Trinity, 3 Malicks

Correlation is not casual um err causal?#@&!

Pursuing Lust in the Web of Language

Relational Networks as a Computational Semantic Model

This post is, in large part, revised and adapted from the “Computation” section of Literary Morphology.

In the middle and late 1960s researchers in cognitive science began investigating natural language semantics using computational techniques. One of the first, and most influential of these was a relational network model proposed by Ross Quillian (1968). His basic idea is an old one, that the meaning of a word resides in its relations with other words, and it is perhaps on that account that Umberto Eco adopted it as his semantic model in A Theory of Semiotics (1976, pp. 121 ff.).

Following Quillian’s path-breaking work a number of researchers developed their own relational network models. My teacher, the late David Hays, was one of them. This post provides a quick introduction to the network formalism and that takes a ‘blitz’ though the work I on did on Shakespeare’s Sonnet 129 using Hays’s network model. This work is reported in more technical detail in Benzon 1976, 1978, and 1981.

Networks, Path Tracing, and Pattern Matching

One of the primary objects of literary computation is simply to compose meaning, on the fly, from the succession of words in the text. This is not the only computational process – we must also consider the sound and rhythm of those words, we must consider style – but that is what I want to concentrate on here. Following a wide-spread convention, though not a universal one, let us think of lexemes (that is, words) as being organized in a relational (or cognitive or semantic) network (see Figure 1) where the nodes are concepts and the links between the nodes are relations between the concepts. Thus dog would be linked to animal through an ISA link (dog is an animal) and to eat by an AGENT (AGT) link, while in the same subnetwork liver is linked to eat by a PATIENT (PAT) link (dog eats liver), and so forth through all the types of concepts and relationships.

Figure 1: Relational Network of Lexemes

The network fragment depicted in Figure 1 is small and quite simple in form. The networks used in computer simulations of human thinking are quite large, but like that in Figure 1, they a composed of simple parts connected to one another in simple ways. In the work I did with David Hays we distinguished between sensorimotor schemas, on the one hand, and cognitive networks on the other (Benzon 1976, 1978, Hays 1976, 1981). Things like dog, pine tree, walk, salt, and so forth, things one can apprehend with the sense, would be characterized by sensorimotor schemas. Sensory schemas regulate how one sees, hears, smells, touches, and tastes things while motor schemas regulate one’s physical interaction with them.

These schemas, in turn, provide the definitional foundation for nodes in the cognitive network. Thus the dog, eat, and liver nodes in Figure 1 would be associated with corresponding sensorimotor schemas. We also had mechanisms for defining abstract concepts over patterns in cognitive networks (Hays 1976, Benzon and Hays 1990).

These days we would say that our account of semantics was embodied. Words are given meaning by perceptual and motor schemes or patterns defined over those schemas.or our immediate purposes, however, the important point is that two general classes of operations can be defined in networks, path tracing and pattern matching (Hays 1977). In Figure 2 we see some network in which a particular path is highlighted. It might, for example, might trace the opening of Shakespeare’s sonnet 129, “The expense of spirit in a waste of shame/Is lust in action …“ Each node would correspond to a word, or perhaps a grammatical relation, and the path in total corresponds to the pattern of network activation set up by the ‘incoming’ string of words.

Figure 2: Path Tracing

In Figure 3 two areas of the network are highlighted, each consisting of five nodes where the pattern of connection is the same in each network. Though these two subnetworks have different orientations and overall proportions in the 2D representation of the network, those are irrelevant; what matters is the pattern of connectivity, and that is the same in each network. Each consists of four nodes connected by five links in the same configuration. Hence the patterns match. Of course, pattern matching implies a second order network, or some other meta-structure, to recognize the match (Hays 1981).

Figure 3: Pattern Matching

Continuing with Shakespeare’s sonnet, one of those fragments might represent the basic lust sequence at the heart of the sonnet while the other embodies the simile that is introduced in lines seven and eight: “. . . a swallow'd bait/On purpose laid to make the taker mad.” Both of these actions involve a three part sequence and so the network fragments representing those sequences have the same form (see Figure 4).

Figure 4: Lust and Madness

Working in this way we could construct a relational network as the basis for performing the sonnet’s meaning. The meaning of the poem would then arise in the interaction between the text and such a network, where the network is implemented in someone’s brain. Think of that meaning is an ongoing and evolving pattern of resonance in millions upon millions of neurons. Such networks, of course, may differ from one individual to another and even where highly similar, may have different emotional resonance.

At Rest

Jerk Avoidance Drill

Preview: The Key to the Treasure IS the Treasure, A Program for Literary Studies in the Current Era

Here is a brief description of a program for literary studies that has been forming in my mind over, say, the last year or so. This conceptualization is very much a work in progress.

