Milton and the Machine

by Simon DeDeo

The first line of Paradise Lost is wrong. Most people, most professors of English, even, don’t notice. Even if they remember the line, which is one of the grandest openings in the language.

“Of man’s first disobedience and the fruit”—the solid man, and then first, ripening as it spreads out on the tongue. The fall of letters like rain in that long, classical, scolding disobedience. Fruit, a wide vowel brought to a sudden close as the enjambement tumbles onward. Do you see the error yet?

The line is a syllable too much. In Milton’s blank verse epic—iambic pentameter, five sets of two-syllable feet—the opening has eleven syllables, not ten.

Most people do miss it—it’s a good game at your next literary cocktail party. Perhaps the ear needs time to adjust. Blank verse may not be as natural as it seems when, fifty lines in, the brain engages and the patterns flow back and forth, from the text to the mind and back, as the mind puts the template on. Even the best musicians need to be counted in.

I never noticed, certainly, when I first read the poem on a Kindle, packed in cacophonous train car on the Tokyo subway. In fact, I never noticed at all, until years later I fed the poem into a machine.

Output from POPE-R, scansion prototype, operating on the opening of Paradise Lost. 1: primary; 2: secondary; 0: unstressed. Lab for Social Minds, Indiana 2015.

I’m a scientist as well as a poet. I study complex systems, which at this late stage in humankind’s knowledge means that I get to study everything we don’t understand. I focus on the products of the human mind, which makes me a cognitive scientist. And I focus, in particular, on what happens when minds get together, and so that means that, with some extremely charming collaborators, I get to study everything from squabbling birds to Wikipedia editors, the spiraling-out of the French Revolution and the Machiavellian strategies of Midwestern gas stations.

People do many things when they get together, but one of the most beautiful is the writing and sharing of poems. So when I fed Paradise Lost to my machine, on a cloudy and humid afternoon in the CUNY Graduate Center—the statistical patterns of the cars honking six stories below are still in my mind months later—I had some serious goals.

I wanted to study surprise, fascination, attention, what draws the eye. I knew about some laboratory studies from the University of Southern California, the so-called Baldi-Itti model, that had found a new way to quantify surprise. Baldi and Itti could take a video clip, and predict where people would look: not at the waving grass, no matter how fascinating, but the soccer player running over it. Not at the rush hour cars, but at the motorcyclist who turns the wrong way up a one-way street.

I wanted to measure Milton’s surprise: his wrong-way cyclists. But in order to do that, I needed to know his structure. We know what people see when they look at a video: light, colors, motion. But what do we hear when we listen to Milton? Far, far more—the words, yes, their meanings and connotations; but also their valence; their rhetorical figures, literal or metaphoric; how they glue together, their syntax, recursive and branching or centipede-like. And, of course, their sounds, their vowels, their consonants; where the stress falls.

As happens to scientists, just as much as to poets, I tumbled down the rabbit hole. I set myself what I thought was a simple goal—I would write some code to find the stress, the lexical stress. Dis-o-BE-di-ence; PA-ra-dise; where our voices linger when the word is on its own. And I would track the rate at which Milton could align that stress with the iamb, that ancient pattern of soft-hard, soft-hard, the metronome of English poetry, and often prose.*

If I had thought a moment, I would have realized how hard the problem is. It’s hard enough to stump Google, as you can verify by listening to Google Maps struggle to turn a string of letters into a sequence of sounds. They might hard-code in a word like Indianapolis, but if you drive in the American Southwest you quickly realize the arbitrary mess of convention, memory, and history that makes Cerrillos Road ser-RI-os, not cer-rill-US. My code struggled and failed; I taught it to search the web for new pronunciations, new accents, spidering out to the Oxford English Dictionary and, eventually, Wiktionary, a hot-bed of nerd phonetics.

This is a blog post, not a press release: I still don’t know where the surprise is in Milton. I have a statistical hold on it, but the error rate is high. It is easy to horrify a scholar—tell them the errors will come out in the wash. And they just might; enough lines of Milton might mean that the code doesn’t have to work perfectly. A ten or twenty, or even thirty percent error rate might be enough to pull a weak signal, a thin mathematics, from this artifact washed up, as it were, on the human beach. But not yet.

