Creating a web-based format for debate and deliberation: discuss?
[ by Charles Cameron — Talmud, hypertext, spider webs, Indra’s net, noosphere, rosaries, renga, the bead game, Xanadu, hooks-and-eyes, onward! ]
.
Let me firmly anchor this post and its comments, which will no doubt shift and turn as the wind wishes, in discussion of the possibility of improving on current affordances for online deliberation.
Let’s begin here:
@hipbonegamer @Aelkus @pmarca Andy van Dam cited “Talmudic Annotation” as ur hypertext pattern since the mid 1960’s pic.twitter.com/F1XKEf9F7G
— Greg Lloyd (@roundtrip) December 11, 2014
**
There are a variety of precursor streams to this discussion: I have listed a few that appeal to me in the sub-head of this post and believe we will reach each and all of them in some form and forum if this discussion takes off. And I would like to offer the immediate hospitality of this Zenpundit post and comment section to make a beginning.
Greg’s tweet shows us a page of the Talmud, which is interesting to me for two reasons:
it presents many voices debating a central topic it does so using an intricate graphical format
The script of a play or movie also records multiple voices in discourse, as does an orchestral score — but the format of the Talmudic score is more intricate, allowing the notation of counterpoint that extends across centuries, and provoking in turn centuries of further commentary and debate.
What can we devise by way of a format, given the constraints of screen space and the affordances of software and interface design, that maximizes the possibility of debate with respect, on the highly charged topics of the day.
We know from the Talmud that such an arrangement is possible in retrospect (when emotion can be recollected in tranquility): I am asking how we can come closest to it in real time. The topics are typically hotly contested, patience and tolerance may not always be in sufficient supply, and moderation by humans with powers of summary and editing should probably not be ruled out of our consdierations. But how do we create a platform that is truly polyphonic, that sustains the voices of all participants without one shouting down or crowding out another, that indeed may embody a practic of listening..?
Carl Rogers has shown us that the ability to express one’s interlocutor’s ideas clearly enough that they acknowledge one has understood them is a significant skill in navigating conversational rapids.
The Talmud should be an inspiration but not a constraint for us. The question is not how to build a Talmud, but how to build a format that can host civil discussion which refines itself as it grows — so that, to use a gardening metaphor, it is neither overgrown nor too harshly manicured, but manages a carefully curated profusion of insights and —
actual interactions between the emotions and ideas in participating or observing individuals’ minds and hearts
**
Because polyphony is not many voices talking past one another, but together — sometimes discordant, but attempting to resolve those discords as they arrive, and with a figured bass of our common humanity underwriting the lot of them.
And I have said it before: here JS Bach is the master. What he manages with a multitude of musical voices in counterpoint is, in my opinion, what we need in terms of verbal voices in debate.
I am particularly hoping to hear from some of those who participated in tweeted comments arising from my previous post here titled Some thoughts for Marc Andreessen & Adam Elkus, including also Greg Loyd, Callum Flack, Belinda Barnet, Ken (chumulu) — Jon Lebkowsky if he’s around — and friends, and friends of friends.
What say you?
December 12th, 2014 at 2:29 am
[Leaving aside the observation that a blog post with sequential following comments certainly falls short of the polyphony you seek…]
.
I’m all for exploring the idea; in Microsoft Research there has been periodic effort to go (far) beyond real-time Tracked Changes on documents, but with no dramatic success, at least as measured by users; and there’s also the fundamental question of “why stick to the document artifact of much broader discourse?”
.
Without enough time (or intelligence) to put enough thought into it (yet), it seems to me that you are proposing something related to the sometimes-sighted “collaborative hermeneutics” movement, which some trace back to Foucault’s lectures on “subjugated knowledges.” The clearest effort I’ve seen at this (though without anything like a GUI on it) is this paper, believe it or not: “A?hermeneutic?analysis?of the Denver?International?Airport?Baggage Handling?System” (included as a chapter in a book, find it here http://www.academia.edu/1034724/A_Hermeneutic_Analysis_of_the_Denver_International_Airport_Baggage_Handling_System). Fun read, but see especially the “Reflections section at the end, the few para’s at bottom of p. 75.
.
