Like Caleb, I found this chapter a little less illuminating and more basic. Moretti discusses what makes can do and how, like networks, they help reveal hidden connections that might be missed if the object of interest was not reduced. One of the things I found useful is Moretti's detailing of maps as flexible tools for analysis; there are a lot of methods for mapping, not just geographically, and Moretti notes that maps can be used to "prepare" works for analysis. I found this interesting as he seems to be entering into another form of reduction, similar to networks. His work with geographical mapping reminded me of our earlier lab with Voyant labs and watching us trace the historical data of the issues across each place-name that cropped up. This chapter particularly interested me since I share a love of geography and a fascination with maps... I cannot wait to analyze Dubliners and I wonder if this could fit into my final project somehow.
It was only a matter of time before Moretti's writing style started grating on me.
I'm going to be quite honest when I say that I'm not sure what Moretti's reasons were for writing this chapter. It seems to be some rather basic level stuff: with maps, we can abstract information from literature and visualize it, placing characters, places, etc. in relation to each other in the process. Also, maps are different than graphs, and the shapes of genres change with time. Maybe I'm missing something, but I just really couldn't see what type of new information Moretti was trying to present here.
I am really looking forward to mapping out some stories from Dubliners though.
(I am posting this even though it is late because I had it written and was delayed!)
I feel like I've kind of flip-flopped since last week, and now I'm apparently on the opposite side of the debate! After deciding that Ramsey's methods are at the very least interesting and perhaps quite useful, I've gone somewhere from "so what?" to an area much closer to "why not?" when consider the purpose of DH investigation. Have pinpointed my irritation as somewhere close to "let's not throw the baby out with the bathwater", I'm willing to admit that, just so long as everything that's come before isn't being devalued (as Jockers kind of did), DH explorations of data could probably provide additional illumination to the study of literature. This, of course, doesn't negate what is probably my main issue with technology, which is simply that it eliminates the physical book, and the feel, smell, etc. that go along with this. Are text mining and Moretti's network theory warm and comforting? Not really. Could they be useful in research? Perhaps.
This is probably why I was willing to give Moretti's discussion of network theory a bit of leeway, even considering the apparent issues many of you were finding. Or maybe I just find the blending of Shakespeare with a digital humanities approach rather charming. In any case, I found some of Moretti's thoughts quite useful, and I especially found that, rather unlike Jockers, his analysis seems to toe the line between close and distance reading quite well--examining both the what happens from just what it means that Horatio's absence can throw the play into chaos.
One thing I do wonder, at times, when we begin to read Shakespeare so very closely, or, I suppose, broadly, when "The past becomes past, yes, but it never disappears from our perception of the plot" (4), is if we're not in some way reaching too far, reducing too much. When we've so far surpassed what the average viewer of the play could possibly have been expected to understand, and surely whose comprehension of the play its success depended upon, does our analysis actually amount to much?
I really hate Moretti's tone, in general, but I admit I liked reading this piece. I think his aims don't quite reach his results, but I think through his engagement with network systems and his emphasis on visualization, he effectively communicates the benefits of graphing networks. I particularly enjoyed his engagement with network theory, particularly that "making the past just as visible as the present: that is one major change introduced by the use of networks. Then, they make visible specific “regions” within the plot as a whole: subsystems, that share some significant property." I didn't quite understand the depth of these subsystems until he started discussing Horatio's relationship to structure.
I thought about how this might apply to my own project, how "networks are made of vertices and edges; plot networks, of characters and verbal exchanges. In plays this works well, because words are deeds, deeds are almost always words, and so, basically, a network of speech acts is a network of actions." I'm working with Stoker's correspondences and I am in fact reducing each letter to its sender. I've struggled with the idea of such a reduction. Am I losing too much? Will I be reducing the field too narrowly? Will the Network itself just be enough? But Moretti has helped me through here with his discussion of the Chinese Guanxi where "nothing major happens here: people talk, walk around, play go, gossip... No interaction is crucial in itself. But taken together, they perform an essential reconnaissance function: they make sure that the nodes in this region are still communicating: because, with hundreds of characters, the disaggregation of the network is always a possibility." And that quote, I think, is incredibly important to my work. By the very fact that a network exists, it suggests interaction.
