STANLEY FISH: There are many ways to think about the digital humanities. By the way, how many of you know something about the digital humanities? OK. That looks to be about a tentative-- people raising their hands as if they were little mice. How many of you know nothing about the digital humanities? Yeah, right. Good, OK.
One of the ways to think about the digital humanities is as a response to the endless announcement in more books than anyone could possibly read, of the crisis of the humanities. Crisis of the humanities has been going on for thousands of years, as far as I can tell. And both the laments, the diagnoses, and the recommendations tend to be dispiriting alike.
Always there's a savior in the form either of a person or of an idea or of an agenda or, given the kinds of academic creatures we are, of a method. And it's the salvation by method that is supposedly offered to us by the digital humanities. Now the digital humanities is a prodigious enterprise. There are many people, many younger scholars especially, who are invested in it. And it keeps issuing manifestos.
So what we have, or what I have here, is the Digital Humanities Manifesto 2.0, which I thought was the third announcement of Hillary Clinton's candidacy, but in fact, is a manifesto where the digital humanists pose themselves against people they call the traditionalists. There are no-- no, that's not quite-- there is only one name mentioned in the Digital Humanities Manifesto 2.0, and I'll read the relevant short paragraph.
What is the traditionalist response? And then there's a list. And it's a list of responses that you're supposed to hear as obviously inadequate and predictable. Passively except the tools handed down from Olympus. Weave lamentations on the decline of the West-- something to which we, as humanities academics, are quite prone. Keep on doing what we have always done until the moment of extinction. That's, in fact, my strategy. And finally, celebrate extinction or usefulness from a well padded, tenured chair, and 401k, a la Stanley Fish.
So I'm very pleased, actually, to be the only name mentioned, and mentioned in an appropriately hostile manner, just as Jonathan predicted. Here we have a whole new race of academics who don't agree with me from the beginning. What could be better? But it's still a task that we have, at least that I have, to explain to some of you for the first time-- who will be hearing this for the first time-- what the digital humanities is or are. And I myself vacillate between those grammatical alternatives.
And to perhaps take issue with some of you in today's audience who may be sympathetic to the digital humanities, or may be, God help us, digital humanists yourselves. I should say in advance that there are a couple of projects associated with the digital humanities with which I have no quarrels. One is the project of attribution, the idea that the digital humanities can give us another technique which will better enable us to identify disputed texts. That is, to identify the author of disputed texts.
And there are, of course, a number of disputed texts in almost any cannon that you can imagine. When I first got into the Renaissance poetry game, there were a number of poems that are disputed as between the authorship of John Donne and Ben Johnson, although I myself cannot think of two sensibilities more different. But nevertheless, there is that dispute. And I think that aspect of the digital humanities is indeed something we should pay attention to.
And then there's that part of the digital humanities that provides very extensive and helpful, pedagogical helpful websites, allowing you to surround a text or an object or a painting with all kinds of historical and interpretive materials, and introduce your students to those materials. And I think that's also a worthy addition to the tools that we have at our disposal. But those are not the versions of digital humanities that I'll be talking to you today about.
And I wanted to, I'm going to frame my remarks on the digital humanities by speaking about legal theory, something that I've been involved in now for many years. And especially about legal interpretation, and the kinds of competing approaches to legal interpretation that now vie for supremacy in the field. Because I see a relationship, at least an organizational relationship, between the problems raised by the competing schools of legal interpretation and the problems raised by the claims and promises of the digital humanities.
The relationship between interpretation as a topic and the digital humanities is a complicated one. Some digital humanists believe that the traditional task of interpretation can be better performed with the aid of powerful computer programs. Others believe that the traditional task of interpretation is tied to an epistemology and to a politics that the digital humanities is pledged to undo. So there are two very different strains of the digital humanities.
But before exploring these two positions, and they are not the only ones in the field, it will be good if we can specify exactly what interpretation is, which is of course a large claim of mine, to be able to specify to you today, in a paragraph or two, what interpretation is, since people have been arguing about that forever. Debates about interpretation, as you know, are found in every discipline. But the debate is most sharply focused in the discipline of law, in part because the object of legal interpretation is often the Constitution, or a landmark statute.
And most people agree that getting the Constitution right is an important project, as opposed, let's say, to my project of 45 years of trying to get Paradise Lost right. And I'm still working at it. There are, at present, three accounts of legal interpretation vying for supremacy in the field. That is, three answers to the question, how do we go about specifying what a text means. And those three answers are related, I believe, in interesting ways to the digital humanities project.
The first and most popular answer to the question, how do we go about specifying what a text means, is that you figure out what a text means by attending to it's language, to the meaning its words had at the moment of their production and/or reception, and to the syntax within which those words are embedded. The foremost proponent of this view of legal interpretation is Justice Antonin Scalia. And Scalia is famous for insisting that interpretation should be tied to the stipulation of what he calls original meaning.
And also famous for arguing that original meaning is to be found, not in the intentions of an author or in the record of constitutional or legislative debate or in any other location external to the text. Original meaning is to be found, Scalia tells us, in the text itself-- at least if it has been well crafted. Now the name of this view is textualism, not surprisingly. And it has a strong commonsense appeal. For isn't the text an object, like a piece of sculpture or painting? And if we look hard enough at it, shouldn't we be able to determine what it means, without going elsewhere into all kinds of foreign precincts?
Textualism, at least as a general thesis, is congenial to the version of the digital humanities that claims to be able to reveal more about texts than could possibly be detected by mere human readers like you and me. The idea, and here is where I begin to try to explain with the digital humanities is or are, the idea is that there are patterns of repetition, of frequency, of contrast, of anything. There are patterns so deeply embedded in text, or spread across so wide a space of text, that no human reader could discern them or have world enough and time to search them out.
The problem then, from the vantage point of the digital humanist, is double-- an insufficiently microscopic vision on the part of the merely human reader, and a life span that severely limits what he or she could possibly read. Enter the computer program, and we are told both problems are solved. For the computer can uncover what is not visible to the naked readerly eye, and it can do it in seconds or milliseconds. Moreover, since vast bodies of text can be digitized and searched, the results, it is claimed, can serve as a much firmer basis for reaching conclusions than the basis provided by the small number of texts a reader could possibly process in his or her threescore and 10.
So for example, the historian Dan Cohen asks, isn't it the case that our conventional scholarship might be anecdotally correct but comprehensively wrong? That is, we might have somehow got it right about Dickens or Tolstoy, but it will just have been a lucky guess. And he says, even if we base our conclusions on 100 books or even 1,000 books, probably beyond the capacity of most scholars to absorb and keep in mind, shouldn't we work not with a sample of text-- and I'm still quoting Cohen-- or a few canonical texts, but with all of it?
Now in response to this question, Cohen and a colleague started working with the database, digitalized, the database of the 1,681,161-- you got that? Write that down-- books published in England in the 19th century. And they have been able to chart the frequency with which certain words appear in the title of those books in different decades. It turns out that books with the word science in the title increase as the century unfolds. I don't know about you, but I'm not surprised.
The word virtue, however, declines in popularity as the decades pass. And I don't know what to make of that. While the word Christian peeks at mid-century, and love troughs, goes way down, at the same time. And there's a lot more of that. OK. Now when I look at this and similar studies, my mind flashes back 40 years to the 1970s, and my attempt, only partially successful, alas, to kill an entire sub discipline. I tried hard, you know, but they kept on getting up, like the slashers in a bad horror film.
The discipline I tried to kill-- and I did wound it, because they did respond to me in wounded tones-- the discipline was and is called stylistics, and it had, and still has I suppose, many versions. The ones I used to teach were computational stylistics, neo-Firthian stylistics-- stylistics that were derived from a famous British linguist-- transformational grammar stylistics-- stylistics based in Chomskyan notions-- speech act stylistics, structural stylistics, pragmatic stylistics, psychoanalytic stylistics, and others.