The Denver example – and much of what I remember of Foucault, and of Kafka’s The Castle – reminds me of the insane recursive arguments we used to have in the good old days of hardcore Kremlinology, particularly in counterintelligence discussions. Without a single author [= Bach], fully intent on guiding and resolving multiple lines from divergence into conlusions, “polyphony” is usually just a polite term for chaos. If you can come up with a GUI or web-based interface that can elicit a communal Bach, you’ll deserve your Nobel. Or at least a Grammy 🙂
December 12th, 2014 at 3:24 am
Charles — Thanks for the invitation, and the running start with your points on 1) many voices; 2) debating a central topic; 3) using an intricate graphical format.
.
I don’t know what Edward Tufte thinks of Talmudic annotation, but from what he’s written on information density, spatial relationships, and pure content without chrome — he *should* like it, and so do I, for the same reasons.
.
As I understand the Talmudic style, it uses “chunking” wisely, so that the polyphonic discussion is clearly centered on the core topic both spatially and in meaning. The outward spiral of annotation use a spatial representation of the time dimension, visually separating temporal layers of commentary, point, and counter point. Frankly, I don’t see a better style for presenting chunked debate and discussion clearly and compacty – but look forward seeing what other folk suggest.
.
If you ask what I *want*, I’ll cheat on “constraints of screen size” and say that I want:
.
1) The same chunked organization (where chunking is either historical or based on an after the fact break down of a speech like the State of the Union or a core text);
.
2) Chunks organized left to right, top to bottom for English. Otherwise for Hebrew, Arabic, Chinese, Japanese…
.
3) On a 10 foot high by 20 foot wide wall where I can read the discussion from a few feet a way, or use magic glasses to zoom closer, pan, or stitch together fragments with a word or gesture.
.
Actually, I want a *room* with four 10 x 20 foot walls with the same properties. A bigger room or connected rooms if necessary — to avoid cluttering the ceiling and floor, and to allow many people to walk, talk, and gesture through the space at the same time. Someone can also figure out how to grow discussions by adding satellite chunks to refactor and extend big discussions while retaining core and satellite topics at the center, and time layered organization.
.
Many years ago at my first job (NRL) my boss and I were tasked as members of a NASA Space Transportation System (STS) review for the Carter administration. At the Johnson or Kennedy space center, we visited an STS project room about with four perfect whiteboard walls bigger than 10 x 20, with diagrams, photos, papers and top ring bound summary sheets covering every surface.
.
As I recall there was no conference table, and no chairs; it was designed for standing meetings, walking from place to place along the walls, with just a hand or pointer to guide peoples attention to a diagram, graph, or report. The whiteboard surface covered everything, even the inside of the door used to enter the room — with a push plate rather than a door knob. The entire ceiling was luminous, no shadows, very even lighting. It was like being in a Kubrick movie.
.
I want that. Covered with instantly updatable, flowing, stitchable Talmudic annotations, with perfect voice recognition and gestural commands for magnification, linking, annotation, and note taking. With space for a dozen or so people to walk, talk, gesture, discuss, and debate.
.
Also: transcending the limits of space and time (people don’t have to be physically present the same room at the same time, but that would be best).
.
With wine, drinks and very tasty food (I recommend Birch restaurant in Providence RI).
.
Is that too much to ask? No!
December 12th, 2014 at 6:15 am
Well, Lewis:
.
I would like to see the format, such as it might be, developed for open use by interested parties, but it might work first as a sort of analog of a magazine, with the editor-analog serving the “Bach” function, and a single topic chosen in the first instance as a proof of concept.
.
I thought at one time that the Israeli-Palestinian conflict might be a suitable topic, lured on by a remark Edward Said made in his capacity as a musician / music critic / essayist rather than as a Palestinian polemicist:
My idea at the time was to have a multi-node game-board (here in a seriousl reduced image):
such that the individual nodes could be named on the board, and each one hyperlinked to a more detailed textual / visual / aural / numeric / filmic account of the node’s meaning. And at the time, I wanted every edge between nodes to represent some linkae between the nodal concepts — but that’s way too restrictive, and if i continue playing with that idea. I’ll make the edge connections voluntary.
.
There’s a completed game about nuclear war in myth and poetry that will give you some idea of how that would work, with ten moves and a final 10-move board with the moves named on it, here on Scribd. If you’d prefer to read about Yeats and ung, try here — or for a lighter 7-move game, War is Sexy, here. In each case, imagine the board as GUI and the moves as linked pages, and you have the concept.
