Attridge and Staten 2: The Sick Rose [+GPT-3]

Bumped to the head of the queue from 2015. Is it GPT-3 week at New Savanna? Ken Lacker quizzed he/she/it (do we need a new pronoun?):

Q: Which colorless green ideas sleep furiously?
A: Ideas that are colorless, green, and sleep furiously are the ideas of a sleep furiously.

Some years later poet John Hollander wrote a poem entitled "Coiled Alizarin":

Curiously deep, the slumber of crimson thoughts:
While breathless, in stodgy viridian,
Colorless green ideas sleep furiously.

The (metaphysical) distance between GPT-3 and Hollander is HUGE, ludicrously monumental. I discuss this poem in considerable detail below.

* * * * *

In my first post [1] I commented on Attridge and Staten’s conception of minimal reading in “Reading for the Obvious: A Conversation” [2]. Now I want to comment on their remarks about their example poem, William Blake’s The Sick Rose. My one complaint is that they didn’t finish the job. By that I DO NOT mean that they didn’t describe – my word, not theirs – everything, for that is all but impossible, and they say so. No, what I mean is that after a nice ambling conversation they didn’t conclude with a summary statement of their points of agreement.

While I’ve thought casually about doing so myself, I’ve thought better of it. But I think I’ve quoted enough of their discussion to give you and idea of what they’re up to.

Before proceeding I should note that I’m not a Blake scholar [3], nor a romanticist, nor even a specialist in poetry. It’s not clear to me what I am. I suppose I’m a theoretician and methodologist who pursues the craft mostly by examining texts in some detail.

* * * * *

The Sick Rose

O Rose, thou art sick.

The invisible worm,

That flies in the night

In the howling storm:

Has found out thy bed

Of crimson joy:

And his dark secret love

Does thy life destroy.

Note: Occasionally there is an ellipsis in their text, indicated in the usual way. In those cases where I introduce an ellipsis into a quoted passage I’ll indicated that by enclosing the ellipsis dots within angle brackets, thus: […].

The Obvious

Henry Staten notes that that, in Attridge’s book, The Singularity of Literature, he asserts that the poem deploys “syntax to achieve an unrelenting forward drive that climaxes on a single powerful word” (66). He then quotes this passage:

The simplicity of the strongly articulated phrasal movement contributes to this experience. The arresting initial statement, “O Rose, thou art sick:” – one line, two beats – is followed, after a pregnant pause, by an extension that takes up the seven remaining lines. This extended elaboration of the opening line is made up of three lines of anticipation, followed by the stanza break which further heightens the tension, and then a four-line arrival. And those three lines of anticipation form a crescendo of intensity – “The invisible worm / That flies in the night, / In the howling storm,” – while the stanza of arrival varies the 1:3 balance of the first stanza by taking the reader through two climactic statements of equal length: “Has found out thy bed / Of crimson joy; // And his dark secret love / Does thy life destroy.” (69-70)

Let me supplement this in a rather pedestrian way. As punctuated, the poem consists of two sentences: 1) the first line, and 2) the other seven. That second sentence thus loops over the boundary between the two quatrains. In some sense, Blake is signaling to us, that first line has a (logical) weight equal to that of the other seven lines. The second sentence would seem to amplify the assertion made in the first. The rose’s sickness, a property of the rose, is caused by the invasion of this foreign body, this worm.

Yet me further observe that the poem is addressed to the rose. The rose is not being talked about in the third person, as the worm is, but is included in dynamics of the poem itself. Thus a second person pronoun appears in the opening line (“thou”) , in line 5 (“thy”), and in the final line (“thy”), but not in line 2-4, which are given over to the worm. Thus the worm and the rose have a different relationship to the dialog itself.

And then we have the use of “thou” and “thy”. I’m not sure how to read that. We are dealing with what linguists know as the T–V distinction (after the Latin tu and vos). In Early Modern English “thou” and “thy” would have been used for familiars and subordinates while “ye” and “you” were used for strangers and superiors. But the use of “thou” and “thy” had pretty much disappeared from ordinary discourse by Blake’s time. But those forms have continued in restricted contexts where they “seem pious and (ironically) more formal and respectful than the everyday ‘you’” [4]. What was Blake signally in this usage, familiarity with the rose, or respect for it?