And that first line? I’ve tried the Milton question on anyone I think will be annoyed when they miss it. Jess, a supremely educated barrister in London, turned the tables on me by finding the solution as she spoke the line aloud. Disobedyence—like the ny in canyon, or the Spanish ñ. Turning a pair of vowels into a consonant is a trick, but Milton’s in good company: Virgil did the same in the second line of the Aeneid. As of yet, a machine hasn’t beat them.

*Perfect iambic tetrameter: there are no fractures in your foot.

agni blog logo smaller

DeDeoSimon DeDeo is a professor of Complex Systems at Indiana University, and external faculty at the Santa Fe Institute. At Indiana he runs the Lab for Social Minds, which studies the present and past of the human species to better understand its future.


27 thoughts on “Milton and the Machine

  1. As a scientist, you may have applied the Scientific Method, and seen if your result is repeatable. It is not. Milton uses this very same word (I am ignoring “disobedient”, though that produces the same result) as follows:

    By sin of disobedience, till that hour

    –Book VI, 396

    Of disobedience; firm they might have stood,

    –Book VI, 911

    And disobedience: On the part of Heav’n

    Book IX, 8.

    In every case, your “problem” only results with your American pronunciation of dis-oh-bee-dee-yents. If you pronounce the word DIS-oh-BEE-dyents, there is no problem.

    And I am confident that, if any line has been looked at by scholars more than other lines, it is the very first.

    Liked by 2 people

  2. Who is to say how Milton pronounced his English of his day, whether one pronunciation was preferred over another? Did dictionaries offer syllabification as a tool for writers or were writers left to their individual ideas for proper usage? Language is a living thing…

    Liked by 1 person

  3. Oxford University Press online indicates that the syllabification of “disobedience” is dis-o-be-di-ence, consistent with the author, but this does not make it the only possibility. It may not be wrong to break the ten syllable rule of the iambic pentatmeter if logic, grammar and pronunciation dictate. Eleven isn’t wrong, just different!

    Liked by 1 person

  4. Pingback: Milton and the Machine | R.M. ENGELHARDT

  5. English is one of the most volatile languages in the world. Greek (which is a major component of my scholarly work) is an example of a relatively stable language: if we went back in a time machine and spoke in modern Demotic Greek with Plato, he would understand us, though he would find our accent rather odd. However, if we were to go back only to the time of Shakespeare we would find his English difficult to understand (as every high school student knows), and by the time we get to Gawayn and Beowulf, it is incomprehensible without specialized study. What is more, the volatility of English results in a wide variety of accents, some nearly completely opaque to outsiders – the Scouse of Liverpool, the accent of rural southern Alabama, Zimbabwean English, and so on – and this is despite the universalizing influence of the mass media, which are imposing an artificial “correct English” on the entire English-speaking world. In Milton’s day there was no such influence, and if one went just a few miles outside of London one encountered varieties of English that were very hard to follow. I have read a scholarly work on the pronunciation of English in Shakespeare’s time, and it was evidently NOT in the declamatory manner developed in the nineteenth century and enshrined today by such actors as Laurence Olivier — it most resembled the accent of the Deep South in the United States.

    As a result, what the OED or any other MODERN dictionary says is the proper pronunciation of “disobedience” is entirely irrelevant. I recently spent some time travelling in North Wales, and I was struck by the vast differences in how English is enunciated there. And, no, we do not have recordings of Milton reading Paradise Lost aloud, but we do have his text, and linguists and philologists can approximate his pronunciations therefrom with considerable precision. (I do this rather often myself with classical Greek.)