.
Note that these are all solo games, though — appropriate software could be used that way, but it’s the conflict resolution possibilities, with two or more parties in tension, that I imagine most interest us here.
December 12th, 2014 at 6:29 am
One other thought, Lewis:
.
I’m halfway through the Denver AIrport piece you linked, and this para quoted from “Kidder” comes close to describing the sort of editorial attitude I’d see as requisite:
My emphasis, then, would be on retaining the best (most powerful & succint) arguments & sub-arguments on each side present in play, with some form of branching structure to drill down into finer & finer levels of detail.
December 12th, 2014 at 6:48 am
Greg – whoa!
.
Are you moving the room with whiteboards wholesale over to the restaurant, or binging the restaurant to the whiteboards?
.
It’s getting late, I look forward to responding properly tomorrow.
December 12th, 2014 at 4:10 pm
Charles – Not just a restaurant! The room would be a shared space – physical or virtual – with affordances designed to promote “civil discussion which refines itself as it grows”
.
1) A limited number of people at one time. A dozen or so people might be physically or virtually (video, audio, text) present at one time, talking, debating, linking, writing on the wall.
.
2) Everything that’s written on a wall is visible to the outside world, as are words written or spoken within the room based on agreed norms, like Chatham House Rules. For example talk within the room might be: a) visible outside the room, or not; b) permanently recorded or not ; c) anonymous (to the World and record) or not.
.
3) If many people are eager to enter the room, there’s a line. People might be given 24 hours or so “in the room” before exiting and rejoining the line if others are waiting. Special guests, honored members, and others might be granted a privilege to enter at will and stay as long as they wish.
.
4) Trolls could be banished permanently or given a temporary time out. Some rooms may have membership by invitation only – voted by members or other rules.
.
5) Drinks and tasty food are rewards for people who go to the trouble to show up in person rather than virtually! Face to face discussion and debate would always be the most fun, and should be rewarded.
.
This is a semi serious response to the problem of civil discourse that scales.
.
A group of people in the same “room” act differently based on the norms for that space, how permanent and visible their actions are, who’s in the room, and the closely readable reactions of others in that space.
.
A dozen or so people at a time in a shared space would likely have a more civil conversation and written discourse than an unlimited number of simultaneous participants. Limiting the number of people in a room at the same time would also limit the crowd who jump into a space base right after an event (like the State of The Union), or are recruited as trolls.
.
Two Talmud questions:
.
A) How was the printed Talmud page shown as an example “moderated” from BCE through medieval and current time? Who decided what commentary was worthy of recording as part of the cannon? Is there a canonical Talmud, or are there recognized variations in content and interpretation?
.
B) How is current Talmud study linked to the Talmud cannon? Have hypertext and the Web influenced Talmud discussion? Like other current scholarly work – or based on Talmudic annotation and discourse?
December 14th, 2014 at 12:42 am
My son returns from his first university semester today. so I’ve been busy and may still be a little slow responding, but wanted to add this, from my Twitter feed:
.
.
Great link — thanks, Ken!
December 14th, 2014 at 2:57 am
BTW, those (“mischievous, laconic”) are the words of Emmanuel Levinas, through whom I first ran across the Jewish proverb:
Succinct, brilliant — revolutionary in the best and most inward-turning sense.
December 15th, 2014 at 4:35 pm
Greg,
.
Before I respond to your Talmud questions, a quick disclaimer is necessary: I have studied Talmud for 20+ years (and even wrote a paper about it in law school – http://www.jlaw.com/Articles/Rodef%20paper-%20Final.pdf), but I am no historian and therefore do not have a full grasp of my dates and timelines.
.
a) There are actually two versions of the Talmud, the Jerusalem Talmud and the Babylonian Talmud. The latter version is typically referred to as “The” Talmud though, as it plays a more central role in deciding Jewish law. The text of the Talmud itself has remained relatively constant throughout the centuries other than some minor censorship and thousands of [relatively] small textual emendations. The commentaries around the page were added over time. Rashi (who explains the text) and Tosfos (a collection of authors who reconcile seemingly contradictory passages) were added first. Then, over time, the other annotations slowly made their way onto the canonized page. Even today, there are new annotations that are added when a new printing of the Talmud is done, although they are mostly reference works and not commentaries.