All of these were united by a single assumption, and this is the assumption that I think is faulty. And since all versions of stylistics and the digital humanities begin with it, the entire project is a nonstarter. And that is the assumption that it is possible to go from data, however collected, to interpretation. That's the assumption. That you can go from data, however collected, to interpretation. Which then gives rise to the secondary and almost inevitable assumption, that the more data, the better based your interpretation will be.
In short, the assumption that formal features-- anything from sentence length to image clusters to key words to definite articles to passive constructions, you name it-- that assumption that formal features carry meaning. Now in my critique of the stylistic project, I took unfair pot shots, and now do it again, at a scholar who after studying the prose styles of Addison, Samuel Johnson, the historian Gibbon, and Macaulay, determined-- and Swift, and Jonathan Swift-- determined that Jonathan Swift's prose style is distinguished by the presence of long series of phrases in apposition. This is, of course, true.
The scholar then delivered his conclusion, which I read. Swift's use of series argues a fertile and well stocked mind. You got that? When I first read this, I stood up and exclaimed, thank you Lord, for having delivered my enemy unto my hands. I observed in print that the movement, in this instance, from formal pattern to interpretation, that is, to an interpretation of Swift's mind and character on the basis of his use of series, is entirely arbitrary, not to say banal.
One might as well have said, and with the same lack of justification, that Swift's use of series argues and anal retentive personality. Actually, that's probably true. Or that Swift's use of series argues a nominalist rather than a realist philosophy. Or that Swift's use of series argues a reluctance to proceed to closure. Or a thousand other things. Once you have the data and just the data, you can't go anywhere. Or rather, and it is the same thing, you can go anywhere at all.
Now writing in a book called The Companion to the Digital Humanities, digital humanist Hugh Craig acknowledges the force of my criticism in the 1970s, but asserts that the more sophisticated techniques now available make possible a new stylistics with what he calls another motivation. And he defines it, the motivation, quote, to uncover patterns of language use, which because of their background quality-- that is, how deeply embedded they are-- or their emergence on a super humanly wide scale, would otherwise not be noticed, unquote.
But if the problem with the old stylistics was that you could not generalize, except illegitimately, from the data, the problem with this new up-to-date stylistics is that it is by no means clear why you should be interested in the data it uncovers at all. Maybe the patterns that have not been noticed before, patterns like the frequency with which particular words appear in the titles of 19th century books through the decades, should have remained unnoticed, because they are nothing more than the artifacts of a machine. Maybe the whole project is makework, with emphasis on the make, rather than the work.
Now it is in response to objections like mine that digital humanist Tom Scheinfeldt makes an argument also made by the stylisticians in the 1970s, that it is wrong to expect too much from a project that is just underway. This is a nice argument, because it indicates, because it allows you to say that failure is a reason to believe that success is around the corner. So Scheinfeldt says, it takes time for new techniques to payoff, and to demand results too soon is to fail to give this new enterprise the breathing room it needs.
Does digital humanities have to produce new arguments now? Does it have to answer all the questions right away? Unquote. After all, he goes on to point out, the experiments of scientists often are just that, experiments-- exercises performed without a clear sense of where they are going, but exercises that, in the long run, sometimes take us to places we cannot now even imagine. He concludes, the digital humanities may not be answering lots of questions currently, but let's just keep at it and see what turns up.
And the most sophisticated theorist of the digital humanities, a man named Stephen Ramsay, in a book called Reading Machines-- and I recommend it to you-- he calls this strategy, that is, of let's run some numbers and run some more numbers and run some more numbers and see what turns up. He calls this strategy the hermeneutics of just screwing around. And I really like that, the hermeneutics of just screwing around.
But the prospect of anything except the belabored or the obvious turning up depends ultimately on a faith I don't have, and that is the faith that data mining will lead to something more than the proliferation of its own operations. Instead of Edmund Spencer's endless work, we will have endless noticings, endless noticings-- look at this, look at that. But digital humanists will reply that endless noticings are more than enough, and are finally more valuable than the narrow interpretive results people like me are hung up on.
Those who say that will have some affinities with the second school of legal interpretation, which calls itself the living Constitution school. That school rejects textualism and Scalia's notion of original meaning, because they bind us, it is said, to the dead hand of the past, and render us incapable of adjusting flexibly to the ever changing dynamics of a fluid, mobile, transformative, and transforming world. Living Constitution proponents in the law also argue that the broad and general cast of the framers language, phrases like cruel and unusual punishment, indicate that their intention was to produce a document that would grow and evolve and be adapted to the need of successive generations.
To the members of the living Constitutional school of legal interpretation, the words of the Constitution are not constraints that limit and cabin interpretation. They are rather prompts to the performance of a creative activity that begins with some problem or urgency in the present, and then stretches and bends the Constitution's language until it can be made responsive to those problems and urgencies. It's a question of what end of the handle you grasp. That is, do you want to past to in some way control and constrain the present, or do you want to begin with the present and to see if you can massage the past until it fits the present's needs?
The best account of the living Constitution theory of interpretation was provided by Richard Rorty, when he said, almost in an aside-- as Rorty always spoke in an aside, even when he wasn't speaking in an aside he was speaking in an aside-- Rorty said that in the act of interpretation, and I'm quoting, what you do is beat the text into the shape that best serves your present purposes. And that for him was the definition of interpretation, and it's certainly, I think, an appropriate definition of the living Constitution school of legal interpretation.
In the digital humanities, the practice Rorty describes-- that is, of working the text over until you can turn it into a shape that pleases you, for whatever reason-- in the digital humanities, this practice is termed deformance, D-E-F-O-R-M-A-N-C-E, or tampering. Rather than view the text as a fixed, stable entity whose integrity must be preserved, which is of course what originalists and textualists do, think of the text as an assembly that can be reassembled by making what the post-structuralists used to call a new cut. That is, a new angle from the perspective of which the text's components are rearranged into new patterns. And then you can do it again and again and again.
In his book Reading Machines, Ramsay points out, and I think this is a powerful point, Ramsay points out that deformance is in fact what we've always been doing when we offer a reading of a text. We notice something and then we set to work. And he describes the kind of work we usually do. We read out of order, we translate, we paraphrase, we look at only certain words or certain constellations of words, and at the end, the text hasn't changed it's graphic content, but the text quite literally assumes a different organization from what it had before. End of my quote from Ramsay. And once that different organization is in place, do it again.
We run procedures, says Ramsay, that have the effect of creating alternative texts that form the basis of still further elucidations and still other alternative texts. The point is not to pin meaning down, which is usually the point of the traditional act of interpretation. The point is not to pin meaning down, but to release it. And here is Ramsay's master statement, our fear of breaking with the text must give way to a renewed faith in the capacity of subjective engagement for liberating the potentialities of meaning.
So rather than being just a technical aid to interpretation, in a statement like Ramsay's-- and his is a modest version-- in a statement like Ramsay's, the digital humanities become liberating. And you might ask, well, what are they liberating us from? And it will turn out to be everything. Now the critic who so imagines himself, that his, imagines the artifact of human culture as radically transformed, recorded, disassembled, and reassembled-- another quote from Ramsay-- the critic who so imagines himself is, of course, himself or herself an artifact of human culture.
His or her apparent fixity is no more real than the fixity of the text, subjected to serial assembly. He or she does not preside from a position of mastery over the meanings whose potentialities are released. He or she is not their author. He is simply one of a chain of endless re significations. Rather than owners of these significations, we are members, says Kathleen Fitzpatrick, another leader in the field, quote, of a fertile community composed of multiple intelligences, each of which is always working in relation with others.
The result, says Fitzpatrick, is decentered texts, decentered authors, and decentered readers. And you may like that. You may think that's a good thing, in which case, you're going to be on the side of digital humanities. Even for me to read decentered texts, decentered authors, and decentered readers is to send a chill up and down my spine. And everything therefore, in the digital humanities vision, at least this version of the vision, is on the wing. It is a vision of principled instability, principled instability. And in the words of still another digital humanist, Franco Berardi, quote, no object, no existent, and no person. Only aggregates, temporary atomic compositional figures, that the human eye perceives as stable for a moment, but that are indeed mutational, transient, frayed, and indefinable.