    If EVERY time the word “disobedience” and “disobedient” appears in Paradise Lost there is what seems to a modern person an extra syllable — see my list above of other uses of the word and you will see this apparent “extra syllable” DOES appear every time — that tells us that he was not making an exception, that in fact he pronounced the end of the word as one syllable, “dyents”, not two syllables (dee-yents). If there were exceptions in which he handled this word as a five-syllable structure, I’d be open to the possibility that once or twice he might have made an exception to the iambic pentameter. But Milton was a very careful poet, and I fail to find exceptions in his iambs – ever. They are just not there. Thus, he did not make an exception here either; that is just a wrong explanation of the situation. The only tenable answer is that English has changed, and that the pronunciation of “disobedience” is not today what it was for John Milton.

    Liked by 3 people

  6. Larry Bole

    Scansion is not a cut-and-dried thing. Neither is iambic pentameter. Many poets, when using iambic pentameter, have occasionally employed ‘variants’, meaning adding a syllable or omitting a syllable in a given line. To say a line of iambic pentameter is ‘wrong’ when this happens is a rather rigid way of thinking.

    As others have pointed out, Milton may simply have elided a syllable. And if he didn’t, maybe he is demonstrating an example of continuing disobedience.

    Liked by 1 person

  7. Lovely to read the comments; I’m absolutely thrilled this short piece has occasioned so many thoughtful responses.

    I do hope it is clear that when I say “the first line of Paradise Lost is wrong”, I mean only from the point of view of a modern-day speaker counting beans: not the sensitive and intelligent readers I go to parties with and meet on the internet. I do find it interesting that we modern speakers (in many, though not all, dialects today) naturally read that line with 11 syllables, but fail to notice. There’s an enchantment to the line that goes beyond the form.

    Larry writes quite simply that “Milton may simply have elided a syllable”; this in contrast to the alternate pronunciation that James notes as a way to explain the four instances in one go.

    An interesting follow-up to Larry’s point is that, on average, Milton’s lines—scanned according to modern pronunciation—are 10.5 syllables. Which means either that, in Milton’s day, English itself was more clipped; or that Milton is pushing his meter to the limit, and demanding more elisions that a reader, modern or contemporary, would naturally make. (In computing that average, I don’t give Milton a pass on explicit elisions such as th’).

    If you expand the figure in the post above, you can find a few other lines to brainstorm, marked as “Extended Line” (“Parse fail” is where the machine encountered a line with a word it couldn’t break into syllables with known stress). Another extended line in that image is “Invoke thy aid to my adventurous song”; the likely solution is “ad-ven-truss”.

    I would love if the readers we’ve gathered, briefly, here, would continue to write. And, James, Allen, Larry, and any others, please feel free to contact me over e-mail if you feel I can be of help to you in your own thinking or reading.


    1. One might also want to take into account how Milton originally wrote it, in terms of spelling, which could affect pronunciation.

      Doing some searching online, I find myself in ‘google books’. One such is “The Poetical Works of John Milton,” edited by David Masson, published by MacMillan in 1890; in which is written in an included essay: ” ‘Ventrous’ and ‘Adventrous’–The first occurs three times, spelt so or ‘vent’rous’; the second six times, spelt so or ‘advent’rous’.”

      Although I’m not a teacher, if a student of mine pointed out the extra syllable in the opening line of “Paradise Lost,” I would be pleased that the student was paying attention, but I would then discuss the issue of variations in iambic pentameter in English poetry. I guess I would have to read and make use of another ‘google book’ I found: “The Strict Metrical Tradition: Variations in the Literary Iambic Pentameter from Sydney and Spenser to Matthew Arnold,” by David Keppel-Jones, published by McGill-Queen’s University Press, 2001.

      For some easier-to-access information about variations in iambic pentameter, the Wikipedia entry for “Iambic Pentameter” has some discussion of this.

      At least one source (I now don’t remember) that I found online said that the variations in Milton’s iambic pentameter in “Paradise Lost” lends a “musicality” to his versification. Anyway, to my mind, ‘irregular’ is not the same thing as ‘wrong’, at least in literary terms, except perhaps to pedants (not that I’m accusing anyone of being pedantic in this discussion).