I don’t know how to resolve the matter, though it is quite possible that a Blakean would be able to do so. In any event, I think the precise valence of the second person address is less important than the fact that the poem is addressed to the rose. What that gives is a poem that begins 1) by addressing the rose, then 2) takes a three-line excursion to talk of the worm in third person and then 3) returns to addressing the rose. We might also read a spatial progression where the image of the invisible worm in the storm invokes large unbounded space, the image of the roses bed locates us specifically in the immediate neighborhood of the rose, while the concluding couplet implies that the worm is destroying the rose from within–even as we are addressing the rose.

Let’s move on to a comment by Attridge:

You may have doubts about the symbolic suggestiveness of rose and worm in themselves, but what about the sexual symbolism in their conjunction? Sexual symbolism in its crudest form–the phallic worm, the vaginal rose, the bed: is this not obvious to all readers, and thus part of what it means to take the poem at face value? It relies on the most basic visual associations to do its work (again, a touch of the child’s imaginative world?). Yet the implications of this interpretive leap are shocking: destructive pest entering beautiful flower to kill it equals human love-making... So my term “literal reading” is a bit misleading: it can include a response to symbolic meanings as well as literal. What I wanted to suggest is that the words mean what they mean, and they never stop meaning what they mean. (Meaning being, of course, a matter of general agreement. There is no such thing as a private language– Wittgenstein’s slogan bears repeating.) Even metaphors, as Donald Davidson insisted, mean what they say and not something else. This is a poem in which a worm enters, and destroys, a rose; and it’s only because worms and roses, and howling storms and crimson joys mean more than entomological, botanical and meteorological phenomena when we meet them in this arrangement that it’s a poem in which a great deal more happens.

I’d put that last notion a bit differently. The reason they “mean more than entomological, botanical and meteorological phenomena” is because they’ve been put in this most particular arrangement, an arrangement that summons their ordinary meanings and rubs them together to spark new meanings.

This in turn makes the notion of a “literal” reading problematic, but we don’t want to give in to “symbolic” readings too foolishly. Henry Staten explains:

I’m very friendly to calling the kind of reading we’re after a “literal” reading, even though it can be misleading (but all our terms can be misleading, starting with “obvious”). I think that the notion of “symbolism” is more misleading than “literal” is. Except when it’s used in the strictest sense, to mean something whose meaning stays more or less constant in each of its occurrences, the notion of the symbol has always seemed to me to introduce obscurity into the discussion of poems. […] You yourself say here that the fundamental conjunction of images “relies on the most basic visual associations to do its work.” Precisely. And I think it’s better not to call this “symbolism,” especially in cases like this one, where the technically non-literal meaning is so reliably linked to the literal meaning. I think it’s highly significant, though, that Blake has not depended on the worm-rose conjunction by itself to communicate the erotic meaning; he has thrown the words “bed” and “love” in there to make it obvious. To read a poem “literally” to me means not to depreciate the figurative meanings, but to

take it at its word and to avoid reading through it or behind it as much as possible.

Somewhat later on Staten invokes structuralism, which is where I got my intellectual start many years ago as an undergraduate at Johns Hopkins:

Another aspect of description with a high probability of consensual acceptance is the patterning in many poems of what the structuralists liked to call “binary oppositions.” There is a primary linguistic energy in such oppositions that requires a minimum of understanding of meaning to grasp, and which is very often obvious or close to obvious (beginning with such traditional topoi as light/darkness and the like); in many cases all we have to do is point to the pairs of words that have such a relation to one another and remind ourselves of what it is about them that makes their meanings clash—a clash that can often be felt right at the surface of the words’ meanings, independently of most subtleties of interpretation. By sharpening our sense of this clash, we make palpable the “force” or the “dynamics” of the juxtaposition of these words in this context. Blake’s poem fairly explodes out of the juxtaposition of these two words: worm, flower.

Later in that same passage he has a parenthetical remark where he talks of howling-storm-worm and crimson-joy-rose, a nice opposition and a nice compact statement of the semantic gamesmanship operating in the poem.

What kind of thing is a crimson-joy-rose? A crimson rose, yes. But a joyous rose, or crimson joy? They don’t make sense. I want to take a digression through that kind of nonsense before continuing on with Attridge and Staten.

Coiled Alizarin

One of the best-known sentences in modern intellectual life appeared in Noam Chomsky’s 1957 Syntactic Structures. He introduced the sentence “Colorless green ideas sleep furiously” to illustrate the independence of grammar from syntax. Chomsky contrasted that example with a second, “Furiously sleep ideas green colorless,” asserting that both are nonsense “but any speaker of English will recognize that only the former is grammatical” (p. 15).