I surmise that, in the future, literary studies will be organized around four foci: 1) description, 2) the newer psychologies, 3) object-oriented ontology, and 4) digital humanities. Do I KNOW this will happen? Of course not. I surmise that it is possible. I also think it would be a good thing, indeed.

1) Description: We need to develop richer descriptions of the texts we study. I’ve blogged about this hear and there, and I note that some folks at Arcade seem to be thinking about these lines. But mostly what I’ve been doing is working at honing my descriptive skills, with texts and with films. I’m working on an explicit justification of description. For now, see my postscript about a handbook for Heart of Darkness.

The primary texts constitute the treasure we study. Full and accurate descriptions of those texts are the key to that treasure. Everything else is built on those descriptions.

2) Newer Psychologies: By which I mean cognitive science, neuropsychology, and evolutionary psychology. I take this to be background knowledge, knowledge about the human mind. The first half of Brian Boyd’s On the Evolution of Stories is exemplary in this regard. Supplement it with material on the computational aspect of cognition and you have a synoptic précis of early 21st Century psychology.

3) Object-Oriented Ontology: Provides a philosophical framework in which aesthetic and ethical issues may be addressed. It’s only in the past two or three months that I’ve been pursuing this line of thought, and so I’ve relatively little to say about it. I note that OOO is grounded in the Continental philosophical tradition, which is where my own roots are. But I’m certainly not averse to contributions from the Anglo-American analytical tradition. But I know even less about that than I do about OOO.

4) Digital Humanities: The digital humanities can provide: 1) New tools for the description of literary phenomena, whether, e.g. through statistical studies of individual texts, or though ‘distant reading’ (in Franco Moretti’s phrase) of large numbers of texts over large periods of time and geographical extent. 2) Tools for richly presenting and annotating primary texts, organizing handbook material, and integrating them with the larger literature around those texts (again, see my postscript to the Heart of Darkness posts). 3) Ultimately the digital humanities may provide explanatory models, simulations, of how the mind encounters texts.* In this function the digital humanities would converge with the newer psychologies.

I do not see robust simulation of literary processes as something that is likely to happen soon. Nor do I have any idea when it is likely to happen. Thirty-five years ago I thought that it would take, perhaps, twenty years to reach that point. I was wrong.

Thinking Novels vs. the Memebots

I was looking through the archives at The Valve and came across an old book event, the one Lawrence LaRiviere White organized around Nancy Armstrong’s How Novels Think. Though I was ‘on staff’ at the time, I didn’t participate in that event. Now that I’m a fellow traveler on the Object-Oriented Ontology train, the title alone intrigues me, as apparently it intrigued White.

OOO sees the universe as comprised of irreducible objects, and ascribes agency to all objects. Therefore, under OOO, novels too must have agency. And they exercise that agency through, what else? thinking. Perhaps feeling and desiring as well, but thinking is sufficient for this post.

It turns out, alas, that White was disappointed in what Armstrong had to say about just how it was that novels DO IN FACT think. And he had so wanted her to come through on that point. Dan Green finds her wanting on that score as well, though he likes her readings of various novels and her overall account of the rise of the novel. (I’ve not read through the other contributions to the event.)

What interests me is simply the fact that the issue has been broached: literary texts think. By way of comparison, here we have Eileen Joy (via Tim Morton):

. . . following the thought of Jane Bennett, I want to say something like . . . texts are objects that possess vibrant materiality; they are “quasi forces” that possess something like “tendencies of their own.” They possess thing-power, and as much as they are able, they strive, in the words of Spinoza, to “persist in existing.” Texts are, in some sense, alive, while at the same time they are, even while produced by humans, utterly inhuman.

Alive, and so, perhaps, they can think, while at the same time they are “utterly inhuman.”

So there we are, living texts and thinking novels.

To which I contrast the homomuncular memebots loosed on the world by Richard Dawkins. As Dawkins tells the story (in The Selfish Gene) he was looking for ‘replicators’ other than the gene and hit upon the idea that culture too is an evolutionary process, one ‘driven’ by memes, the cultural replicator. In time the term went viral and now we have it, all over the place, memes memes memes and more memes.

Peekaboo

The are the good old days

Angular Handstyle

HD Postscript: Toward a Heart of Darkness Handbook

Now for something different.

For some time now I’ve been thinking about the need for ‘handbook level’ information about texts and, correlatively, a professional discipline organized to produce such handbooks. The purpose of this post is to look back over what I’ve written about Heart of Darkness and suggest some things that should be considered for inclusion in such a handbook.