      Liked by 1 person

  8. Yes, I hate to say it because Mr. DeDeo seems so inversted in this, but I agree with James Audlin that the author is wrong. It is common to encounter different pronunciations in old English poems, and Mr. Audlin is absolutely right in his logic that since this occurs in the exact same way with every usage, Milton clearly heard, pronounced, and counted it as ten syllables each time. I will add that I wouldn’t call the first line “wrong” even if it did have 11 syllables (Shakespeare’s Iambic Pentameter is full of extra syllables, for example), but I think we can conclude with near certainty that for Milton and his contemporaries it had 10.


  9. The machine is the pedant, I think, whose exacting mind provides a novel point of reference. We can’t fudge its output; the reasoning may be wrong, but its strict reliability provides a meter-stick that reveals a great deal.

    One question we don’t have an answer to, and which the conversation is circling around, is the extent to which the deviations it finds are due to differences in pronunciation (James) vs. the need for “unnatural” elisions (Larry) vs. (finally) the possibility that Milton wants to break the line itself (Allen).

    We really don’t know how Milton spoke or (rather more importantly) how he imagined his readers to speak. We’ll need to cross-reference words; if we mark the syllables like this, for each occurrence, do we regularize the line? The tension between (say) explaining away N unusual lines with one pronunciation revision declines with N; if I can make twenty lines “perfect” by revising “disobedience”, I likely should, but what about two?

    On this note, James, you write that “But Milton was a very careful poet, and I fail to find exceptions in his iambs – ever.” Really, ever? The opening foot of a Miltonic line tends towards inversion—it is hard for me, at least, to not put the accent on “sing” in “Sing, heavenly muse”, or to not say FAV-oured instead of fav-OURED in “Favoured of heaven”.

    I’d be curious to hear thoughts on this aspect of the problem: the relation between lexical and poetic stress (in our first runs, Milton seems to align very little with what we can reconstruct).


    1. I just read that in Elizabethan English “Heaven” was pronounced as one syllable, which shows how impossible it is to do a complete analysis of stress of a language no longer spoken and never recorded. It also accounts for what you have wrongly called an “extended line” in lines 6, 9, 27, and 30, etc. I have to disagree with NJE. It does not seem that the author was aware of and accounting for the pronunciations of the time.

      But, if you want to discuss the line in modern American English—”Sing, heavenly”—the phrase would be SING, HEAV-en-ly, MUSE, Stress, Stress, Unstress, Unstress, Stress. While the verb command “Sing” would ordinarily be stressed, it is “HEAV” which absolutely must be stressed because that is the stressed syllable with that poly-syllabic word, and within a strong iambic pattern, that could coax one to not stress “sing” and allow it to fall into the macro pattern. That effect might have been still stronger when HEAVN-ly was a two syllable word.

      In line 13, I suspect that “adventurous” would be one syllable less than you have accounted for, and in line 18, the function of the contraction th’upright is to contract it to two syllable not three. None of these lines are extended.

      I cannot immediately account for line 17, however.


      1. As I pointed out in a previous post, according to David Masson in his “Poetical Works of John Milton,” the word we now know as “adventurous” appears six times in Milton, ‘spelt’ either ‘adventrous’ or ‘advent’rous’ (and three times as merely ‘ventrous’ or ‘vent’rous’).

        By not including the third syllable of the modern ‘standardized’ spelling of the word “adventurous” (and under the influence of social media, who knows what the ‘standardized’ spelling of words will be in the future?), I assume Milton was spelling the word as he took it to be pronounced (without the modern third syllable), or wanted it to be pronounced, rightly or ‘wrongly’.


  10. P.S. I’ve only read excerpts of “Paradise Lost” in English Lit classes in high school and college. I know that Milton is a great stylist in English poetry, but I’m not going to read the entire poem even now.

    By the way, I remember (at least I think I remember, from when I was living in NYC) hearing a talk by a woman who wrote a memoir about setting herself the task of writing out “Paradise Lost” word for word in longhand. Does this ring a bell with anyone?