Now in an earlier essay, I describe this vision of the digital humanities, as the agency of a powerful, liberating force, as theological. There's a religious, there's a religion here. And let me explain what I mean by theological with the help of three 17th Century poems. The first is by John Milton, and it's called At a Solemn Music. And in that poem, Milton re-imagines the moment when God's harmony, the music made in concert by all parts of the universe, the music of the spheres, is disrupted. Everything was fine, Milton says, until-- and now I quote-- disproportioned sin, jarred against nature's chime, and with harsh din, broke the fair music that all creatures made, unquote.
Now in the phrase, disproportioned sin, disproportion is not the adjectival modifier of sin, but the definition of sin. It is sin, Milton tells us, to stand alone, to not be in proportion with others, to stand out, to claim to be a single voice in possession of one's own thoughts and words. That is sin. It is, on the other hand, virtue to be heard only as a member of an ensemble, to exchange the false joy of being a solitary singer for the true joy of singing along with all creatures in a corporate existence that knows no mine and thine.
Milton's hope at the end of the poem is for a day when nature's chime is again sounded by a universal choir in which each member finds his or her identity in the loss of identity. And I quote the magnificent four lines that end this poem, oh may we soon again renew that song and keep in tune with heaven, til God ere long to his celestial consort us unite to live with him and sing in endless morns of light. Now that's good.
The difficulty for mortals of experiencing this gain in loss, that is, of losing individuality, but as more than a recompense, being in tune with the music made by all creatures, the difficulty for mortals of experiencing this gain and loss is highlighted by the Anglican poet George Herbert, in three short remarkable lines from his poem, The Flower. Here are the lines. We say amiss, this or that is, thy word is all if we could spell. And I'll do them again. We say amiss, this or that is, thy word is all if we could spell.
We say amiss, that is, by saying at all, by predicating, by indicating here's a this, and there's a that. We say amiss by picking things out from one another. The better course, if we could only follow it, would be to see the oneness in difference. Thy word is all. Instead, we as fallen creatures, keep on trying to figure things out, to spell, but we will only learn how truly to spell when we stop spelling, stop trying to keep separate entities that exist, and are known not frontally and directly, but diacritically. That, of course, is easier said than done.
Discursive language is by definition linear, confining, unidirectional, dedicated to delivering just one insight at a time. It is resolutely, that is, discursive language of the kind that I am attempting to speak now, is resolutely thin. And it draws from the poet John Donne this exasperated exclamation, in a poem called The Anagram. Donne says, quote, if we can only put the letters but one way, in the lean dearth of language, what could we say? Not much.
And that is why Donne, like other poets, sets himself the task of bursting the bonds of predication by means of puns, double entendres, complicating allusions, bivalent syntax, embedded acrostics, kaleidoscopic images-- anything that disrupts and arrests the drive to closure. Anything that opens up the expansive vistas that language, when conventionally deployed, always narrow. Now the reason that language is the way it is, that is, lean and dedicated to a dearth of meaning, rather than to a kaleidoscopic multiplicity of meaning, the reason language is that way is because we are that way.
That is, we are limited creatures whose movements and thought and action are always toward a point in the future. Meaning, meaning is always elsewhere and never apprehended in its fullness. We are never where we want to be, fully realized and connected to everything. Our activities are always bringing us to somewhere, but not to every where. We are, as creatures, hopelessly narrative, capable of following and enacting a story line from its beginning to its end, but incapable of following all living and infinite number of story lines simultaneously and without rank or ordering.
Attempts to bypass the limitation of finitude, of mortality, of temporality, of discursiveness, are as old as human kind itself, and included in a list that is only partial, hallucinogenic drugs, fasting, meditation, and forms of art that invite us to climb the mystical ladder leading to a union with the divine. Now in some of its more apocalyptic moments, the digital humanities ask to join the list that I have just read. Here is Carla Hesse, another digital humanist, in a statement that holds out the same promise, whose realization in a far distant future Milton celebrates at the end of At a Solemn Music.
Carla Hesse writes, what appears to be emerging from the digital revolution is the possibility of a new mode of temporality, one in which public exchange through the written word can occur without deferral. That is, without linearity, in a continuously immediate present, a world in which we are all, through electronic writing, continually present to one another. So that's the promise, not just a new tool-- that is, we have machines that can allow you to see more data and perhaps draw from the data that you have amassed better conclusions-- not just a new tool, but a new mode of being, is what the digital humanities offer us.
Not just a refurbishing of the humanities project, but a total transformation of it, which also orders a transformation of its practitioners. What is at stake, declares David Perry, another digital humanist, is not the object of study or even epistemology, but rather ontology, the big O. He concludes, the digital changes what it means to be human. And the humanist Rafael Alvarado tells us, now has, quote, the opportunity to immerse yourself in the transductive plasma of interpretation, where ideas and their expressive vehicles can be mapped and remapped in a variety of forms and frameworks, a giddy play of practice.
I hate every word in that sentence. That is, especially the part that begins, transductive plasma of interpretation, and ends with the giddy play of practice. OK. Now I have to turn the page so I remember why is it that I hate that. Ah yes, here it goes. Phrases like transductive plasma of interpretation and the giddy play of practice will make them members of the third school of legal interpretation, the school to which I belong, very nervous. That school is called intentionalism.
And it is fair to say that intentionalism, as a model for the interpretive effort, or as an account of it, stands against everything that the more visionary version of the digital humanities is for. Intentionalism is against radical openness, decenteredness, instability, dynamism, transformation, fecundity, and multi vocality. When Kathleen Fitzpatrick describes our texts as growing, even after they have seen the light of day, and Mark Poster calls for rearticulation of the author from the center of the text to its margins, an intentionalist like me will think, what are they talking about? Texts don't grow. They're not vegetables.
And if authors are not at the center of texts, anchoring and giving them point, what is at the center of the text? The answer, offered by the digital humanists and offered without embarrassment, seems to be nothing, nothing whatsoever. Intentionalism is also against that version of the digital humanities which is an extension of Scalia's textualism. For intentionalism denies textualism's basic tenet, that texts, in and of themselves, bear meanings. Or as linguist Martin Joos put it many years ago, text signals its own structure.
That is the most concise definition of textualism that I have ever seen. Text signals its own structure. But intentionalists assert that there is no text apart from or prior to the assignment of intention. Absent that assignment of intention, which may be a positive act or a default assumption, there are just black or white marks, swirling smoke signals, scribbles with crayons, the play of light and shadow, or some other merely physical phenomenon. To see my point, imagine walking along a street and seeing on the sidewalk some marks that apparently spell out H-E-L-P.
You wonder as you come upon this configuration, what does it mean. You know, is it a genuine cry for help, is it a tribute to a Beatles song, is it an outrage against the ravages of neoliberalism? What is it? OK. But then let us suppose that you find out that the pattern you're looking at was produced by the random droppings of fluids from the gutters of a building. Do you continue to wonder what it means? Do you continue to interpret it? The answer is no, because unless you believe that buildings and gutters are capable of communicating-- and there are some horror movies in which that is the case-- but unless you believe that, you will no longer think that somebody was sending you a message.
The marks that you see don't form a word. They merely resemble one. They are physical accidents and don't mean anything, because no purpose of agent meant, that is, designed or intended them. Now you can see where this is going. For an intentionalist, the fact that data mining can uncover hidden patterns undetectable by the mere human reader is cause not for celebration, but for suspicion. A pattern that is subterranean is unlikely to be a pattern that was put there by an intentional agent. And if it wasn't put there by an intentional agent, it cannot have meaning.
And if it cannot have meaning, there's no reason to be interested in it. It's finally no different from random droppings that just happen to coalesce into a shape. I don't know why it's so hard to turn the pages here. For an intentionalist then, interpretation is reserved for one off acts performed at a particular time in response to the effort of some purposive being to convey a message. Perhaps the simplest case is the grocery list given to you by your spouse or partner. And this, to my experience, is one of the most terrifying experiences.
