Saturday, January 14, 2006
Maps, Iconic and Abstract
Though I am rather interested in maps, I admit to being a bit puzzled by the maps chapter of Moretti’s book. Part of my problem is that I am comfortable, not only with the iconic maps of physical space that Moretti uses, but with the notion of a cognitive map as used by psychologists and neuroscientists, and with abstract maps of conceptual spaces used by cognitive and computer scientists. I have given considerable thought to the use of those abstract maps of conceptual spaces in the study of (literary) textual semantics. Moretti is not doing that sort of thing in this chapter, which is fine. My problem is that I don’t quite see how to relate what he has done with iconic maps to thinking that I - and others - have done in terms of abstract conceptual maps.
I say this, not as a criticism of his chapter, but as a statement of my difficulty in engaging with that chapter. This document is a set of comments around and about that, but arriving at no particular conclusion. It is an essay into possible sites of exploration.
Let me start with the cognitive map, which is a neuro-cognitive structure through which an animal "represents" (the word is problematic in certain intellectual domains in the cognitive sciences) the external world (see URLs below). The nature of this representation is of considerable interest to psychologists and has attracted a good deal of attention. This work is closely related to the mechanisms by which animals navigate from one place to another. I have, at one time or another, read some of this literature, but I am not expert in it. The following remarks are meant to be no more than reasonable and indicative.
Animals are creatures of habit. They live in relatively fixed territories where various places serve various needs. There is a privileged place that serves as home base and there are various other places where the animal finds food and water, or danger. Many animals migrate with the seasons, sometimes over considerable distances. Finally, we should note that most animals are considerably more sensitive to odors than we are. They smell their way about the world. How is all this represented in the animal’s brain?
There is no reason to believe that cognitive maps are like iconic maps except, rather than being inscribed in the dirt, or on a rock, or imprinted on paper, they are somehow inscribed in neural tissue. They seem to be more like lists of significant places intertwined with bearings and headings between one place and another. The vital significance of these places is part and parcel of the map; the "map" is not a neutral spatial substrate to which vital significance is later attached. The space of cognitive maps is not merely about physical position; it is about needs and satisfiers, vantage points and opportunities for action.
Given the importance of local and distant geography to animals, it is clear that the neural systems that map that geography must be sophisticated and complex (and have evolved over 100s of millions of years). Other than the system for interpersonal relationships, these may be the most sophisticated neural systems we have. The myth narratives Levi-Strauss has analyzed in, e.g. The Raw and the Cooked, would appear to intertwine these two neuro-cognitive systems, the geographical and the social.
Vital significance seems central to Moretti’s maps. It is not just that things happen in different places, but that different kinds of things happen at different kinds of places. And where people tell stories about the world around them, those stories will reflect the vital structure of that world.
One wonders how conscious were the writers of the geographic structure of their world. Some years ago Donald Norman asked graduate students living in a certain building - some of them for years - to sketch a map of their apartment (p. 139).* Many of them made a significant error in their sketch, even thought they were sitting in the apartment when they made the sketch. Yet, they had no trouble getting around in the apartment building. Their mental model was adequate to that task, but not to the task of making a sketch. Would the author’s of Moretti’s texts have been able to make accurate sketches of the territory they wrote about? We do not know. But being able to make such sketches is quite a different skill from being able to navigate through the territory. In drawing his maps Moretti may thus show us something about the geography of those worlds that the writers themselves did not (quite) know.
They need not have been conscious of the lay of the land - as it appears to map makers - in order for it to be implicit in their narratives. If activities are arrayed in the world in a certain spatial pattern, and the narratives faithfully represent those activities, then that pattern will be implicit in the narrative without the author ever having given explicit thought to the structure of that array. If the usage of geography changes over time, then stories told over the same time span will reflect those changes. Again, without the author’s being explicitly aware of this.
This is at the center of Moretti’s arguments in the maps chapter. And perhaps that is what so interests him, that the narratives accurately reflect things of which the authors were not, need not have been, explicitly aware. Those patterns turn up in the narrative because that is how the world works. The world creeps into the mind in ways that exceed our explicit grasp.
Because our lives are lived in geography, many of our narratives include journeys within them or take the form of a journey. Some of those narratives even include maps as part of the text - I’m thinking of e.g. Treasure Island, King Solomon’s Mines, or, rather more recently, Perdido Street Station. And then we have those squiggles in Tristram Shandy which map, not the places of Tristram’s or Toby’s or Walter’s life, but the digressive mode of the telling.