A network can reveal and highlight new patterns. As Moretti states, "you make a network of a play, you stop working on the play proper, and work on a model instead: you reduce the text to characters and interactions, abstract them from everything else, and this process of reduction and abstraction makes the model obviously much less than the original object." I'm moving into creating a model of Stoker's correspondences to visualize the network he was in. I think I was incredibly nervous about the implications of such a reduction and subsequent visualization, but I feel a little more at ease now.
Last week, we could find that Matthew L. Jockers explains about the difference between close reading and distant reading by making comparison of relationship between microeconomics and macroeconomics. According to him, close reading is similar to microeconomics because as microeconomics deals with individual market and individual person’s preference for the commodity, close reading tries to examine the individual text, and infer its significance from the text. On the other hand, in case of distant reading, like macroeconomics which is number driven learning, it deals with the issue of literary text in terms of numerical based data. We can find such comparison from Franco Moretti’s writing, “Literary Lab” and how such idea of close reading and distant reading can operate in Moretti’s thought.
The drawback of distant reading which uses numerical data is, as Moretti confesses, that it cannot deliver the weight of the dialogue. Although his idea of networking between characters can draw the relationship between each character, it cannot show us the aesthetics or social meaning from the line that has been uttered in the characters. For example, this distant reading cannot deliver the vivid imagery of morning given by Horatio’s speech in Elsinore or character’s mental complexity in soliloquies. In addition, according to him, his study can reveal the power distribution in the character, because all of the characters cannot be considered in general same, from his study. However, this might neglect the implication of diverse characters who do not show up often. In case of gravedigger, his significance cannot be found in his graphic line, but as we read his speech, we can find interesting point that this ordinary person can outweigh the main characters in his wit and this shows disruption of hierarchy of knowledge in that era. For this reason, Moretti’s way of reading may miss out some important scene in the play.
Of course, we can find some meaningful thought from his study. His study can easily visualize us how to conceptualize the structure of characters’ relationship. For example, his examination between the people of Hamlet, and the people of Claudius makes us convince how the plot goes on with these characters. His emphasis on Horatio’s role is also interesting, as he explains that his status shows us well the fact how the court is surrounded by characters from different countries and examined by them through Horatio. He also does not dismiss the importance of close reading by focusing on Horatio’s role. From Horatio’s line, he briefly explains that he operates as a person in the court who does not behave like people in the court, but like a figure of the state who does not elaborate his speech. Such reading style of speech can be possible, only because he used close reading.
I am having a hard time mustering up any strong feelings about Moretti's concept of network theory in either a positive or negative manner. On one hand, I think creating a network of the character relationships or whatever elements is a rather nifty notion. I can see it being extremely intersesting and useful, and Moretti's methodology seems to be egalitarian, giving each word the exact same value and weight as every other (3). However, I also share a lot of Steven's frustrations, as none of the conlcusions that Moretti comes to are particularly earthshattering. Yeah, removing some characters from Hamlet raises questions about the dyanmics between characters and why it is that some people are more central than others despite being secondary characters in terms of plot. But, as Steven said so well, this isn't really new?
I think these network maps are super cool, and could definitely prove useful. But I'm just not sold on the necessity of it all.
I kinda liked this reading. I also kinda didn't like this reading. (Good start to this post, right?)
Look, I spent a decent amount of energy last class harping on Jockers, and a decent chunk (can you use chunk in academic writing?) of what I didn't like about Jockers is also what I don't like about Moretti, namely that they both get so far ahead of themselves in their arguments. (Steven pointed out several instances of this in his post.) Can one, as Moretti claims, say "nothing about Shakespeare’s words – but also, in another sense, much more than it..." (4). Can or should a protagonist really be considered "non-anthropomorphic" (11)? What is the benefit of seeing "different genres... as different shapes" (10)? I understand and even appreciate the enthusiasm, but there is a fine line between ambition/excitement and pedantry. Sometimes Moretti and Jockers toe that line; sometimes they cross it (ahem, "layman").
On the other hand, there were various aspects of this essay that I really enjoyed. I think that the network graphs and histograms are enlightening to say the least, and the article offered some much needed inspiration. Imagine creating a network graph of a novel like Catch-22. It would be fasinating to see the shapes that arise from an exercise like that. I also found Moretti's discussions of clustering, symmetry/assymetry, and the cultural import of network graphing through the concept of guanxi to be interesting as well.
