Close and Distant Readings Commensurate with Close and Distant Writings


It occurs to me that one consideration regarding close and distant reading is that there are limitations of “zoom” based on the work itself. The closeness or distance with which a work can be read must be commensurate with the level of minute detail and symbolic density (for close reading) or abstract generalization (for distant reading) infused in the work by the author. That is to say, the reader cannot perform such a close reading that would involve attributing significance to individual words or phonemes or patterns of these constituent pieces if the author did not consider these minutiae when he or she composed the work. It would be like zooming in to an image to the point where the whole screen is a blur of individual pixels: while some inspired viewer might find some highly subjective interpretation of the pixellated blob, we can not say that that blob is representative of the larger, zoomed-out image in which the resolution befits content. Close readings thus work well with works whose authors, through attention to detail, have infused each component of that work, be it a phoneme or a brush stroke, with intention.

The same principle holds true for trying to scale your perspective of a work above and beyond what the author designed. In the same way that poets often design their works to be scoured microscopically, novelists often generate their works with grander messages, meanings, or morals intended to emerge out of and beyond the literal meaning of the individual words, sentences, paragraphs, and chapters. These works, in this sense, lend themselves to being read in ways greater than the sum of their parts. It is this function, to keep with the example of the novel, that enables satire and other forms of socio-political commentary. While some media and works, by design, lend themselves to very distant readings, as with microscopic readings, the cultural-consumer can be telescoping in their reading of a text. Readers can slather grander meanings onto texts never intended by the author, and can use these readings of texts for their own philosophical or political agendas.

As with too close a reading of a text, too distant a reading can allow the reader to infuse/impose meaning into a text that, though unintended by the author, can be actually meaningful to the reader. When this is done frivolously or in a way that doesn’t impact others, these hyper-readings can be beneficial, allowing the reading to subjectively experience greater meaning than otherwise might have been possible from a given text.

Close and Distant Reading: A Look Into the Future

One of the core values of the humanities, especially the part that focuses on literature is “close reading”.  Close reading is a careful interpretation of text with emphasis on particular details such as individual words, syntax, and the order in which sentences and ideas unfold as they are read. This practice helps the reader get a better understanding of the meaning of a text. But how should one analyze all the writings of an author or all the books written in a specific time period? There is not enough time to carefully read that much text. Digital humanist have made efforts to enable the understanding of literature not by studying particular texts, but by aggregating and analyzing massive amounts of textual data. Search engines and various digital tools give us the ability to learn more about literature by letting the computer process the texts. This practice is called “distant reading” and it is sometimes criticized for distancing the human from the reading. Distant reading can’t interpret the semantics of a verse in a sonnet like a human would, but it can measure the occurrence of certain words or phrases in over a million sonnets in an instant.  Distant reading is not completely distancing the human because it still needs humans to make sense of the computer’s work. The second criticism to distant reading is its reliability and bias. Context is everything and current text mining tools still need to work on that aspect of data analysis. According to Ted Underwood, text mining is a “focused form that only shows you what you already know to expect.” I believe that we need distant reading because we can’t read all the books in the world but we also need to develop better technologies that will allow us to make distant reading more intuitive and human-like. We need artificial intelligence (AI).

AI technology would merge close and distant reading. Imagine we teach a machine to think like a human. It would read a sonet and understand its meaning, metaphors, themes. It would appreciate its beauty and language. Most importantly it would do all of this in a split second and be able to actually “close read” all the sonnets in human history before a human would finish reading even one. The distant reading of the future would not distance the human from the reading. It would put the human – its creator – at the very center of it. This would only be possible if the AI is given a human-like consciousness, where it’s self-improving and developing, learning (mimicking) from others – like a human but much faster with no physical constraints of a fragile human being. Of course, having real AI technology raises a whole new set of issues (see The Terminator). Stephen Hawking said: “The development of full artificial intelligence could spell the end of the human race”. We can, however, use the concept of AI to improve our current distant reading technology and make distant reading less distant but equally or even more efficient. This is already happening. Last year inventor and futurist Ray Kurzweil joined Google to breathe intelligence into Google Search. He is leading a project aimed at creating software capable of understanding text as well as humans can. More information about this project can be found on the MIT Technology Review or on The Guardian.