But this nonsense was not just any nonsense. It was not like Shakespeare’s Dogberry contradicting himself by asserting that “for the watch to babble and to talk is most tolerable, and not to be endured” (Much Ado About Nothing, III.iii.35-36). Nor was it like Louis Carroll telling us that “’Twas brillig and the slithey toves/ did gyre and gimbal in the wabe.” Those are just nonsense words. At its syntactic core Chomsky’s nonsense was a pair of what philosophers call category mistakes.

It makes no sense to assert of ideas that they have any color whatsoever. Ideas can be new or old, complex or simple, limited in scope or broadly applicable, and so forth, but they cannot be black, white, red, blue, green, or any other color. Ideas aren’t the sorts of things that can have color. That is one category mistake. The second lies in attempting to have ideas sleep. An idea can awaken someone’s interest or it can be so boring that it puts that person to sleep, but the idea itself cannot sleep—though it might lay dormant in someone’s soul, which is very close to sleeping, but not quite. These two offenses are then conjoined with other sins—the contradiction between “colorless” and “green” and the quasi-categorical mismatch between “sleep’ and “furiously”—to complete the impression of nonsense.

And a very convincing impression it is. It really doesn’t make sense. That property, however, turns out to depend on context. John Hollander, the poet, brought sense to Chomsky’s line by providing it with two lines of context and a title, “Coiled Alizarin.” The title itself presages the poem’s method, for alizarin is a red dye and, as such, cannot sensibly be said to coil—though an artist might be put in mind of coils of Alizarin Crimson paint, which is very common, squeezed onto the palette. Here is the poem, which is dedicated to Noam Chomsky:

Curiously deep, the slumber of crimson thoughts:

While breathless, in stodgy viridian,

Colorless green ideas sleep furiously.

The poem works by establishing a context in which Chomsky’s category mistakes become natural; category error thus become the poem’s modus vivendi. And that, we will see in the next section, seems to be what Attridge and Staten are saying about The Sick Rose. But before we get to that I want to elaborate on the notion of a category mistake. For it implies that language has categories and that those categories impose restrictions on the kinds of assertions we can sensibly make.

Back in graduate school I did some moonlighting in the Linguistics Department, where I had become a member of David Hays’s research group. One of the things we discussed was the conceptual structure known as the Great Chain of Being. But we were interested in it, not so much as a conceptual trope in Western thought, but as a basic organizing principle of natural language semantics.

Consider the table below (adapted from Bloom and Hays [5]). The object in the left-most column is considered to consist of an assignment between the object immediately above it and the object to its right. The general idea is that language has a built-in structure of ontological categories (so called natural kinds) organized by a pattern of assignment relations.

The left-hand column is recognizably the Great Chain and, at this level of argument, rows 1 and 2 are relatively unproblematic. But what about those souls in lines 3, 4, and 5?

The first thing to be said is that they are there in Plato and Aristotle and continue in the Western intellectual tradition into the seventeenth century – we’re now talking of the Great Chain as an intellectual trope. The second thing to be said is that they are obviously essences, the conceptual purpose of which is to account for the behavior of the beings possessing them – we’ve switched back to basic semantic structure. If a plant grows, then, it is the vegetative soul that is causing the growing. Things do not have souls and so cannot grow; plants do have them and so can. Similarly, if a dog is sniffing the air and following a scent, it is the sensitive soul that is the source of agency. And so it is for the natural person, speech, and the rational soul [6].

Beyond this, I note that this ontological structure is explicitly a structure of conceptual inheritance, but it is “orthogonal” to the inheritance in the ISA paradigms of AI knowledge representation – ISA, that is to say, “is a” as in “a dog is a mammal”. In that sort of inheritance, dogs inherit all the properties and capacities of animals (for a dog is an animal) and have, in addition, the unique properties and capacities of dogs, rather than cats or gerbils. In like fashion, collies inherit all the properties and capacities of dogs, and lassie all the etc. of collies.

The assignment relation (ASN in the diagram) carries inheritance in the opposite direction. The animal node inherits the properties and capacities of plants (life and growth) via assignment just as the plant node inherits the properties and capacities of things via assignment. It is this inheritance via assignment that gives the Great Chain its conceptual punch. And it is inheritance that makes these structures “living” conceptual organisms rather than mere “bins”. But a proper understanding of this is more than I can convey in prose, even a great deal of prose. One must work with such models in order to appreciate their properties.