.
b) Current Talmud study would be instantly recognizable to a visitor from the distant past; the “chavrusa” [pair of individuals studying Talmud together] learning out of an ancient text still forming the core experience. However, given the abundance of new technologies – chief among them Bar Ilan’s Responsa database – preparing for a lecture can often involve a lot of database querying, copy/pasting, etc., and has definitely shifted tremendously in the past decade or so.
December 15th, 2014 at 6:57 pm
Thanks, Jonathan. I am reading your article on din rodef with appreciation.
December 15th, 2014 at 10:09 pm
Charles,
.
It seems to me that the Delphi Process used to iteratively explore sets of issues could be used to achieve the end you seek. Would require a facilitator (or a team) to moderate the process and to propose new aspects to explore during a particular iteration. Voting/ranking would be used to identify both areas of agreement and conflict. One could build a brainstorming feature into the process.
.
Example: facilitator asks the group for ideas both in support of and against a particular position. The results could be grouped and combined and the results assesses.
.
The entire process could be done asynchronously and online — utilizing some existing forum processes and some new ones designed specifically for the collaborative, dialogue process that the Delphi Process represents.
December 15th, 2014 at 10:38 pm
Jonathan – Thank you! I’ve heard stories of Talmud scholars who could put a pin through pages and tell what words the pin hit – on both sides of the paper.
.
From an information modeling point of view, the canon and Talmudic annotation seem to provide a stable coordinate system (link coordinate system) which has a wonderful ability to present temporal layers of annotation as a dense spatial layers modeling temporal layers of analysis, spiraling out from from the core topic.
.
Such a stable coordinate system and corpus can be both an aid to memorization and navigation (like Cicero’s mnemonic method), and the basis for dense indexing, linking and external analysis. Links to stable coordinates don’t break. It seems like the right direction for Charles goal of “online deliberation” where craft and economy of language are used to create the corpus. Not so great for reddit volume and velocity.
.
The Talmudic style seems best for a community of scholars, perhaps guided by the hand of a J.S. Bach or higher power. A different corpus could be continuously refined and refactored as it grows, less like the Talmud, more like Charles’ gardening metaphor.
December 16th, 2014 at 3:57 am
Greg,
.
Two quick points:
a) the stories you have heard are accurate. I have seen scholars perform the “pin trick” myself. But these are mere party tricks; true Talmudic scholarship involves the ability to take a complex topic, instantly boil it down to its essence, and either prove or disprove the concept based on a seemingly unrelated “sugya”, or topic.
b) to build on your point about a stable coordinate system facilitating memorization, it is impossible to overstate the impact of having each folio in the Talmud taking a slightly different visual form has on memorization. When I am trying to locate a passage, I first recall where on the page it is found, then what the page looks like, and that can usually help me place its general location in a vast sea of similar-looking text.
December 16th, 2014 at 10:09 pm
Jonathan – Yes, memorable visual form (map like), stable coordinate system, and literal rather than encoded content seem to be distinguishing characteristics of Talmudic annotation – or at least the characteristics that I like.
.
Other codifications have stable coordinates (like numbering of the US Code) and spatial clustering of commentary (like inline footnotes with printed pages), but I doubt that spatial configurations of footnoted pages are truly memorable.
.
The Talmud seems much more visual, analogous to topographic survey maps, but built with literal content rather encoded using contour lines and symbols. Like survey maps, the landscape of the Talmud is stable and memorable, subject to refinement not radical change – unless you consider Mount St Helens.(1)
.
Associative memory based on visual form, location, and content – “That reminds me of a different place that looked like X” may help find seemingly unrelated “sugya” quickly.
.
Because the Talmudic representation is literal rather than symbolic, it’s much more densely encoded that a 2D map or the symbolic diagrams Charles describes, where you need to remember the content behind each symbol.
.
The Talmud uses pure content and spatial encoding of commentary and temporal relationships, which is why I think Tufte should like it!
.
(1) A friend in Oregon worked as a mountain rescue volunteer, and keeps a framed copy of his old Mt Saint Helens topographic map on the wall of his office. Showing the mountain top that’s no longer there.
December 17th, 2014 at 5:04 pm
Ward, all:
.
I’m behind on some tasks and doing furious catchup, hope to reinvigorate my end of this discussion once my “desk” is “cleared”.