Those squiggles are more abstract in their import. This leads me, in Shandian fashion, to ring structures, which are about the mode of telling. In the small, we have a rhetorical figure called chiasmus. For example, "well knows knows well" in Shakespeare’s Sonnet 129: "This the world well knows, that none knows well . . . . " But such inversions appear in longer texts. (And they are an explicit part of the arsenal of melodic variation and development in baroque counterpoint.) A narrative will unfold through a series of steps to a mid-point and then trace its way back through the same series of steps, but in reverse, thus:
Mary Douglas has been investigating ring structure in books of the Old Testament, while I have found it in Osamu Tezuka’s graphic novel, Metropolis. The fact that rings are symmetrical about a mid-point suggests to me that they may ultimately depend on the cognitive structures we use for use for spatial navigation. If you travel from location A to N and then back you will pass the same landmarks on each half of the journey, but in reverse order. If things of interest and consequence happen to you along the way, both going and coming, then you may have a narrative that is interesting as well.
But the ring structure of Tezuka’s narrative cannot be explained through fidelity to an external geography. For it is not physical movement in space that takes the form of a ring, nor the deployment of locations in space. Tezuka’s narrative is more like a picture within a frame3 within a frame2 within a frame1, where you start at the outermost frame, move to the next, the next, then the picture, and then you turn around and go back out through the series of frames. We can think of frame1 as the public world at large; frame2 is an underground world of evil conspirators and robot slaves; while frame3 is a children’s world, of home, schoolhouse, and playground. (The picture itself has starting revelations.) Beyond the fact that frame2 is underground, there is no explicit sense the geographic relationships between these realms nor even of different locations within each of the realms.
There is no particular reason to believe that Tezuka achieved this effect through conscious deliberation. The ring structure I’ve outlined is not at all obvious; uncovering it takes a bit of analytical work. Whatever forces are at work in this ring seem to me more mysterious than then material forces working on the English and German geography arrayed in the narratives Moretti examines. In those narratives the mind mirrors the external world. In Tezuka it is not clear that the mind is mirroring anything at all.
Note that these ring-form narrative are different from the rings Moretti talks about. In his discussion of village narratives he illustrates physical rings in the geography that is the setting for narratives. Certain things happen in the center, other things happen closer to the periphery. But nowhere does he talk about individual narratives having episodes arrayed in a cyclic fashion.
Thus the geographic ring and the narrative ring are different and independent phenomenon. Yet there is the possibility that the narrative ring ultimately depends on neural structures that arose for the purpose of navigating the physical world. The narrative ring is more abstract. Let us now consider a different kind of navigational abstraction.
The Method of Loci
[Here I simply quote from an old article of mine on Visual Thinking**]:
The locus classicus for any discussion of visual thinking is the method of loci, a technique for aiding memory invented by Greek rhetoricians and which, over a course of centuries, served as the starting point for a great deal of speculation and practical elaboration - an intellectual tradition which has been admirably examined by Frances Yates. The idea is simple. Choose some fairly elaborate building, a temple was usually suggested, and walk through it several times along a set path, memorizing what you see at various fixed points on the path. These points are the loci which are the key to the method. Once you have this path firmly in mind so that you can call it up at will, you are ready to use it as a memory aid. If, for example, you want to deliver a speech from memory, you conduct an imaginary walk through your temple. At the first locus you create a vivid image which is related to the first point in your speech and then you store that image at the locus. You repeat the process for each successive point in the speech until all of the points have been stored away in the loci on the path through the temple. Then, when you give your speech you simply start off on the imaginary path, retrieving your ideas from each locus in turn. The technique could also be used for memorizing a speech word-for-word. In this case, instead of storing ideas at a loci, one stored individual words.
The method of loci has become a central part of our memory improvement lore. Further, the effectiveness of this technique has been verified in psychological laboratories. According to Ulric Neisser, a cognitive psychologist, it even works for people who deny that they have mental images.
What is interesting about this technique is that it involves the deliberate creation of a cognitive model that is affectively and motivationally neutral. One memorizes a particular physical space simply as a set of loci and paths between them. The loci have no intrinsic significance. Rather, they are mental "pigeon holes" in which one can place things that do have significance. What kind of "things"? Anything you can imagine.
Notice, in particular, how there is no necessary relationship between the topographic structure of the loci and the logical structure of whatever is committed to memory through those loci. If you were to listen to someone deliver a speech which they had committed to memory using the method of loci, there would be nothing in the speech that would betray the building used to establish the loci and the path used to traverse them. It would, in principle, be possible for two people to deliver pretty much the same speech, while using different buildings as their memory model.
That, in general, is the problem we face in trying to figure out how the mind deals with literary texts. However visible the texts, the mind’s mechanisms are hidden. Moretti’s maps tell us something of how the mind finds the world. But just how is it that the world makes its way into the mind there to be transformed into texts? That process remains invisible.
* Norman, D. A. (1973) Memory, Knowledge, and the Answering of Questions. Contemporary Issues in Cognitive Psychology: The Loyola Symposium. R. L. Solso, ed. New York, John Wiley & Sons: 135-165.
