoying is the condescending tone of the piece.

Sure, literature might be able to be codified to find out how many commas, conjunctions or cliched phrases fill a novel. But what a computer cannot do is pull meaning from art. Why? Because meaning is individual. Thus, the phrase: What does this mean to YOU? Even if we were to all believe that Picasso's work is wonderful, there are still gradations within the word wonderful, not to mention the people who look at his work and think it’s garbage.

I think what the author missed, or doesn’t get, is that words have always been only an approximation of what we feel and observe as human beings. So now we’re going to have some computer tell us what Jane Austen intentions were? Please. Whatever the computer comes up with, there will be a million people who have a slightly different take on the issue.

Thus, yes, we still need critics, they are still valuable, because what they perceive cannot be forced into the exacting digital terms of a computer. And isn’t that just like life? So much of it is NOT black or white, so much of it is NOT exacting. There’s a lot of gray. But then again, the computer knew that, right?

"Many things that are countable, do not count.
Many things that count, are not countable".
.....Albert Einstein

Algorithms are nothing more than poetry or musical scoring. The Algorithmician selects discrete elements that are woven into a finished piece that the algorithmician feels with achieve his personal motives.

Algorithmicians, like poets and composers, will achieve discernibly different results working within the same genre. Think of the Blizzard That Wasn't recently in NYC. The European Model missed the mark, the American Model didn't. One problem, two algorithms, two different results. Beethoven vs. Lennon & McCartney. Keats vs. Robert Frost.

One of the many hubrises of the computational crowd is that they think that natural science and its practitioners are objective and non-interpretive. Just take raw data and plug it into the formula. Nonsense.

Doctor Leroi and his normal scientists would do well to consider the branch of humanities known as the history and philosophy of science and mathematics. Sitting at the knees of the giants might instill a little humility.

"If the rudiments of a new cultural science are visible, so are its limits. There is one great difference between human and natural things: The former have meaning; the latter do not. That is why the humanities are filled with critics and the natural sciences are not: Critics tell us what artifacts mean."

True, that. Reducing Beethoven's String Quartet Opus 131 to, say, a statistical printout of how many C-sharps occur in the opening fugue reveals nothing
about the work's meaning. To learn more about that, I would turn to a critic such as Joseph Kerman. Meaning is the essence of what the arts and humanities have to offer and the Digital Humanities simply miss the point.

As a psychologist who has seen, first hand, who limited quantitative methods are for understanding the mind and behavior, I'm quite sure that those who do believe in quantitative methods will not be fazed by past failures.

Promissory materialism reigns in consciousness studies (we may have absolutely no idea how a non-conscious, non-living "something" that we can't define, gives rise to sentience, and self-awareness, but by "God" we will). I'm sure that those who believe in the catechism of promissory materialism will have no difficulty transferring their faith to the domain of the humanities.

And commenters will no doubt prove that those who believe in the catechism are simply incapable of understanding what is lost when the immeasurable world of quality is reduced to quantification.

The faith is so widespread it is simply taken for granted. One day, religious scholar Huston Smith was talking in the MIT cafeteria with a colleague who taught science at MIT. The scientist was trying to articulate the essential difference between the humanities and the sciences."

"I've got it!" he finally exclaimed. "I count, and you don't."

Well, now that humanities scholars finally can "count", we will finally have achieved the ultimate nihilist goal: Everything can be counted and nothing will count.

Science may answer questions that can be formulated in a digital context. But the humanists, whether they are scientists or liberal artists, must imagine the road ahead, contimplating the alternatives.

This pean to digitization proceeds upon an important assumption--that the subjective criticism of the sort practiced by Bloom is functionally dead and that in the conflict between the old and the new forms of criticism the old must inevitably cede ground to the new.

Humanistic studies suffers from the perception that it is merely about subjective interpretations, and since we all have subjective interpretations it is of no value; it's the "I could do that" problem. Political criticism (e.g. Marxism, feminism, etc.) driven by ideological agendas, has largely had its day and audiences are hungry for something new--something that seems freed of the subjective and ideological biases of much recent criticism.

Digital humanities appears to be the answer to that problem. The digitization of texts and quantification of literary studies have indeed opened up new avenues of research. At the same time, we should be wary of suggesting that this is the answer to the perennial problem of the value of humanistic criticism.

Mr. Leroi solves this problem with another assumption--that criticism should be a science. If you accept this assumption, then this is a persuasive argument, but there is another, overlooked option: to understand criticism as an art.

When the critic is seen as an artist instead of a (social) scientist, the "I can do that problem," while it does not disappear, shifts its ground of value.

Could quantification play a role in the humanities? Sure, but in a complementary manner, not as a wholesale replacement for established ways of conducting humanities scholarship. The author falls into the trap of creating an artificial divide between the sciences and the humanities. And along the way, the author loses sight of a key point: much of the humanities is about perceptions and worldviews. This is the realm of interpretation, one which science cannot touch. So, let's celebrate the ways scientific approaches complement the humanities, but let's also avoid the trap of assuming science is the new be-all and end-all answer to age-old questions regarding interpretation of humans' place in the universe.

If a more scientific approach can end the reign of postmodernism, subjectivism, and all these other unproven psychological theories which have been accepted like a religion, I'm all for it.

Yes, let's turn everything into data that can be measured and quantified. I don't know enough about this field to understand if it will or will not replace traditional scholarly research but I can only compare this essay to Bruni's essay on the humanities from yesterday. Where he talks about the impact on his understanding of King Lear by a gifted teacher:

"She showed how that simple request harbored such grand anguish, capturing a fallen king’s hunger for connection and his tenuous hold on sanity and contentment. And thus she taught us how much weight a few syllables can carry, how powerful the muscle of language can be."

We care too much about data, and not enough about those emotional experiences that advance learning. This emphasis on data, rather than the individual child is what is destroying education in this country. Perhaps we will learn more by applying algorithms to literature and art, but will we lose the emotional connection of those disciplines? Are we merely machines? Are we attempting to explain away creative thought and make it merely a mechanical process?