  11. NJE

    I think that some of these comments miss the point here. The author is undoubtedly aware of the English pronunciation of the word “disobedience” and the problem it poses for his project–he says as much. He also indicates that the post is a blog post, and the implication is that this is an exploratory project. The interesting thing at the heart of this is that there are fascinating questions about the poetics of surprise that we may be able to observe with a new set of computational linguistic tools. These tools do not replace other kinds of literary, linguistic, or historical work, but they do introduce us to another possible way to see patterns in poetry and language. The post shows us an honest account of some of the difficulties of doing that kind of work, and the opening is nothing more than a provocative account of an initial impression that is later called into question.

    Liked by 1 person

    1. Concordances existed before computers, as well as analyses of such things as irregularities in the iambic pentameter of such works as “Paradise Lost”

      For a while now, there have been a number of articles written about such projects as using a computer to do these things faster and easier than was done in pre-computer days. Other such projects have attempted such things as charting changes in social attitudes by counting the use of certain words, representing different social attitudes, in print (books, magazines and newspapers) over lengthy periods of time. The computer does this fast and easily, but it doesn’t help in the interpreting of the meaning of changes in frequency of the usage of certain words.

      There have even been attempts (not yet totally successful) to ascertain whether an otherwise anonymous writer is male or female by analyzing the language being used, for which task a computer would appear to be ideally suited.

      All very interesting, but that begs the question of the validity of underlying assumptions, such as labeling something as being ‘wrong’ vs.’irregular’, or what constitutes ‘male language’ vs. ‘female language’.


  12. Excellent. I find that English iambic pentameter is not my “bag,” I should study more. I too have read only excerpts of “Paradise Lost” (book X), I found it boring as all get out! I prefer the alexandrine verse of Baudelaire’s “Les Fleurs du mal” of which I am currently stuck on “Les Bijoux”. Baudelaire rhymes “aimer” with “mer” (they do not rhyme!). But they do look alike. Perhaps our difficulty is in the number of different pronunciations for disobedience in centuries old English, both five and four syllable versions might garner support? Just asking…


    1. Allen, I understand what you mean about “Paradise Lost” being “boring.” I’m reminded of Gary Snyder’s poem, “Milton by Firelight,” which starts out by quoting Milton: “Oh hell, what do mine eyes with grief behold?” and ends the first stanza with Snyder saying, “What use, Milton, a silly story / Of our lost general parents, / eaters of fruit?”


      1. Thanks Larry, I didn’t think I was alone. I am finding the work more interesting as it is non-rhyming blank verse, I didn’t recall noticing the lack of rhyme when I first encountered the poem. Greatly influential to modern thinking.


  13. Again. Even today we find multiple pronunciations of the same word, “aunt” may be pronounced “ant” or like “aren’t” (without the “r”.), “Oral” may be “ore-ul,” or it might be pronounced “aural”. It is sufficient to say that even today we don’t all agree on linguistic matters.


  14. Milton’s blank verse as used in “Paradise Lost” does not rhyme! Wonderful! I found a line from Book II, line 4, “Showers on her king, barbaric pearl and gold,”. An eleven syllable line by today’s syllabification, (spelling perhaps modernized?) I don’t know if there are any more…


  15. I found two more lines of undetermined length of syllabification. Both come from Book I, lines 15 & 17. I quote: “Above the Aonian mount, while it pursues” and “And chiefly thou Oh spirit, that dost prefer”. Continued reading finds more examples of multisyllabic verse. The poem is available online.


  16. Then there is the fundamental question of whether the information produced by such computer analysis is significant or trivial. Does the fact that there are a significant number of irregular lines of iambic pentameter in “Paradise Lost” mean that Milton is in some way deficient as a writer of blank verse?

    My impression is that critics have known for a long time that there are a number of irregular iambic pentameter lines in “Paradise Lost.” But they probably never worked out the overall average of the number of syllables per line in the poem. What new insights are arrived at by knowing this?


    1. I see. Yes, the information is new to me, so I like the confirmation that my writing, which tends to be off ten (often) or off twelve (alexandrine rhyming verse translations) is not so bad being “imperfect”. I was not aware of the number of eleven syllable lines might be contained in Milton’s work, and I do try hard (when writing hexameter for example) to get my lines to be just right…


  17. Pingback: NRR 2015: Looking Back | the AGNI blog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s