However, I still find there to be some traces of interpretation devaluing here that rubs me the wrong way. "This is not a long term solution, of course," Moretti writes, "but these are small networks, where intuition can still play a role; they’re like the childhood of network theory for literature; a brief happiness, before the stern adulthood of statistics" (3). I don't know about you, but I've kinda wanted to avoid that "stern adulthood of statistics" - in fact, that's part of the reason why I chose to study literature in the first place. If one is completely honest about their process, I see no reason why intuition or interpretation shouldn't be a valid and honored part of literary studies.
Moretti's problems are evident by the time he gets to section 4. He is explaining network mapping in the context of Shakespeare (Was this a thing? Do a lot of DH'ers use Shakespeare as some kind of control by which to prove their DH tools?), and highlights a pretty serious concern -- how do you weigh interactions between characters? How do you highlight that weight? What about one way communication? I would assume that you could just use arrows (one-way arrow means a one-way conversation, bold/different colored arrows show different levels of communication, depending on an included table/chart), but apparently that's not a thing. So alright, that's a problem that should be solved. I can see that, 100%.
But then section 2 comes along and Moretti seems to lose the plot. It starts off well enough in the beginning sentences, when he says that using a network map like this can reduce the play into something that is far less than its consituent components. Basically, we shouldn't forget the context in which all of these character interactions take place in. Alright. I'm on board.
But then, he says "a model allows you to see the underlying structures of a complex object. It's like an X-ray: suddenly, you see the region of death of Figure 5, which is otherwise hidden by the very richness of the play" (4). What? Is he saying that it's particularly hard to note the fact that everyone in Hamlet's court is killed at the end of the play? That was noted back in my junior Shakespeare class in 2010. I wrote it off as miswording, perhaps a bit of overzealousness even, but he continues. He says on page 5 that "What is done is never undone; the plot as a system of regions; the hierarchy of centrality that exists among characters; finally - and its the most important thing of all, but also the most difficult - one can intervene on a model; make experiments. Take the protagonist again. For literary critics, this Figure is important because it's a very meaning-ful part of the text; there is always a lot to be said about it; we would never think of discussing Hamlet - without Hamlet. But this is exactly what network theory tempts us to do: take the Hamlet-network, remove Hamlet, to see what happens: And what happens is that the network splits almost in half."
Forgiving the awful writing, this reads a lot like "The only way people can even think of these types of problems, let alone actually see how Hamlet survives without Hamlet as a character at its center, is if they are using network mapping." That is...incorrect at best. Tom Stoppard's Rosencrantz and Guildenstern Are Dead did this rather humorously (but also effectively in a literary context) back in 1966. Or is he, perhaps, saying that these types of questions are network theory questions? I am honestly confused, and his writing does not help matters. This reads a lot like "Only Digital Humanities can solve these problems," when it should say "Digital Humanities can help us visualize these problems." I am willing to admit that I have simply misread him, that I'm missing a fundamental humility or angle of his, but even in the final section he says "So, from its very first section, the essay drifted from quanitification to the qualitative analysis of plot: the advantage of thinking in terms of space rather than time; its segmentation into regions, instead of episodes; the new, non-anthropomorphic idea of the protagonist; or, even, the "undoing" of narrative structures occasioned by the removal of specific vertices in the network" (11). This is why I'm so frustrated. This angle of literary investigation isn't new!
He does take a moment to address my issue with his position. Apparently, he has been asked often "Did I really need [network theory] to speak about Horatio and the State?" to which his answer is "No, I did not need network theory; but I probably needed networks. I had been thinking about Horatio for some time -- but I had never "seen" his position with Hamlet's field of forces until I looked at the network of the play....What I took for network theory were less concepts than visualization: the possibility of extracting characters and interactions from a dramatic structure, and turning them into a set of signs that I could see at a glance, in a two-dimensional space" (11). His answer to the problem is more an excuse for his bias -- he couldn't even think of these questions before he utilized network theory/mapping, so network theory/mapping is presented as the only way to see these questions are even there. People have been talking about character relationships -- and the centrality of protagonists -- for long before Moretti happened upon using network mapping on Hamlet. He should reflect that in his writing, or at least acknowledge that this is not the only way for people to think of the relationships between characters.