** William Benzon (1990) "Visual Thinking," in Allen Kent and James G. Williams, Eds. Encyclopedia of Computer Science and Technology. Volume 23, Supplement 8. New York; Basel: Marcel Dekker, Inc.: 411-427.
Cognitive map URLs:
A classic paper from 1948, which I’ve not read myself:
A classic and very influential book, which I’ve read in précis form in Brain and Behavioral Science in 1979:
Method of Loci, Wikipedia:
Just thought I’d add two papers arguing against the concept of a “cognitive map.”
http://pigeonrat.psych.ucla.edu/200C/Bennett Do Animals Have Cog Maps.pdf
Also, I’m not sure how relevant it is to Moretti’s claims, but research by Gordon Bower and others has shown that represented physical distance (e.g., between two rooms in a house) is reflected in reading times, such that reading a sentence about two objects that are physically distant from each other (on opposite sides of the house, for example) takes longer than reading about ones close to each other (e.g., in the same room). I’m don’t know if anyone’s looked at this on larger scales (maps of cities or countries), or if anyone’s looked at production as opposed to comprehension.
Thanks, Chris. The two papers look most interesting. I’ve only glanced at them, but I’m sympathetic to them. The notion of a “cognitive map” is, indeed, a vague one. The Bower result is interesting and seems consistent with the by-now classical work on mental rotation.
A question I’ve been interested for quite a while now is if there’s any way of combining the cognitive scientific concept of a cognitive map with Jameson’s very impressionistic and yet suggestive appropriation. People who use the latter tend to nod at Kevin Lynch, and those working on the former seem to be exasperated at the prevalence and fuzziness of the Jamesonian legacy. I don’t think it has to be this way, though.
I’m afraid I can’t help you Jonathan. I read Postmodernism some years ago—in one of my recurring attempts to give (some verson of) Marxism a try—and have never had any occasion to cite the book—through, upon inspecting the book, I notice that I have a good number of marginal notes. I certainly didn’t see anything in that book that read like he’d appropriated anything from cognitive science. Meanwhile, Joseph Carroll, the Darwinian literary critic, has been talking of cognitive mapping (as the adaptive function of literature). I’m pretty sure he didn’t get the idea from Jameson, nor do I think he’s picked it up from the cognitive scientists either.
As a term “cognitive map” is pretty general. To give it much meaning you pretty much have to cite specific bodies of work.
Any attempt at a rapprochmont between folks like Jameson and cognitive science would have to face issues of conceptual style and of granularity of analysis and interest. As a point of information, I note that one of my fellow graduate students, Mary White, did a dissertation in which she did a fairly detailed cognitive analoysis of the world view of a millenarian commune. I suppose George Lakoff’s recent trips into political conceptualizing (not to mention the accounts of Western philosophy in Philosophy in the Flesh) might fit in here. For my taste, that work pushes conceptual metaphor theory past the breaking point.
Jameson’s use of the term comes from Kevin Lynch’s The Image of the City and first appears, as far as I know, in a brief essay in Marxism Beyond Marxism. Lynch was not a cognitive scientist but was writing about how people orient themselves in cities.
If there are, as the papers Chris cites suggest, inherent tendencies in the way space is imagined in narrative, this would be a good candidate for an atomistic unit of cultural (narrative) transmission, or at least potential constraint that shapes variation.
This continues to be a very interesting discussion.
Thus the geographic ring and the narrative ring are different and independent phenomenon. Yet there is the possibility that the narrative ring ultimately depends on neural structures that arose for the purpose of navigating the physical world. The narrative ring is more abstract.
I have done a bit of work applying some of the principles that are used in cartography (in re: spatial representation) to the understanding of historical (temporal) representation. Of particular interest has been a process that cartographers call “generalization”—an unfortunate term because it seems so, well, general. “Generalization” in cartography is process of reducing a complex and detailed map to a simpler map, and it involves all sorts of enumerable practices: deletion, outline simplification, symbolization, feature displacement, etc. (Cartographers became especially conscious of these practices when they first tried to get computers to do generalization; it turns out to be harder than you’d think.) You can draw very close parallels between all of these spatial practices on the one hand, and practices that occur in temporal (historical) representation on the other, and I would include literary narratives in the general category of temporal representations.
You’ll find some of my thinking on these topics in:
“Systematic generization, historical fate, and the species problem”
This paper is a bit stiff, but interested readers can certainly pick and choose from parts they find of value. The page above includes a link to the pdf version which includes the necessary illustrations.
Thanks for the pointer, Bob.
Here’s a link to an article in Style that reviews the literature on ring composition—taking a bemused stance toward it, I think—I’ve not yet finished reading it:
A great article--had the pleasure of studying narrative theory with Jim Paxson.