It is one of the most terrifying moments in life, when your partner gives you a list and says, go to the store and get these. You know that disaster of some kind will follow. Let's say now, your job is to figure out what he or she means. And your ability to do the job, one of the hardest I know, depends on your ability to keep in mind the kind of person he or she is. If the list says swordfish, and there's no swordfish to be had, will tuna do? If it says orange juice, can you safely get orange juice with pulp?
Now intentionalists will think that with respect to the question of interpretation, there is no difference between the grocery list and Paradise Lost. The task is the same, to figure out what someone meant by these words, although you might decide that one instance of the task is more difficult than the other. But just which one presents the more difficult problem, Paradise Lost or the grocery list, will be an open question for those who have been married, as I have been for a long time.
But of course, there are those who would balk at treating a grocery list and Paradise Lost as the same kind of objects subject to the same kind of interpretive inquiry, into a single intended meaning. There is a long tradition, as you all know, in which literature and especially poetry, is defined by its capacity to bear many meanings, even an infinite number of meanings. It is literature if it is polysemous and resists being reduced to a single intended message, resists, that is, being interpreted. Interpretation, in this view, is a violation of what art is.
It is, in the famous words of Susan Sontag, reactionary, impertinent, cowardly, and stifling, unquote. To interpret, Sontag declares, is to impoverish, to deplete the world in order to set up some shadow world of meanings. Away with shadow worlds of meaning, she cries. Quote, let us experience more immediately what we have. We must learn to hear more, to feel more-- and I'm still quoting her-- in place of a hermeneutics, any theory of interpretation, we need an erotics of art. That is, an understanding of art that stresses play, proliferation, and fecundity, not as a means preliminary to a discursive end, but as that end itself.
Now in its most lyrical moments, some of which I have provided you, that is what the digital humanities calls us to, an erotics of art. Indeed, it might be said that the digital humanities, rather than offering itself as a handmade of art, aspires itself to be an art form, and one more inclusive and more universal than any other. The obstacle to this lofty ambition is to be found, digital humanities tell us, in the gate keeping mechanisms maintained by the Academy and the publishing industry, maintained and designed to limit both the number of authors and the kind of work they do.
You will not be surprised that digital humanists are sworn enemies of copyright, are sworn enemies of any gate keeping procedures, sworn enemies of editors, sworn enemies of selection processes that supposedly winnow out the wheat from the chaff. Essays by digital humanists are full of contemptuous references to quote, pencil humanists-- I like that one, pencil humanists-- knee jerk humanists, to entrenched senior faculty, to artificial constraints that serve professional rather than human ends, and to the poverty of a discursive field that is hung up on print and linearity, and values product over process.
Everything about traditional humanities practices conspires to stifle the natural and exuberant creativity of the signifying creatures we are, or so say the digital humanists. One traditional practice that has been the target of criticism is the writing of academic papers, either by ourselves as full fledged academics, or by our students. My former colleague, Kathy Davidson, noticed that the same students who wrote the most turgid and unreadable papers were elegant bloggers.
And she asked herself, what if bad writing is a product of the form of writing we require in school, the term paper? What if bad writing is not necessarily intrinsic to the student's natural writing style or thought process? What if research paper is a category that invites, and even requires, linguistic and syntactic gobbledygook, unquote? Another digital humanist, Mark Sample, is even more severe. Quote, the student essay is a twitch in a void that means nothing to no one. Nowhere but in school would we ask someone to write something that nobody will ever read.
Now I might put Sample's point in a different, slightly more generous way. A student essay, or the essays that we write as faculty members, a student essay is a specialized piece of work written in conformity with specialized standards and objectives. It is not written for everyone, but only for that small population that can participate either as a producer or a consumer in a disciplinary transaction. Writing a paper is a learned skill, and not everyone need learn it. But those who want to learn it should be encouraged, not mocked. It is an artificial skill, the very opposite of natural, as Davidson charges, but so is blog writing and any other form of communication made possible by the emergence of digital tools.
Digital humanists are fond of citing McLuhan, Marshall McLuhan, in passing, but they miss his point. There is no message without a medium, and the medium does not merely convey, but shape and limit the message. The term paper is a medium for a message that is foreordained, scripted, and constrained. But Twitter is a medium equally, if not more, foreordained, scripted, and constrained. One could say, if one wanted to say things like this, that a tweet is today's sonnet.
If Davidson's students write gobbledygook in their term papers, the fault lies not with the medium, but with the failure at once to respect it and to innovate within its generative limitations. When those same students turn to blog writing, they are not displaying what Davidson calls a natural style as opposed to the cramped style of an academic exercise. They have exchanged one learned and confining practice for another learned and confining practice.
Mark Sample declares, quote, I don't believe that my mission as a professor is to turn my students into miniature versions of myself, unquote. Well, if that's not his mission, to impart and expand the skills that earned him his degree, then maybe he's in the wrong line of work. Because that's what we all do. And that's what we should be doing, trying to turn our students into miniature versions of ourselves. Sample inveighs against the academic fixation on words, and he says, why not focus on images, sounds, why not objects? And the answer is, no reason.
If he develops, as indeed he tells us that he has, alternative forms of writing in which his students become quote, aspiring Rauschenbergs assembling mixed media combines, more power to him. He should know, though, that he is still producing, or attempting to produce, miniature versions of himself. He should know that-- I'm sorry. I'm sorry. No one naturally goes around assembling mixed media combines. Everyone who preaches digital liberation should be required to read Wordsworth's sonnet, Nun's fret not at their narrow convents room, once a day, and take note especially of these lines, in truth, the prison unto which we doom ourselves, no prison is.
And my conclusion, after this quick and dirty survey, well, if the digital humanities stakes its claim on an ability to perform or perfect the standard task of interpretation, as we have known it for so long, then it won't work. Because the digital humanities is wrong about what texts are, wrong about what meaning is, and wrong about the possibility of doing without intention. And if the claim of the digital humanists is even stronger, and reaches toward the reinvention not only of humanistic commentary, but of human beings themselves, then I can recognize and even honor it as the latest effort to escape the limitations of mortality.
That is, limitation of being confined within a narrative life, and so being always at the beginning or the middle or the end. And the idea is to forsake that limitation and ascend to a realm of full and immediate knowing. But as I have argued above, that won't work either, short of revelation or an intervention by deity, which I do not rule out but have not yet myself experienced. In short, the larger, more ambitious version of the digital humanists is the latest chapter in an old and noble dream.
Will that dream be realized, at least professionally? Will the outsider status so many digital humanists complain of be shed, and be succeeded by a brave new world in which all the department chairs and all the deans and some of the president's will be digital humanists? The answer to that question lies with actuarial tables. And as I look around at the roster of digital humanists and compare my age with their youth, I can confidently say that for good or ill, the future, and especially the future of funding, belongs to them. Thank you.
SPEAKER 1: Thank you, very much, Stanley. As you can imagine, he is happy to entertain questions, hear responses, [INAUDIBLE], objections, or indeed, even possibly agreement on some of his points. Please. Yes, in the back.
AUDIENCE: Hello, Mr. Fish, my name is Henderson. And I don't have a question, but clarification on your points about digital humanities. It seems like the problem that you have is moving from [INAUDIBLE] data, to interpretation, producing meaning, and god forbid, a new ontological being. To re-mention one of the theories that you have mentioned, Frank Moretti, who is an Italian scholar [INAUDIBLE], who also began a new form of reading called Distant Reading.
STANLEY FISH: Yes.
AUDIENCE: He would say that understanding literature, not by studying particular texts, but by aggregating and analyzing massive amounts of data, it seems then that [INAUDIBLE] is are you against the part of analyzing data, or and against the form of aggregating data? Because I believe that in the form of digital humanities, I understand your point against the production of signification, but isn't the digital humanities in a sense creating a new form of data use that we use as a powerful tool within the school of humanities?