And suggest is I all can do. It’s not merely that literary studies doesn’t have such handbooks (though casebooks serve such a role to a limited extent) but that, if we did, it would not be possible for any one person to enter something into the handbook. All material would have to be vetted in some way. For, when I talk of ‘handbook level’ information I’m talking about consensus among expert specialists: THIS is what we know about, in this case, Heart of Darkness.

Literary studies is not such a discipline. Moreover, if there is a consensus on anything, it is that the discipline cannot, in principle, be organized in such a way because our subject does not permit it. Yes, there is historical information about authors and texts, that admits of consensus. But little else does, or at any rate, little else of significance.

Handbook as Description of a Sharable Promptuary

For literary studies (and kindred disciplines such as film studies and cultural studies) has been driven by the search for meaning, and meaning just doesn’t seem to admit of cumulative knowledge. In the informal extreme, each reader makes his or her own meaning for any given text. That is in fact how the profession has conducted its business, the understanding being that each one is entitled to “roll their own” reading so long as they can come up with a coherent rationale for it.

The last thirty years has seen a wild proliferation of rubrics under which coherent rationales can be constructed. They’re called readings. And, again I’m speaking informally, the meanings set forth in such readings are taken as explanations of the text:

Q: Why does the text exhibit this or that feature?
A: So it can express this or that meaning.

Beyond this, one family of critical approaches, reader response theory, has emerged that explicitly asserts that each reader has their own reading.

Here’s what one reader response critic, Normal Holland, has to say about the role of texts in readers experience (Five Readers Reading, pp. 286-287):

Every reader has available to him what the writer created—the words-on-the-page, that is, the promptuary (a store of structured language) from which he can build an experience. To be sure, the promptuary includes constraints on how one can put its contents together, but these constraints do not coerce anyone. One is always free to go to the extreme of total delusion . . . But most reading is not solipsistic. . . . [Readers] work with what is publicly available—the words-on-the-page as a readiness of structured information. They make from that sharable promptuary a completely original and private experience.

That’s what I want to describe in the handbook, the sharable promptuary.

Weave

Ontology, Woven in the Text

We’re heading toward Patterson, William Carlos Williams’ modern epic lyric. The poem’s guiding conceit is stated early in the text:

Yet there is
no return: rolling up out of chaos,
a nine months’ wonder, the city,
the man, an identity—it can’t be
otherwise—an
interpenetration, both ways.

With that identity, man=city, you can see that ontology is in play. But we are, specifically, headed toward Book V, which is an extended discourse on the Unicorn Tapestries in The Cloisters museum in New York City. They depict the hunting of a unicorn. In Medieval sacred symbology that was a vehicle for the capture and crucifixion of Jesus of Nazareth, aka the Christ. Williams, with full knowledge, of course, treats it also as a figure for the defloration of the bride.

So there we have it, at trinity: 1) a story of an imaginary beast, 2) a story of a quasi-historical man-god, and 3) a story enacted many times by many people in many places, a common story of man and woman. As he weaves these stories through one another, Williams is conscious, always, that he is dealing with a tapestry made of threads, in animate material stuff. So we have three stories woven into one physical tapestry woven into one (meta)physical poem.

Let’s approach gently.

First, let us review, the Great Chain of Being, classically. Then remind ourselves about Latour Litanies by finding something like them in Robert Burton’s Anatomy of Melancholy. Then through language to Williams.

The Great Chain of Being as Conceptual Structure

This is a bit of mildly technical cognitive science from the archives. It's relevant to a current run of posts about ontology.

Lakoff & Turner (1989) have used a cognitive account of the Great Chain of Being in the analysis of proverbs. I’d like to elaborate on the commonsense structure they call The Nature of Things and suggest an extension.

The Assignment Relation

Consider the table below (adapted from Bloom and Hays, 1978), which moves up the Great Chain. 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 ontological structure is a pattern of assignment relations.

1)	substance
2)	thing	form
3)	living being (plant)	vegetative soul
4)	animal	sensitive soul
5)	natural person	rational soul
6)	social person	role

The left-hand column is recognizably the Great Chain and, at this level of argument, rows 1, 2, and perhaps 6 are relatively unproblematic. But what about those souls in lines 3, 4, and 5?

[In the annoying manner of math texts, I leave it as an exercise for the reader to determine whether or not, and if so, just how, juristic persons (corporations) are to be incorporated into this ontology, perhaps as types of social person. I note that this particular issue is under dispute in American politics.]

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. 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. If a plant grows, then, by [the metaphor mapping] EVENTS ARE ACTIONS, 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 which is the source of agency. And so it is for the natural person, speech, and the rational soul.

What it is