I have been struggling to see how people could be so against DH and its uses. After having read this paper, I'm beginning to understand.
I've long entered a debate with close family members over objective/concrete evidence and its often assumed absence in literary criticism. But I was struck by Ramsay's argument, in particular by his definition of data mining. According to Ramsay, who borrows a definition from Witten, "data mining is the extraction of implicit, previously unknown, and potentially useful information from data." This data, processed through algorithms, destills to "regularities" and "patterns" (185). It continues, "Many patterns will be banal and uninteresting. Others will be spurious, contingent on accidental coinicidences in the particular dataset used. And real data is imperfect: some parts are garbled, some missing. Anything that is discovered will be inexact." And this inexactness made me think of Derrida's Archive Fever, notably how the data in an archive is also far from perfect, subject to irregularity, and will never be approached from a neutral, purely rational, or objective standpoint. Interpretation will always be encoded in the human, in the moment, from our place of cultural understanding. It will never occur in a vacuum, always informed by data and previous patterns or discoveries.
Likewise, Ramsay comments on this lack of neutrality as several of my classmates have already discussed. Ramsay states that software cannot even be neutral as "there is no level at which assumption disappears" and instead, argues that this "lack of neutrality" should be "assert[ed]" with "candor, so that the demonstrably non-neutral act of intrepretation can occur" (182). If I see Derrida as looking to encode the cultural archive, I see Ramsay on a smaller scale, attempting to show how Derrida's truisms (or at least what I see as truth out of Archive Fever) also function on a smaller scale in data extraction. Both comment on the subjective human-process of patterning and privileging information.
I have admitted previously to being somewhat of a skeptic (or, at the very least, a doubter) on the topic of “The Digital”, since I am often finding myself rather alarmed by the rapid progression of technology within what would seem just the past 20 years or so, and regarding which I usually wonder if the benefits of convenience necessarily outweigh the potential negative effects that we perhaps do not even fully understand yet. I am thus probably a bit of a—gasp—traditionalist, in this respect. So on some level, the process Stephen Ramsay describes in “In Praise of Pattern” showcases just what I had feared: that much of the purpose of such Digital Humanities projects just might be for the sake of mere curiosity. What, I was poised to ask, might be the cost of such curiosity for its own sake? (I will still ask this, but maybe with less apparent certainty/disdain, as if I already know the answer, because I honestly don’t.) (I do know, too, that I am oversimplyfying the issue and there are certainly reasons besides curiosity alone, and that curiosity may, in fact, be a blessed thing at times.) After all, Ramsay notes himself that stage directions for Shakespeare’s plays are notoriously sparse, and we also have reason to believe that sets often did not change from scene to scene, instead leaving any mention of a change of scene to the imaginations of the play’s viewers. He thus explores the fact that in order to explore the topic he was interested in a certain level of guesswork was necessary: And so, faced with this computationally intractable buffet of confusion, I did what any good humanist scholar would do: I guessed” (182). But for what ultimate purpose is it truly to work at “guessing” in such wise?
These were my thoughts upon initially encountering Ramsay’s article. What I didn’t expect, however, was to be struck by the genuine earnestness of his argument. I found it particularly refreshing what he writes about neutrality, for instance: not that we should strive eternally for neutrality in our work—which I sometimes feel is desired but inevitably impossible—but instead that we should acknowledge openly the inherently subjective nature of something created by another individual person (in this case, software) that even so does not discount the value of one’s work: “This does not imply that the software should be neutral, as many tools and web sites in digital humanities try to be. It cannot be neutral in this regard, since there is no level at which assumption disappears. It must, rather, assert its utter lack of neutrality with candor, so that the demonstrably non-neutral act of interpretation can occur” (182). Additionally, I was very charmed by the sense of discovery Ramsay describes and claims to experience while occupied in The Search (he even calls these moments of unearthing something new “epiphanies”—as strong a description as I ever heard!), and the very human—even endearing—manner in which he describes first feeling rather sheepish about discussing his curiosity with those outside the field of English, before ultimately finding greater worth in his own internal yearning for discovery. The candor with which Ramsay discusses the particular scholar’s joy in the serendipitous encounter was, I found, very compelling in this piece, and, from my perspective, quite a nice argument for the worth of such research methods.