STANLEY FISH: The short answer to that question is no. Although that is the claim. And you're quite correct, of course, that Franco is the proprietor of something that he calls Distant Reading, which is another-- actually a colleague of his pointed out-- will relieve us of the obligation to read it all. You don't have to read books at all. We don't have to do that. All we have to do is manipulate data.
Now the question when you manipulate data, I think, is the question of what is it that guides your manipulation. If, in fact, what guides your manipulation is the kind of intuitive, hypothetical, interpretive guesses that we've always trafficked in in the humanities, then it's hard to know why having endless banks of inert data helps refashion or refine that task. I mentioned the work of Mr. Cohen, the historian named Cohen. Here's a more recent paper that came out of the Stanford Literary Lab, that Franco and others have developed.
And it's a paper entitled, A Quantitative Literary History of 2,958 19th Century British Novels: the semantic cohort method. Now what is the semantic cohort method? Well, it turns out to be a method-- by the way, just as a piece, I don't know, something that's almost, if you pardon the word, aesthetic. When I come upon an essay that has a page in it like that, I want to reach for my gun.
Now what the authors of this-- there are two authors-- what the two authors of this essay have done is they have digitized these novels, and begin to run some numbers on them. And what they found was the following. They found that at the beginning of the century, that is beginning of the, end of the 18th century, beginning of the 19th century, there are a lot of what they called seed words. Seed words which formed a semantic cohort around civilized behavior within stable families in stable, largely rural societies.
But they also discovered as the century progresses, the seed words become less descriptive of general character formations and of moral imperatives, and more descriptive of the rich, and to some extent, the nitty gritty feel of the city. Now in other words, what these people have discovered is that if you read Jane Austin, there are many ways in which it will be different from the experience of reading Charles Dickens. But for them to have-- and these guys know this-- for them to have reached that conclusion is to have had that conclusion waiting, not in the wings, but at the beginning.
Because in order for them to ask questions of the digitalized texts, the huge masses of data, in order for them to ask questions, those questions have to have a direction. And that direction is provided by the very intuitive hunches that they themselves-- that is, all digital humanists-- deride. And what do they say here? OK. Now I'm not going to-- there was a paragraph there that I thought was particularly revealing of what they do. Yeah.
The experience of having to revise our interpretation taught us one of the common pitfalls of interpreting large banks of data, and that is the tendency toward validation. We start with some notion of what we're going to find, and then lo and behold, we find it. Because it is difficult to bridge the distance between signal and concept, we tend to read data in terms of the concepts we already have at hand. In our case, a standard historical narrative that draws a clear line between Victorian and late 18th century British society. Quite true. Quite correct.
And when these authors say, we tend to read data in terms of the concepts we already have at hand, I want to say, as opposed to what. In other words, what other way of reading is there? What other way of reading is there, except for reading as-- in terms of the concepts we already have at hand. The digital humanists, digital humanities, is one more effort to refute, or rather to turn back, the criticism always made of humanist work, that it is not sufficiently scientific or precise or objective. And here we now have a way of joining those critics and showing to them that we too can engage art texts, or our objects of attention with the same scientific rigor that we find in the natural sciences.
It just doesn't work. Or rather, anything that does work works because-- you have again, and I've said this three times already-- already imported the hunches or intuitions that the digital humanist method is supposed to supplant at the very beginning. So what you are attempting to bypass or transcend is guiding your effort. So what you do get is long essays, written by earnest people. I've never read a group of more earnest young scholars. Because they see salvation. They're offering us something.
They see new vistas opening up, long papers that deliver so little. They deliver absolutely nothing. Another one that I recall was a guy who digitized all the mid-century American novels, and then did some searches on the corpus. And discovered to his surprise that there was a large number of foreign place names in these novels, novels by Hawthorne and Melville and you know, Anna Warner and a whole bunch of other people. A large number of foreign place names.
The conclusion that this critic reached was that, therefore, American literature was less narrowly self centered and parochial than it had traditionally been assumed to be. But it only takes a moment of reflection to say that that conclusion is an entirely bogus one, because there are any number of reasons why foreign place names might turn up in a mid 19th century American novel. They might be turn up to be pushed away. We don't engage in any of that fancy stuff. Some kind of wink and nod, Mark Twain move, for example. Or they might be place names that were dropped by characters in these novels we are clearly intended to dismiss as superficial or pretentious or any other number of things.
But you see, my point is always the same. You can't get from the data to anything if you're really playing the game that you claim to be playing, honestly. And the corollary is that you can get from the data to anything, because you have instead of instituting new constraints, you have released constraints entirely.
AUDIENCE: You mentioned funding at the end, which seems to me a really important consideration.
STANLEY FISH: Yes.
AUDIENCE: When I think about the digital humanities, I think about the very real technological shifts that took place, and have driven pedagogy in higher education-- so the shift toward online classes, and the MOOCs now. So the people going into teaching are forced to have to confront a very different landscape of pedagogy. And so when I think about digital humanities, I think about attempts to address that and respond to it, but also ways of providing scholars means of forming communities online, outside traditional institutions, which have been so defunded that the normal job path one have taken in the past, of having a stable, traditional type of job isn't necessarily available for everybody.
So I think that there's a whole aspect of digital humanities that your sort of manifesto approach doesn't really address, because of the hugely shifted-- the funded isn't merely cynical. I mean, if it's gone, it's gone. So if people go to the technology to try and have tools and means of making community and dealing with new pedagogical issues. So I was hoping maybe you might address that aspect, which doesn't really seem covered by the--
STANLEY FISH: You're quite right. You're quite right in several ways. First of all, I don't believe that there's cynicism involved. It is true that you're more likely to get funding for a project if you can dress it up in digitally humanist clothing. Now whether or not that's a set of affairs that will remain, or is simply passing, I don't know. So, and it's also the case that as in the glorious halcyon days of the emergence, in the late '60s and early '70s, of theory, and with that emergence being directly responsible for the formation of this institution, it became incumbent upon departments of English and comparative literature and others to hire a theorist, even though many of those departments didn't know what one really was, and were fairly certain that if they found out, they wouldn't like it.
Now I think now, so and I'm always for things that might allow-- open up job opportunities for my colleagues, and especially for my students. So that's a good thing, even though its basis may finally be specious. But this other aspect which you spoke of, the way in which digital techniques and online teaching, MOOCs, and everything else that's associated with online teaching, may be a response to the changing conditions of higher education in general, and especially to the changing conditions which have been changing in a downward trend since 1960 in the humanities.
I was cheered when I heard a friend of mine-- some of you may know, Andy Delbanco of Columbia University-- give a talk last month in San Diego, where I was doing some teaching. And just in passing, he said, and of course, MOOCs are dead. I didn't ask him what he meant, because I was so happy to hear the news that I thought if I asked the question and got a really detailed answer, my happiness might evaporate. And I didn't want that to happen.
But you're quite right. I didn't address that set of questions. And I really am not competent to address that set of questions.
STANLEY FISH: You could stop right there.
AUDIENCE: I've said all you need to say. Back to that article that came out of the Stanford Literary Lab, I was thinking about what pay off it might have. And that when I read it, I recall it sort of talking about Raymond Williams's thesis on noble communities, so it's more actually engages sort of that--
STANLEY FISH: Oh, you've read this?
STANLEY FISH: Good.
AUDIENCE: For a graduate course, of course. But at the time, I was sort of thinking-- so the thesis is that Raymond Williams was right. OK. But also I was kind of thinking-- we got to talking in the class about kind of the role that this kind of scholarship might play as kind of quantitative confirmation of a hunch. So I was interested to hear what you thought about that, maybe a more generous reading about what this actually offers. It kind of reminds me of, you know, mathematical proofs, where if someone has the hunch, and puts together the proof, and we'll get the quantitative proof later, which is not [INAUDIBLE], but does some kind of work.