Categories in The Sick Rose

Thus when Blake begins the poem by addressing the rose directly he is violating the conceptual structure of the Great Chain. Roses are plants and plants do not talk nor can they understand language; only humans do. That opening gesture may be highly conventionalized, but no matter. It also jumps a level in the Great Chain. The next three lines (that is, ll. 2-4), as we’ve already remarked, aren’t obviously addressed to the rose; for all we know, they might be addressed to the world at large. But “thy bed” in line 5 tells us that the rose is still being addressed. And it is being addressed in the final line as well: “they life destroy.” The whole poem, then, takes the form of an address to the rose.

Literally, it’s nonsense. Poetically, though, it is something else.

While Attridge and Staten never mention the Great Chain as such, its categorical discipline haunts much of what they say. Here we have two passages by Derek Attridge. The first:

Doesn’t the empirical approach in fact break down pretty quickly as we try to imagine that flying worm that takes wing in stormy nights? This seems to be something like a generic worm, rather than an identifiable creature; a living entity prone to arouse disgust and fear. (Worm phobias are not uncommon, and the reluctance to touch worms is, I imagine, quite deep-seated among a large number of people, at least in Western cultures.) So yes, give the empirical phenomena their due, but recognize that literal meaning doesn’t equate with phenomenality of that sort.

The notion of a flying worm isn’t nonsense in the way that that of sleeping idea is, but ordinary worms do not fly. This worm thus does not belong to the natural order of things.

DA raises more category issues:

Your last parenthetical point is very interesting, and it raises the question–which isn’t easily answered – of which terms in the poem are literal. Rose we can assume is literal […] Sick is not as commonly used of plants as of animals, but it’s pretty close to being literal. Invisible can’t be entirely literal, but we can understand it as “unseen”–though not entirely, as “invisible” carries a whiff of the magical, or the perhaps the eldritch. (Is this not an association that would be generally understood?) Then flies: if the worm literally flies it’s not a literally a worm; if it’s literally a worm, it can’t literally fly. […] Night and howling storm are presumably literal; but then found out suggests a willed activity hard to ascribe to a worm, however noxious, and as you point out bed is metaphorical. So, in fact, is most of the second stanza, until the last line: this is part of the poem’s movement towards the adult and the erotic. Our obvious, minimal, “literal” reading has no option but to proceed figuratively in order to bring the meaning back to the empirical scene of flower-killing pest: “love,” for instance, has to be understood as a metaphor for the pest’s physically intimate connection with the flower, as well as its need for what the flower can offer it. But it doesn’t stop meaning love as we understand it in the human domain. I believe it’s this mental stretching and wrenching that makes reading the poem– no matter how many times we do it – such a powerful experience.

My point is simply that “this mental stretching and wrenching” is tugging at and violating the boundaries of the system of categories that organizes language. Poetry, of course, does that. The trick is to do so in a way that is ultimately orderly.

On that howling storm, I agree that it is literal, but I can’ t help but wonder if that has been transferred from noises – howls – omitted by animals or humans. I note that there is a slight difference in usage as between animals and humans on the one hand and storms and such on the other. When used as a noun humans and animals can emit multiple howls, but “the howl of the wind” will always be singular.

Back to Staten:

The case of “howling storm” is a more interesting divergence between us, one that brings out sharply the way that singularity comes into even the weakest reading—but at what I would consider a pretty “high” level. I don’t look at our divergence as something that has to be resolved, but as something that opens doors to further dimensions of reading, working systematically on the basis of the obvious that we do agree on. We agree that “howling storm” belongs to the same rhetorical register as “flies” and, to some degree, “invisible,” as predicates that don’t fit your empirical garden-variety worm; and we agree that, nevertheless, the garden variety worm has to underpin this rhetorical flight.

That is to say, the garden variety worm is one that plays within the ordinary boundaries of common “garden variety” linguistic categories. HS continues:

“Invisible,” however, as you’ve noted, could mean simply that we can’t see the worm that is eating the rose (he’s presumably eating her from inside, and in the dark of night), and flying, while it isn’t a characteristic of worms, is a characteristic of many noxious bugs, so that “worm” starts to sound like a synonym for “bug,” “vermin.”

At this point I’m getting a bit antsy over the detailed literalness of this discussion – “eating her from the inside” – but with “vermin” we hit pay dirt. I’m thinking of the structuralist account of taboo offered by Edmund Leach in the 1960s [7] where creatures are taboo if the don’t fit a culture’s category system. Leach, however, isn’t talking about the Great Chain categories; he’s talking about how things and events are assigned a place in social relations. He’s talking about taboo. Why do the French eat horse meat but the English do not? It can’t be biological properties that are at issue; it must therefore have to do with social categories. Well, vermin is a general term for creatures that don’t fit the system of social categories.

