In a series recent blog posts, here, here and here, I have explored John Perry and David Israel's notion of an information architecture given in their paper Information and Architecture (pdf). In particular I have been exploring their notions of a coincident architecture and a combinative architecture. A coincident information architecture is an information system in which the architectural relationship between two indicating facts or signals induces a relationship in their respective indicated contents. Thus, to take a canonical example, a machine that simultaneously measures the weight and height of a person, by the fact that it is built in such a way, induces a relationship between the indicated contents of each measurement, namely that the belong to the same person. In a combinative architecture, the architectural relationship between indicating facts or signals reflects rather than induces their respective indicated contents. Hence, to take a canonical example, medical documents belonging to the same patient will be put into an envelope together, this relationship of being in the same envelope being a reflection of the fact that the documents pertain to the same patient. See my blog posts and their paper for more details.

It strikes me however, that this notion of coincident and combinative information architectures may find an application in the analysis of ritual. A few quick (and possibly naive examples):

Coincident Architecture

In order to be a coincident architecture between signals or indicating facts, the relationship between their respective indicated contents must have been induced by the architectural relationship between the signals or indicating facts. For example in some kinds of magic the ritual paraphernalia may be manipulated in some manner for the purpose of bringing about a desired effect upon the persons and object indicated, such as when a curse tablet is placed in a grave. Rites of separation often include acts of cutting, which might be thought of as inducing or causing the separation between the things or people that they represent.

These examples require much more careful exegesis. Instead I would take a (hopefully) more straightforward example of a Roman Catholic priest dressed in his priestly robes giving holy communion to the congregation. We might identify two contents:

1. The fact that the man is wearing the dress of a Roman Catholic priest carries the information that he is a priest of the Roman Catholic Church.

2. The fact that the man is giving holy communion the congregation indicates that he has or has assumed the ritual authority to perform that particular ritual act.

The underlined portions we call the indicating facts or signals, and the italicized portions the indicated contents.

We can connect these two contents by observing the (hypothetical) connecting fact that the man who is wearing the dress of a priest is the same man who is giving communion to the congregation, and therefore the man who is the Roman Catholic priest has or has assumed the ritual authority to perform that act. The induced relationship is that of identity.

Inducing an identity in this way may seem trivial, but it seems to me that such relationships are integral to ritual power and to the establishment of authority. In this case, the relationship between Church authority and holy communion and ritual authority.

Combinative Architecture

In a combinative architecture, the relationship between indicating facts or signals reflects existing relationships. The image in a mirror reflects the visage of the person in front of it. The crown on a king's head reflects the position he occupies. The fact that the priests stand before the congregation with their sacred paraphernalia reflects the positions they occupy as intermediaries between the congregation and God. The order in which warriors are called may signal the honor that their leader accords them. The boy who survives hardship to become a man demonstrates both his courage and resilience. All of these require more careful examination, and a legitimate claim can be made that these all play roles in inducing exactly what they reflect, something which should not be taken as abnormal. On the contrary, combinative and coincident architectures are frequently coupled.

The big-man who slaughters a thousand pigs for a feast demonstrates that he is a bigger man than the man who could only slaughter five-hundred. One cannot deny that this slaughter contributes to his status. But here I want to point to the material and economic constraints of the achievement itself. The fact that the first big-man was capable of slaughtering a thousand pigs for a feast is a reflection of his wealth and power. The fact the second big man was capable of slaughtering fiver-hundred pigs for a feast is a reflection of his wealth and power. The fact that the first big man was capable of slaughtering more pigs than the second is a reflection of the difference in wealth and power between them. This is not much different than a person's lifting 500 lb cannot help but to
attest to her greater strength than a person's lifting only 150 lb.

Final Thoughts

This is only a crude entry-point, more suggestive than substantive. My anthropology books are packed away; I no longer have Roy Rappaport's "Ritual and Religion in the Making of Humanity" on hand, and I'm afraid I do not have much time to engage the anthropological literature on ritual. A shame. I welcome this community's suggestions, or contributions to such an analysis, especially those like John McCreery who have spent so much time studying ritual.

I have sometimes found that maintaining the distinction is trickier than one might imagine. What Perry and Israel mean by induce is not always clear. Magic that is meant to cause an effect might be said to meant to induce an effect, but I am not sure that Perry and Israel would accept this interpretation. Also, Perry and Israel are very much interested in truth, facts, and such. Information is true content, content being a more general thing, something with which I suspect Rappaport may have agreed. But truth has an uneasy place not only in anthropology in general, but in the interpretation of ritual and magic in particular. Yet, I think that a fairly straightforward accommodation of the two views can be found, at least under fixed interpretation. More on that perhaps at another time.