STANLEY FISH: Well, I can see, for example, someone might have a thesis about the 19th century novel. And that theses might, how shall I put this, find a useful point of corroboration-- but would it be corroboration-- in the findings of Professor Cohen and his colleagues, that is, the findings about in which decades of the 19th century this or that word appeared often in the titles of books. So you might work something out, and then have a footnote, in the way in which often our footnotes-- how shall I say-- have that opportunistic taint to them.
That is, we know that this is what we're doing is elaborating an idea that is dear to us, and we don't want to let it go. And when we find something in the public record to which we can attach it in a way that seems to suggest a confirmation beyond our own interpretive desires, we seize it. But I'm not sure that there's more to it than that. At least, I can't see it. I don't think that this entire enterprise has evolved beyond Swift's use series argues a fertile and well stocked mind. That's it.
That's the whole game from beginning to end, unless they're playing the other game, which is the game that we're all going to become-- the game that was presaged by the millennial hopes that attended the introduction of the internet. We're all going to be democratized, false hoods and lies were going to be banished because we would all be our own infinitely expanding fact checkers, and all the rest of that. So to use a well-worn political phrase, how well did that turn out? Not very well.
AUDIENCE: Hi. I was just getting to the question. There seems to be a parallel, in the way I hear it, between your argument and something that's happened to the game of chess over the last 50 or so years. Decades ago, Arthur C. Clarke writes a story about a super computer that destroys planet earth because there's this problem originating on planet earth that would take longer than the universe itself, despite that it only had six operators. And the six operators, of course, are pawn, knight, king, queen, rook. And so, of course, the earth gets destroyed rather than possibly infect the computer with this problem and cause it to neglect its other concerns.
But then we have computers now that can't maybe not solve the game of chess in every variation, but that can play chess. And so, I think that if we had-- and part of the reason we have that is because we taught the computers different ways of thinking-- thinking being a kind of metaphor for what the computer [INAUDIBLE].
STANLEY FISH: Look, you're giving away too much by saying that. If you say that thinking is a kind of metaphor, then you're not going to be able to salvage anything in the way of a claim. Either it's thinking or it's not. I mean the Turing test, as you all, I'm sure, know, the Turing test is if, you know, you're shielded from-- you can't see the person who's on the other side of a curtain or of a door, and you're engaging in conversation. And if you can't tell the difference between a human interlocutor and a computer as interlocutor, then, says Turing, you have reached the state where humans are-- where computers are thinking and operating like humans.
And of course, this has produced a whole lot of literature, including a famous set of essays by the philosopher John Searle, pointing out that there's a difference between what's coming back and forth and the ability of either interlocutor to mean anything by what's coming back and forth. So it can't be a metaphor. It has to be real thinking. Some of you may know a book that is now quite old-- not as old as I am, but old-- called, by the philosopher Hubert Dreyfus, come two editions, What Computers Can't Do. And Dreyfus was writing at the time when the chess playing abilities of computers were being touted.
And he had a very simple analysis, simple at least at the-- it was complexly worked out-- for a very simple analysis of what computers can't do. And he says a computer cannot be in a situation. A computer cannot be in a situation. And we are always and already-- to resurrect another old phrase, we are a kind of old friend almost-- we are always and already in a situation. Nor, both Searle and Dreyfus would have added, can you devise a way of getting computers into situations. You can't have.
What you can do, of course, is devise computers that can calculate in infinitely more powerful ways than we can. Now I know there are a lot of people in the AI world who believe that those limitations, which they believe are attached only to digital computers, and there are other kinds of computers, those limitations can be transcended. But I don't think that they can be. And I give as my anecdotal evidence the fact, which may condemn me in some circles, that I am the daily, and one might say religious reader, of the New York Post.
I love the New York Post for many reasons. But of course, one of the reasons I love the New York Post is for the agility of its headline writers, the agility and ingenuity of its headline writers. And I'll tell you one of my favorite New York Post headlines of the last few years. You may recall the name, Bernie Madoff. And so there were endless stories about Madoff, but there were also stories about Madoff's associates, who were attempting-- many of whom were trying to, attempting to avoid jail by ratting each other out.
Now Madoff's right hand man was a man named Ezra Merkin. Ezra Merkin was his right hand man and was involved in a huge number of the transactions of were later found to be illegal and deceptive, and amounted-- not amounted to, were Ponzi schemes. So the Post writes a long essay, not a long, well a full, half page piece about an interrogation that Merkin underwent at the hands of a congressional committee. And here was the headline. His name is Ezra. Ezra Pounded. Now, I will not do this, I promise you. I can do five pages on Ezra Pounded, you know.
And I can get to, I can do it. I would end up talking about some of Pound's Cantos, anti-semitism, Ezra Merkin, whole bunch of things. A computer, I don't care how sophisticated it was, could never get from Ezra Pounded to anything. But what the readers, what the writers, the New York Post writers were counting on is that at least some, and perhaps a significant number of readers, would get it immediately. And not only be amused by it, but perhaps provoked to thought by it. So that's my very anecdotal response. Let's try someone else.
AUDIENCE: Thank you very much for your talk. So it seems that your-- is your criticism-- so I have two questions. One, is your criticism of the digital humanities, then essentially that they are religious without being theological, in may ways?
STANLEY FISH: I think that's quite acute, yes.
AUDIENCE: Second question.
STANLEY FISH: Is that a good thing or a bad thing?
AUDIENCE: I don't know. I mean, it seems like that's a problem for you. I'm not so sure if it's a problem.
STANLEY FISH: Well, it's just that if I'm going to worship something, I'd rather it be God than a machine.
AUDIENCE: Right, but there's like paganism, [INAUDIBLE] but there's like subtleties there.
STANLEY FISH: Right.
AUDIENCE: Among the subtleties, I think, this is the second question, for the digital humanists, one of the things that's precisely being denied is this category of moving from the qualitative to the quantitative, and vice versa. They'll say that numbers and statistics is no different than meaning making. Just like it's no different than moving from physics to chemistry to biology, when speaking of the Big Bang. So the stakes are already set up against them if we're going to judge them based on cohesion and comprehension and a language of agency.
STANLEY FISH: Well, right. And I think the last thing, that last remark is center, the language of agency. Because the language of agency, slash intentionalism, temporarily, linearity, is my language, and the language that I'm defending and indeed celebrating in this piece. That's where, I think what you're saying is very much like what Stephen Ramsay, as I quoted him, is saying. He's saying that this is what we do anyway, deformance, as he calls it, or tampering, or the hermeneutics of just screwing around.
But that's not what we do. That is, at least I don't think it's what we do. Now I'll, of course, be open to testimonies to the contrary. But it seems to me that what we as humanists try to do is figure out what things mean. We try to figure out what is meant by some object or set of objects or set of events in relation to some agent, which might be a single author like Milton, Herbert, or Donne. Or it might be something like the zeitgeist, the spirit of the age. Might be the Holy Ghost, who is inscribed his pencil on the fleshly tables of your heart. Might be any number of things. But that's always the project.
For Ramsay, just playing around is good enough. And there'd be an argument for that, because it leads to what he values-- exuberance, vitality, fecundity, a kind of what we might call a kind of interpretive joy. And I'm too much of a Puritan, in several senses of the word, to be able to swallow that. So I think I have to acknowledge the force of your question. And just in response say something like, here I stand, to go back to the--
AUDIENCE: I have two questions. [INAUDIBLE].
SPEAKER 1: Can you speak a little bit louder so people at the back can hear you?
AUDIENCE: OK. So maybe you can reiterate. If there are any overlap between intentionalist school and digital humanist, or you think they're totally oppositional.
STANLEY FISH: Let me try the first one first, and I will allow you to answer. I think they are oppositional, at least in terms of their rhetoric. Because everything that the digital humanists are against, which are the gatekeepers, people deciding that some meanings are more relevant or more, deserve more attention than others, the idea of a temporal process, a step by step movement toward the determination of significance-- all of these things, the idea that people, and this is very important-- the idea that people can in some sense have ownership of their own ideas, ownership of their own readings, or ownership of anything.