Close and Distant

In my attempts to assess the benefits/drawbacks of close and distant reading I’ve found that I tend to limit my view of each method by reducing them to two examples, each from an individual’s perspective while employing the methods during research. The first, the example of close reading, involves an individual analyzing a small selection of work, doing so with care in repeated reading of the entirety of a text, highlighting as he/she reads, ultimately interpreting the work using a personal filter or specific lens. This form of reading, the reading most of us are familiar with, is reading for the purpose of understanding the form (structure) and function (intended effect) of the text, with special attention being paid to context and the “meaning” or impact of the works content. As individuals we bring our life-experience to the table when we attempt to analyze, in this way we are better at understanding the humanity in a body of work. However, though we excel (at times) at relating to the text we also excel at ignoring, rejecting, or overlooking potentially important information. Because we are limited both by how much we can read and our comprehension/retention of that which has been read, we aren’t great at recognizing patterns present within a corpus. Our focus on a specific work or area limits us to a “canonical fraction” of the larger body of work it belongs to.

The second example, the example of distant reading, sees an individual involved in research requiring the ability to search within a corpus to find patterns and thematic or canonical connections. You’ll notice my description of this example is quite short, the length reflects my understanding of it. I’ve never really researched using distant reading methods, and until this course hadn’t really been exposed to it as a concept, and even after reading the assigned material find that it is difficult for me to wrap my head around. I recognize the potential in distant reading, it allows us to do what we never could, we can ask questions before unanswerable and establish more quickly connections and patterns within a vast collection of work. We can even find connections between corpora, something of tremendous impact for historians as it can take what might have been considered unrelated works and find links that perhaps we never could. I fear that this method, however, is detached and cursory. Where close reading involves a relationship between the humanity of the work and the reader, it strikes me that the same connection isn’t present in distant reading, because it isn’t reading in the traditional sense.

Using both close and distant reading while researching, I think, is an exciting prospect. The inherent negative aspects of the respective methods seem to balance out when the methods combine.

Of course, there is still the issue of the politics of curation, but I’ll save that topic for in-class discussion.


While searching for information about distant reading I came across this article, which I think you all may find interesting:





In Defense of Digital Moderation

Technology is ever-increasing the ways in which we can view and research cultural artifacts, but is this a good thing? Before the widespread proliferation of the printing word, scholars could study near everything with relative ease. Books were expensive, and there wasn’t many of them. However, after book production exploded, it was far easier for scholars to focus their study towards a highly curated pool of ‘acceptable’ materials. It was hard, perhaps even impossible, to study everything, so, instead, expert filters were established in order to cull what was worthy from that which was not, but these filters were little more than a stopgap. As is now incredibly evident, filters are quickly failing. There’s too much interesting of everything to study, and without filters, we’re forced to study everything so we can filter everything ourselves. As technology has advanced, so have tools which enable a more distant, mass reading of material, but, as mentioned before, is this a good thing?

There is definitely merit in distant reading, don’t get me wrong. I think new techniques broaden perspectives and bring usually-needed variation to the echo-chambers of academic disciplines. In fact, I’d encourage anything which incites such broadening. I do, however, have a fear. We’re swimming in a sea of studiable material, and not only is it daunting, it also is stressful. Academics and scholars are those who should know things others do not, and there is pressure, be it professionally, socially, or psychologically, to know one’s devoted field of study completely. Often times one’s career hinges on the fact that one knows more, or at least as much as, everyone else in that field, and as the surfeit of useful material quickly rises, so lessens the ability of scholars to maintain their study.

It’s unsurprising that digital tools and techniques have been developed to aid in this task. If I can’t read all the books, maybe a computer can help paraphrase, and, to a limited extent, this has worked. With the mass digitization efforts such as those of Google, and a litany of others that I’m most assuredly unaware of, the problem is solved, right? Computers have saved us. The sea has been dammed. Paradigms have shifted, and this novel new world, as shiny as it might seem, makes me iffy.

Helpful as digital technologies are, there’s real risk that, if over relied upon, they’ll  encourage people to minimize close reading, as those of the past were forced to do. As already stated, the broadening of perspectives resultant from a multitude of methodologies is the true value in developing new forms of learning, seeing, and researching, but what’s the point if we simply trade one form for another? Every form has inadequacies, and it behooves us to not only utilize, but also encourage, nurture, and support all methodologies in order to best understand our world, ourselves, and our past. Developments in scholarship need to eliminate inadequacies, not supplant one for another.

Personally, I think digital tools need to be used with some conscious mediation. Just because the wheel was developed millennia ago doesn’t mean people stopped, or valued, walking. In the same way, scholars needs to value both new and old forms of scholarship. Both have their place, and both should be practiced well into the future.