REFERENCES

Israel, David, and John Perry. Information and Architecture. In Proceedings of the Second Conference on Situation Theory and Its Applications, edited by Jon Barwise, Jean Mark Gawron, Gordon Plotkin,
and Syun Tutiya, 147-160. Vol. 2. Stanford, CA: Center for the Study of
Language (CSLI), 1992.

Views: 30

Comment

You need to be a member of Open Anthropology Cooperative to add comments!

Comment by Jacob Lee on September 16, 2010 at 8:49am
Hi Huon,

Is 'combinative architecture' a fancy way of saying that two things are connected to each other in the eye of the beholder? And coincident implies something more the relationship - result depends on the fact that you can't have one of the elements without the other?

Pretty close. A combinative architecture is not just a way of saying that two things are connected in the eye of the beholder. It is more general than that. A combinative architecture combines signals (indicating facts that provide clues about some state of affairs) in a way that reflects some relationship that exists between the facts that are being indicated. It may or may not be necessary that the signals be structured in that way.

We can differentiate between induced combinative architectures where the relationship between signals is induced by their indicated contents, like the reflections of two geese flying over a pond, and conventional combinative architectures where the relationship is conventional, like a bundle of records put into a labeled folder. In an induced combinative architecture, the relationship between signals is necessary, more or less, and whether or not a beholder of these signals infers the corresponding relationship between the contents is irrelevant. In a conventional combinative architecture, the relationship serves as a hint at the relationship between contents indicated by each signal.One might say that signals are being arranged in such a way that the relationship becomes what the eye beholds: it may suggest that a similar relationship exists between their respective contents. It will suggest this to those who are attuned to the necessary constraints in play.
ex. 1.
The fact that the first and second sentences occur in the same blockquote above and is attributed to Huon Wardle is supposed to carry the information that Huon Wardle said both of those things, as he did. The fact that they occur in that order in the blockquote is supposed to indicate that he said those things in the same order, as he did. But to someone not attuned to the practice of blockquoting, that information might not have been conveyed.

ex. 2.
A frequent situation in user-interface design is the need to design an interface that reflects the architectural relations of some set of features but where such a relation in the user-interface design is not induced by that architecture. To take an example from Donald Norman’s book The Design Of Everyday Things, the control knobs of a stove are ideally arranged so that each knob simply maps (geometrically) to a corresponding burner on the stove: e. g.
B1 --------B2
| / (burners)
B3----B4
K1--------K2
| / (control knobs)
K3—-K4

where Bi maps onto Ki one-to-one. But such an arrangement of knobs is hardly required by the arrangement of burners.
A coincident architecture is pretty much what you described, except that the relevant point is that the relationship between signals causes a relationship between their indicated contents. A physical relationship, for example, such as the physical arrangement of two flying geese, the ground, and the sun that determines that the physical arrangement of geeses' shadows on the ground be just so.

The fact that the man is wearing the dress of a Roman Catholic priest carries the information that he [... is going to a fancy dress party.] [...has delusions of grandeur] [...is protesting against the Catholic Church's policy on contraception]

In my blog post here I did not make clear everything that I should have. A signal (indicating content) carries information that x, provided that the signal is factual (is an indicating fact) and that a number of background constraints and conditions are met. So, I was not saying that it could not indicate any of those things, or indeed only one thing. To correctly infer any of these things is going to require a whole host of background conditions being met.

In any case, anthropologists prefer rich situational descriptions rather than sparse ones.

I am crucially interested in how people (and other organisms) get on with the business of life. All organisms process environmental information, becoming attuned to natural flows of information in a variety of ways and at different levels of complexity. For humans, this crucially involves interpretation of situations you encounter in the world within a web of (mostly shared) meanings, and in which social or institutional facts are highly pertinent. Signals, by which I mean any discernible feature of the world that one can use as the basis of inference, are found everywhere in human life, from artifacts, to gestures, to our speech and our art, but also in our constructed and managed and unmanaged 'natural' environments. It seems pertinent then to understand how information is put together in the natural and socio-cultural world. I brought ritual into it because I was inspired by Rappaport's analysis of ritual in terms of information. It seems to me that ritual performance often provides a semi-formalized structure or architecture of signals, and that these relationships between signals would bear this sort of analysis. Politically motivated performances can consciously or unconsciously manipulate relationships between signals in order to suggest particular inferences/interpretations. The relevance of these relationships seems particularly clear in the reinforcement and justification of exercises of power.

It is not clear that the first thing indicated induces the second - on the contrary everything exists in a tautological relationship. Roman Catholic Priests wear a certain dress/RC priests are allowed to officiate ritually - rituals are officiated over by people wearing the right gear. These are just stereotype facts distinguished from the field of possibilities then added to each other as if in a logical relationship. You have to wear the right clothes if you are going to officiate at a ritual might be the general point here. But, the right clothes are already tautologically tied to the meaning given to Roman Catholic Priest etc.