One of the things that interest-- very interesting in the last 40 or 50 years, and many of you in this room will know about this I'm sure, is that there's been a general attack in humanistic and other disciplines on the idea of originality. A general attack on originality, the idea that you can own something, either in the sense of having it copyrighted, or you can own something in the sense of having originated it as an academic, and believe that it's a piece of currency, which should always bear your name.
And again, I must say that I'm invested in that idea because I've been happy, I've been fortunate enough, a, to live for a long time. And b, during that time to have come up with a couple of things, which because of the game that we're in, have not been patented, but I am nevertheless very jealous of them. And when somebody uses the phrase, interpretive community, and doesn't cite me, I get very unhappy, because that's mine. I invented it back in 1967, or some other dark age. I invented it.
And in the academic world, originality is our currency. But at the same time that originality is our currency, many people have been demonstrating over and over again that there could not be, and should not be, any such thing as originality. And of course, these people are putting forward this idea that there is no such thing as originality, as an original idea, to which they then have some kind of claim.
So I think that, I think that for us, as humanists, or for us as people who are looking at artifacts and events of history or other objects of our attention, we have a set toward the project that is antithetical to the digital humanists. And they are correct when they despise it, and when they think of it as undemocratic. When someone calls me undemocratic, I get very happy. What they believe that to be democratic is the general goal of all of our activity. Your second question?
AUDIENCE: My second question is about the appropriateness of comparing grocery shopping list to Paradise Lost.
STANLEY FISH: Yeah.
AUDIENCE: So, if you understand the meaning of each item on the shopping list, in the end, it is very unlikely you get all the items from the same store. You always need something, because [INAUDIBLE].
STANLEY FISH: Yeah. You got that right.
AUDIENCE: So you always need something, or you get something extra. You don't all the time obey or follow on the list of items. The thing about interpretive exercise as a performative one, if you understand each item, each word by Milton, do you feel very guaranteed to produce a writing, interpretive essay?
STANLEY FISH: I'm sorry. Say that again. I didn't get that last part.
AUDIENCE: If you understand each word by Milton, the original intentional meaning.
STANLEY FISH: Of each word?
AUDIENCE: Yes. And do you think that is a guarantee for you to produce something?
STANLEY FISH: No, no, not at all. No, no, no, no, no, no, not at all. Milton himself was an intentionalist, by the way. But he was an intentionalist who precisely wouldn't believe that each word is what you should focus on in trying to determine the intention. Milton, as some of you will no doubt know, wrote a number of divorce tracks in the early 1640s. These were prompted by the fact that he married an attractive young woman and found that she was unworthy to be the wife of John Milton.
Now of course, he would have found that to be true of any woman he ever married, which is something that Robert Graves saw in a novel called, Wife to Mr. Milton. So Milton writes a series of divorce tracks to demonstrate why it is that he should be able to-- why it is that he should be able to divorce without having to meet any kinds of standard, external to his standard, of having decided that this helpmeet was wrongly chosen by him, and so must be sent away.
Now what stood in his way, what stood in his way, because most questions of this kind were simultaneously legal questions and theological or biblical questions. So the question is, what does the Bible say? And always, those who argued against divorce, either and especially against the kind of radically open divorce that Milton urged, was to quote a line from, I believe Deuteronomy, which says that a man can only put away his wife for reasons of fornication. In other words, the only way that you can legitimately put your wife away, that is, turn her away and divorce her as a unilateral act, is if you have caught her in the act of fornication.
And that might seem to be a difficult hurdle for Milton, but he has no problem with it at all. Because, he argues, that when Christ spoke, he was not to be understood in terms of the literal significance of the words he was producing at the moment, but rather he is to be understood in the context of the entire project that he was put on Earth to further. And he points out, in the manner of a literary critic, that when Jesus said this, he was speaking to the Pharisees, who were attempting to tempt him. They wanted him to say something which would mark him as a morally lax person.
And in order to respond to them and to beat them at their own game, he came up with a stricter standard than even the standard that they proposed, the standard by which you can only put your wife away if you caught her in the act of fornication. So, Milton says, that sentence read literally is made for Pharisees. But, he says to his readers in an unaccustomed moment of generosity, you are not Pharisees, therefore this does not apply to you. And as far as you are concerned, you virtuous person as I am, you can put away your wife for any reason you like.
I mean, I'm not saying this is a persuasive argument, but man, it's massive. It's worthy of-- Seneca wrote a series of essays, Controversiae e Suasoriae, in which the whole point was to imagine an argumentative situation in which you wouldn't think that anyone could argue himself or herself out of it. And then to figure out how it's done. So it's not a question of each individual word. And look, we're playing this drama out right now. And I don't know, has the Supreme Court today issued any of its long awaited rulings?
Because in the Affordable Care Act case, King v Burwell, which is the case which will decide whether or not the Affordable Care Act can be sustained, at least in the form that it now has, everything hangs on a phrase, a four word phrase, exchange-- no, a five word phrase-- exchange established by the state. Exchange established by the state. You should read the oral arguments in King v Burwell. They're wonderful. And it's a 900 page-- no, it's a 2000 page law, the Affordable Care Act.
And there's these four words. And the literalists, of course, are arguing that those four words have a clear meaning, and that meaning means that the machinery now set up in 36 states whereby all of the mechanisms of the Affordable Care Act flow through federal agencies, that those 36 states are not in conformity with the law, and therefore their entire system of Affordable Care Act must be ended. And the argument on the other side is the Milton argument. It's the spirit of the act.
You should read those four words in the context of what the act, and what the framers of the act intended generally to produce. This decision will come down, if not today, than next week or perhaps the week after. But I think either today or next week. And when you read it, what you're going be able to see, what you will see is that what the Supreme Court is doing is debating the very issues of interpretive method that I've been trying to describe today.
AUDIENCE: Thank you very much for your talk. I was not familiar with digital humanities, and you actually managed to scare me.
STANLEY FISH: I thought you were going to say, you're going to run and--
AUDIENCE: I'll tell you, for me, every time names and words turn into numbers, I start to get a chill. And I was wondering, because you said that their problem is that the articles aren't accurate. But it seems to me, the project that seems to, the underlying project is the erasure of the poetic. Because you were talking about, you were quoting Donne and Herbert on spelling. And I remembered [INAUDIBLE] line, you have to spell it wrong to read it right. For me, poetry is deforming, is deformation.
STANLEY FISH: But that can go in both directions. That can go in the direction of the gentleman whose two excellent questions posed a few moments ago. I don't want to get into arguments about, well, in fact, I would love to. I myself have, of course, a theory of what poetry is. But again, I won't burden you with it. But for me, the issues here are conceptual. I'm trying to figure out what do these guys think they gain by this. And now one of the things they gain is what the possessors of all new methodological engines gain, and that is they gain a kind of social/professional position of one-upmanship over the rest of us.
If someone were to now rise in the audience and say well, how sophisticated are you as a computer user, how many banks of data have you ever worked with, and other similar questions, I would have nothing to say. Because the answer would be that I don't know a thing about it. All I know is that, or at least all I think I know, is that the project itself is hopeless. Although one can see its attraction, both in its instrumental version-- that is, we can help you do what you've always done better-- and it's more ambitious version, we can help turn us all into interconnected nodes in a wonderful universe where relationality replaces temporality, structure, and gate keeping.
Both of those visions seem to me to be uncashable. And that's what I've been saying today.
AUDIENCE: Yeah. Thank you. I was wondering, it seems like in sketching out your paper, you are really kind of reiterating very old arguments [INAUDIBLE]. So I'm just wondering if, and obviously you yourself [INAUDIBLE] talking about the law and talking about stylistics and things like that. Now I'm just wondering what specifically, do you see this just as an old argument, or is there anything specific to the digital?