Back to HS:

(And among the bugs, of course, there is the case of the wormish caterpillar who emerges suddenly as a moth or butterfly).

Yes. The mystery and wonder of the caterpillar and the butterfly is that of continuity of substance but change in form. HS:

It also seems unlikely that a real bug would fly at night (are there nocturnal bugs?); [Fireflies – BB] but, again, this can be read as a figure of its invisibility to the speaker, its “secretness.” But it’s hard to imagine a real bug of any kind that flies in a howling storm.

HS on the storm:

Whether or not it’s excessive, however, “howling storm” is clearly unnecessary at the level of meaning, in the sense that nothing is made of this supernatural dimension in the entire culminating stanza. The bedroom drama of “dark, secret love” (a phrase that, incidentally, strikes me as, along with “invisible worm,” the strongest in the entire poem, the very center of its power)

Yes.

doesn’t go well with a howling storm, and even worse is the fact that a howling storm would be furiously tearing petals off the rose even as the worm loved it to death (at which point the whole thing starts to sound ridiculous to me). Now, someone might reasonably object at this point that none of these problems arise if one doesn’t try to literalize the poem as I’m doing; but if there’s no literal scene to underpin the poem it dissolves into a shapeless goop of symbolism on which the interpreter is free to impose a form of her own devising.

Fiddle-faddle! I so object. I faced this kind of problem years ago with “Kubla Khan”. It’s a very visual poem, and seemingly very specific. But if you try to draw a map of Xanadu, you can’t do it.

What’s going on in The Sick Rose has the appearance of literal accessibility. But it isn’t. The problems Staten has been raising pretty much mean that any literal scene one chooses is going to have an arbitrary aspect to it. In what way is an arbitrarily imposed literal scene an improvement over free interpretation of symbolic goop? And, in any event, what’s to prevent the free interpreter from simply ignoring the literal scene? No such norms currently exist in the profession.

Beyond this, what could such a story possibly tell us about Blake’s techne in the poem? Does
Staten think that Blake had some such story in mind when he wrote the poem? I doubt it myself. I’d think he worked directly with the words themselves, their meanings, rhythm, and general sonic envelop.

In any event, we certainly aren’t dealing with symbolic goop. Blake is manipulating the category system of English; that’s his techne. The Great Chain gives us a conceptual focus for that category structure. It is by deploying things against the logic of that system that Blake elevates these concrete things to more abstract and general significance. Just which abstract significance this or that person wants to adopt, that seems to me a secondary matter. What is most important is simply that Blake drives the physical into the metaphysical.

Staten goes on worrying about the literal details, but finally arrives somewhere interesting:

It just now occurs to me to wonder if we should take the worm’s nocturnal flight in the storm to be, as I’ve always unreflectingly assumed, the flight by means of which he discovered the rose’s “bed of crimson joy,” or whether it’s mentioned simply by way of characterizing him as “the worm who is such a hellraiser that it’s nothing to him if there’s a howling storm” (and it’s worth noting at this point that I’ve all along been unreflectingly referring to the worm as male and the flower as female, which seems an obvious assumption, but of which today perhaps we should be a little cautious). If the latter, then the dark secret lovemaking need not be imagined as taking place in a storm; in favor of this reading is the frequentative “flies.” And now that “howling storm” is removed from the immediate scene it feels less disruptive, and I become able to hear it at just the same level of hyperbolic volume as “crimson joy,” which thus starts to feel like a suitable binary partner with it. (“Howling-storm-worm, meet crimson-joy-rose.”) But now the poem begins to feel too much like erotic melodrama! […]

And it seems to me that that’s what we’re dealing with, an opposition between howling-storm-worm and crimson-joy-rose. It only becomes a melodrama if we try to imagine, in some detail, the worm flying about and penetrating the flower. Why do that? It seems to me that the need to imagine such things is a product of choosing discursive interpretive writing as our intellectual tool. Maybe our colleagues in the cognitive sciences have – or could be motivated to elaborate – other conceptual tools along the lines that I’ve outlined in the previous section.

Note that while roses may perfectly well be and often are crimson, it is a category mistake to talk of “crimson joy” and “bed of joy” is questionable as naturalistic description. The same with “dark love”, though “secret love” is fine – which Staten will mention a bit later.

Let’s continue on with Staten:

The power of the poem rests in the way it is able to arouse so much of our response to the fatality of human eroticism while keeping this response tethered to the non-human (and now I start to feel that it’s going too far even to assign gendered pronouns to the worm and the rose).