The stereotyped facts that you refer to are not instantiated in any individuals. These are exactly the kinds of background knowledge and constraints that must be operative in order that information be carried. But they do not carry information in of themselves, at least not in this context. Consider; if any Catholic knows all of this, then why not just imagine the ritual act? Why actually have someone perform it? Rappaport asked much the same question. It is the fact that it is some particular person who is wearing the clothes that works as the signal, not the abstract notion of someone wearing the robes of a priest, not the abstract notion of a priest giving communion. The ritual might not have been performed. Moreover, the stereotyped facts as you have called them may be contested.

Why might Charlemagne have been upset that "On Christmas Day, in 800 A.D., while knelt in prayer in Saint Peter's, the pope crowned Charlemagne emperor by placing a gold crown on his head"? Because in so doing, the pope both signaled his spiritual and political authority over Charlemagne and obligated Charlemagne's to defend the Church against its enemies. The situation is rich in informational significance. It was not just another recurring instance of a generally accepted ritual pattern, but an act that arguably had repercussions for centuries.

Many political ads are much the same, especially placards for county supervisors. We might describe some stereotyped facts about them, such as the inclusion of the name of the candidate and her or his picture. So why would a politician put their name and face on a stereotypical ad instead of someone else's? Because they want to be the relevant particular of the placard's indicated content.

Information is being carried because some grounding situation is supporting various stereotyped facts. I say, yeah, nuns look like X, oh wow, Sara looks like X, is she a nun? Could be. But, no, its Halloween, and the outfit is too cheap. Or, oh, George is sprinkling holy water on the congregation. I didn't know he was a priest! Information crucially involves both particulars and stereotyped facts (or as I like to call them, types).

Why do we need the 'architecture' metaphor apart from the fact that it sounds fancy and exciting?
A couple of comments. First, within the engineering, computational and informational sciences, calling something like this an architecture is pretty standard. Architecture may have once been a word reserved for buildings, but no longer. Therefore, it is arguably no longer a metaphor. It is perhaps another way to refer to its structure or organization.
As for the specific terminology, it is the terminology that John Perry and Davis Israel adopted in their paper. I am not especially fond of the terms combinative and coincident. I would prefer something more straightforward like, reflecting arch. and inducing arch.
Comment by Huon Wardle on September 15, 2010 at 3:57pm
n a combinative architecture, the architectural relationship between indicating facts or signals reflects rather than induces their respective indicated contents. Hence, to take a canonical example, medical documents belonging to the same patient will be put into an envelope together, this relationship of being in the same envelope being a reflection of the fact that the documents pertain to the same patient.

Is 'combinative architecture' a fancy way of saying that two things are connected to each other in the eye of the beholder? And coincident implies something more the relationship - result depends on the fact that you can't have one of the elements without the other?



1. The fact that the man is wearing the dress of a Roman Catholic priest carries the information that he is a priest of the Roman Catholic Church.

2. The fact that the man is giving holy communion the congregation indicates that he has or has assumed the ritual authority to perform that particular ritual act.

The underlined portions we call the indicating facts or signals, and the italicized portions the indicated contents.


The fact that the man is wearing the dress of a Roman Catholic priest carries the information that he [... is going to a fancy dress party.] [...has delusions of grandeur] [...is protesting against the Catholic Church's policy on contraception]

It is not clear that the first thing indicated induces the second - on the contrary everything exists in a tautological relationship. Roman Catholic Priests wear a certain dress/RC priests are allowed to officiate ritually - rituals are officiated over by people wearing the right gear. These are just stereotype facts distinguished from the field of possibilities then added to each other as if in a logical relationship. You have to wear the right clothes if you are going to officiate at a ritual might be the general point here. But, the right clothes are already tautologically tied to the meaning given to Roman Catholic Priest etc.

If I can indicate (point to) a fact of some kind then that means I already know what that fact is about. I already know about the Catholic Church otherwise I couldn't point to the fact in the first place.

However, I can make what I have pointed out a little more meaningful if I can find some other ideas and concepts or facts 'coincident' that give some further possible lines of interpretation. But that is called interpreting a situation - you start with something noteworthy then add more kinds of information until you have an interpretation.

Why do we need the 'architecture' metaphor apart from the fact that it sounds fancy and exciting?
Comment by M Izabel on September 14, 2010 at 2:26am
Is this how we want anthropology to look like? Logicians and computational linguists are doing this kind.

Translate

OAC Press

@OpenAnthCoop

Events

© 2018   Created by Keith Hart.   Powered by

Badges  |  Report an Issue  |  Terms of Service