STANLEY FISH: It's an old argument in a way, but of course, it's very new. Because the claim that these are new tools which can do extraordinary things is, of course, a correct claim. But the issue of whether, the issue of how to move from the observation of formal features to some kind of interpretive conclusion, and then the next step, and what kind of reasons can be given or evidence amassed in support of your interpretive conclusion. This is a very old question. And it will be continued to be debated.
And it's a question that's also the relationship, a question about the relationship between linguistics as a discipline and literary interpretation or literary studies, at least, as they have developed since the end of the 19th century. There was a time back in the late '60s and early '70s where people in our line of work really did believe that Chomskyan linguistics, and especially with its talk of generative structures and deep and surface structures and semantic trees, could provide a secure and firm basis for literary interpretation, and replace the impressionistic and essayistic writings, either of critics of the 19th century, or even of formalist critics of the 1940s and '50s with something more firm, something that people would respect, something that legislators might have to recognize. So it's an old game and a new game at the same time.
AUDIENCE: I just want to register my skepticism of your [INAUDIBLE] too deep for individual human [INAUDIBLE] in their lifetimes are necessarily the kinds of things that are not intentional. Because it seems to me that you're open to the idea [INAUDIBLE]. So it seems to me that we might think about [INAUDIBLE] like literary texts, let's say. So why not treat that set of data as itself a text that we ought to do the work interpreting if we're interested in [INAUDIBLE]?
STANLEY FISH: Well, you'll recall that I said that I had no quarrel at all with the usefulness of digital computer techniques to the problems of attribution, which are, in effect, problems of fingerprinting. The question is, once you have identified, for example, the fact that in a certain author's text, a definite article or yeah, definite article is going to turn up every five words, whereas in the text of other persons who are used for a control group, the statistics are quite different. And I think if you can find things like that, and a number of them, you can go a long way.
And this is, of course, the same kind of technique that people use in trying to determine whether this painting was done by one of Rembrandt's underlings, or by Rembrandt. Once you've done that, once you've done that, let's say we'll stick to the literary example, do you then want to say that what you found is significant of something other than the fact of an identifying tick. And I'm not sure how you can do that. I'm not sure how you can, unless you begin to have some kind of theory of the definite article, which I cannot even imagine what a theory of the definite article would be. So no, I'm afraid I remain unrepentant. Look, it's getting late, isn't it?
SPEAKER 1: Couple more questions.
STANLEY FISH: Right.
SPEAKER 1: [INAUDIBLE] with very red hair.
AUDIENCE: So I should probably, in terms of context, say that I'm a computer scientist, not a literature person. So I'm coming in [INAUDIBLE].
STANLEY FISH: You really know what you're talking about, as opposed to me.
AUDIENCE: That's debatable. But I was curious, first off, because a lot of the things that you're talking about, from the perspective of somebody doing artificial [INAUDIBLE] and I think more and more, the idea of common words [INAUDIBLE], and I think more and more, I think, the computer science and the humanities communities are agreeing that that's not always the best way of [INAUDIBLE] information.
Because as you described, there's the [INAUDIBLE] long enough, you'll find something that you can talk about [INAUDIBLE]. But I think that more and more often there's also papers that will take [INAUDIBLE] hypothesis and say well, this literary movement [INAUDIBLE]. This kind of [INAUDIBLE] element in the way that [INAUDIBLE] evolves over time in two different countries, or something like that. [INAUDIBLE] Do you have any thoughts on that sort of hypothesis [INAUDIBLE].
STANLEY FISH: Give me the hypothesis again, because everything that you said sounded perfectly reasonable, which of course is distressing to me.
AUDIENCE: So, one of the things that I saw recently was [INAUDIBLE] but there was a group of people, I think David Smith might have been one of them, that were looking at the movement and structure of stream of consciousness text. And they were saying, well, in different texts that we've seen so far, that are dealing with specific analysis, the detail of close reading, we're seeing [INAUDIBLE] types of things that are being repeated are the signs of stream of consciousness.
So if we model those and what that should look like, and we look at these other types [INAUDIBLE] find things that are stream of consciousness like, and then analyze them and see [INAUDIBLE] English are then pulled out of say, Japanese texts as being similar, and things like that. So they're doing hypothesis testing, instead of saying, let's count up the things and see whichever--
STANLEY FISH: I think, yes, they are doing hypothesis-- what was that? Hypothesis testing, is that your phrase?
STANLEY FISH: Yeah. I think that's perfectly reasonable. And that also corresponds to what I said when I talked about the paper that you and I have both read. Your name is?
STANLEY FISH: Oh, That's good. That you and I have both read. That is, if you begin with some kind of hypothesis and intuition, and then use that intuition to generate some questions that could be put to a corpus, then you're playing the interpretive game as it has always been played. And it seems to me that you might have, that there might easily be something useful and satisfying about it. But that's kind of modest compared to the claims of the people that I have, from whose texts I have read you quotations.
SPEAKER 1: Maybe one last question, Stanley? Yes.
AUDIENCE: Professor Fish, this question is a kind of a question that has so many pieces from so many questions that have been asked. [INAUDIBLE] digital humanities as a project, right now it is at a stage where it is analyzing things that have been written. They took technology from the computer, it was initially used to analyze the games that people had already played. And then [INAUDIBLE] Deep Blue, which was playing against Kasparov [INAUDIBLE]. So do you see in 20, 25 years from now, a computer generating a new text from the same data that [INAUDIBLE].
STANLEY FISH: No, I don't. I see that as analogous to the old, what if 300 monkeys sitting around typing in a room after 20 years produces Shakespeare? The answer to that is that it isn't Shakespeare. What they have produced is something that resembles Shakespeare. Scalia, when reviewing a book by a friend of mine, Stephen Smith, who is a legal theorist at the University of San Diego, resisted for some of the same reasons that digital humanists would resist, resisted the idea that only persons can make meaning, i.e., intentionally.
And he gave the following-- he, Scalia-- gave the following example, which he thought-- and I should say, by the way, as I begin to mock Scalia, that he's a friend of mine. But nevertheless, hey, what is scholarship for? He, Scalia, thought that he had refuted Professor Smith by this example. There's a factory full of workers. And it's well known that if a certain bell is rung, what the ringing of the bell means is that, for whatever reason, the building must be evacuated. OK, you got that?
The bell is rung. And, says Scalia, it wasn't rung by anyone. It was rung by accident by a monkey. He said, by a monkey. And he says, doesn't that demonstrate that, and that the ringing of the bell was understood by those who heard it to mean, it was a sign, to mean evacuate the building, even though a monkey accidentally, even though a monkey accidentally rang the bell.
The answer to that question is easily offered in the form of another question. What if the workers in the factory were in the process of exiting from the building when over the loudspeaker came the information that the bell was rung by a monkey? Would they continue to evacuate the building? The answer is no, because they would now understand that the ringing of the bell didn't mean anything. Because no intentional agent designed it. And of course, then there are cases of the monkey that is trained by somebody to ring the bell, which would make that somebody the agent.
Look, I want to thank you all for your attention. Wait a minute. I'm not through. But I want to say something to you. I have been associated with the School of Criticism and Theory now for 38 years, since 1977. And I can tell you, some of you who are here, from what Jonathan said, some of you are here in your first day, first official day of the School of Criticism and Theory, that there is no experience like it in the academic world.
I've been to my fair share of exotic places where supposedly great conferences and other things are happening, in Italy and France and et cetera. But there's nothing like the School of Criticism and Theory. It is, to use a word that is overused these days, truly transformative. Not only will the School of Criticism and Theory engage you in a level of thought that you would have thought beyond your capacities, but of course your capacities will then be enlarged, as will your friendships, and as will the prospects for your professional life and for your personal happiness.
And that is a large thing to offer in the name of an institution, but I offer it confidently, and I say to you, congratulations for being here. Thank you.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact email@example.com if you have any questions about this request.
Stanley Fish, Floersheimer Distinguished Visiting Professor of Law at Cardozo School of Law, spoke at Cornell on June 15, 2015, as part of the School of Criticism and Theory public lecture series.