And this tethering to the non-human is forcing category mistakes all over the place, starting with the first line in which the rose is addressed as through it were human.

Yet it remains the case, as you’ve convinced me, that there is a third semantic force-field operating in the non-empirical qualities of the worm. The whole thing is set vibrating with an overtone of the fabulous by these qualities; and “howling storm” is what insures the sounding of this additional note (which “invisible” and “flies” by themselves don’t quite reach).

Those vibrating non-empirical qualities, that’s what you get when you monkey with the language’s category system. That’s what makes this poem work. It’s not symbolic goop; it’s highly specific manipulation of categories and their affordances.

Let’s give Staten a final whirl:

There is, for instance, the salience of line seven: I agree entirely with you that this comes across as “the center of the poem’s power.” How does it achieve this salience? It is, for one thing, the only line with three major content words (Rose-sick; invisible-worm; flies-night; howling-storm; found-bed; crimson-joy; dark-secret-love; life-destroy).

Yes.

Rhythmically this produces in the second stanza the familiar AABA pattern, where a particular configuration is repeated, then varied from, and then returned to with a feeling of closure.

Yes.

The rhythm of the line itself is also exceptional, with three strong stresses instead of two slowing the pace: “secret”, although it doesn’t carry a beat, is too important semantically to be de-emphasized (as a chanted reading would do), and this produces a marked tension between metrical expectation and rhythmic realization.

Yes.

Semantically, too, the line stands out: not only does it have two adjectives qualifying a single noun, but they set up a little eddy in meaning. We would normally expect secret in the collocation dark secret to be a noun, so it’s a surprise to find that it’s an adjective; we then understand secret love easily enough, but what is a dark love? Then there is the echo in sound between two words in the final lines: love and life. Normally, these would belong together, and the sonic echo would reinforce the semantic contiguity; what the poem does, startlingly, is to turn them into one of those oppositions you mentioned–love becomes the enemy of life. More than that, actually: one of the other things we haven’t talked about is the force of the word destroy, which suggests much more, and much worse, than the mere ending of a life.

Not to mention the fact that it rhymes with “joy” two lines before.

At this point we’re almost at the end of their conversation about The Sick Rose, and we’re near to the end of mine as well. I would like to say more about Blake’s manipulation of the Great Chain category system, for that’s clearly where the action lies in the poem. But I fear that, at this time, I would only talk around it, without really getting at the heart of the matter.

Alas, the heart of THAT matter doesn’t lie within the interpretive practices of literary criticism as traditionally conceived, whether the interpretation is minimal or maximal. It’s not a matter of interpretation at all, but rather of understanding the mechanisms of mind and language in terms other than the discursive prose of literary criticism. It’s at this point that I think we really do have to recruit practitioners of the newer psychologies – cognitive, evolutionary, and neuro – to our aid.

References

[1] William Benzon, Attridge and Staten 1: What is Minimal Reading? New Savanna, blog post, December 21, 2015, http://new-savanna.blogspot.com/2015/12/attridge-and-staton-1-what-is-minimal.html

[2] Dereck Attridge and Henry Staten, “Reading for the Obvious: A Conversation”, World Picture 2, Autumn 2008, http://www.worldpicturejournal.com/WP_2/Attridge_Staten.html

[3] If you want a quick look at the criticism on this poem, Nelson Hilton, has posted a useful document that indicates the gist of 20 or so explications, http://www.english.uga.edu/nhilton/wblake/SONGS/39/39perry.bib.html

[4] T–V distinction, Wikipedia, accessed 23 December 2015, https://en.wikipedia.org/wiki/T–V_distinction#English_2

[5] Bloom, D. and D. G. Hays. Designation in English. In Anaphora in Discourse, edited by John V. Hinds. Edmonton, Alta.; Champaign, Ill.: Linguistic Research, 1978, pp. 1-68.

[6] I discuss assignment in more detail in an old working paper, Ontology in Knowledge Representation for CIM, Computer Integrated Manufacturing Program, Rensselaer Polytechnic Institute, Doc.# CIMMW85TRO34 (January 1985), https://www.academia.edu/238610/Ontology_in_Knowledge_Representation

[7] Edmund Leach, Anthropological Aspects of Language: Animal Categories and Verbal Abuse, in Reader in Comparative Religion, ed. by William A. Lessa and Evon Z. Vogt, New York: Harper and Row, 1965, pp. 206-20.