improve the conversation, improve democracy
The Case for the Canonical Debate
why knowledge should accumulate
Those of us working together on Project Gruff (and related projects — see the list in the text below) have as our mission the creation of a platform and related tools that can improve online discussion and debate central to the liquid democracy movement as conceived by Democracy Earth Foundation. As a community, we've kept our eye on other projects (such as Kialo, one of our favorites, and Debategraph, just to scratch the surface). We have also kept tabs on academic research on argumentation, and traditional theory.
Reining in the Web
One of the main principles that we are working with is that Knowledge should accumulate.
Unfortunately, this does not seem to be a direct concern of argumentation theory: the discussion focuses principally on the construction of premises and arguments, and the evaluation of their effectiveness. The focus is on the structure of the debate as a conversation between two or more parties trying to reach a conclusion in a specific instance. No guidelines are given for reusing these conclusions outside the scope of the specific event.
In the domain of popular discourse, we see similar problems in terms of approach: rarely is a dispute settled on Facebook, and then used to inform future decisions via, say, Twitter. Expert opinions are generally concentrated in long-form text on blogs, news sites, and places like Medium, and evidence is scattered to the four winds of the Internet.
There are approaches like ARG-tech's AIFdb, and the Socratic Web which attempt to rein in the confusion by connecting points in the wild related to a single thread of discussion. Whether or not they will be successful remains to be seen, but we are excited about what they are doing.
The Great Attractor
As for our own effort, we are taking inspiration from another project that has had phenomenal success at accumulating the collective knowledge of the people: Wikipedia. By defining a single, canonical place on the Web for everyone to contribute their piece of the puzzle, they have managed to create the largest encyclopedia in the world.
Most of the debate-oriented sites, including Kialo, are taking this approach of providing a single place to come together and debate. In many ways, our effort is no different, and, ironically, the more sites there are attempting the same goal, the less successful we all will be: Wikipedia's success, to some degree, is due to its own success!
Project Gruff is named after my own personal effort to create THE place for debates. However, I soon discovered that I am not alone. Through the Democracy Earth community, I have been in contact with the people behind such projects as Debate Map, Reason Score, WikiD, The Wikilogic Foundation, CollectiveOne, The Internet Government and Digital Peace Talks. While we all have our projects and approaches, we have seen a significant overlap in our goals and our philosophy. Each project at the moment is serving as a test bed for our many ideas, while we finish our white paper towards an effort to building the definitive platform (just to be clear, some of the projects above have unique goals outside of the scope of our project, and will live on in collaboration with the final effort).
The Right Foundation
As mentioned at the beginning of this article, most efforts in modeling debates have chosen the approach of a directed graph, or an argument tree, of arguments for and against the premise of the debate.
This is an easy-to-understand and sufficient model for debates that are held in a single context. Unfortunately, it is not enough for the objective of helping our knowledge accumulate. Consider the example above, the never-ending discussion on whether or not astronauts actually landed on the moon. Although we should have come to some sort of consensus on this topic, there are always those who will disagree.
Now consider a new debate: "We can land astronauts on Mars." One of the first arguments you might expect to see in this discussion is "We landed astronauts on the moon." However, by starting a new debate, given the simple model above, we would have to repeat the whole discussion on whether or not the Apollo landings really happened all over again.
This is the second way in which debates need to be made canonical: they need to be reusable so that we can just “link” a previously-discussed claim into the new ongoing debate. It turns out this is a rather simple thing to do. If you think about it (and argumentation theory tells us this), an argument is essentially a premise (a supposed fact) used in the context of a debate to prove or attack a claim.
Here's an argument for proving the moon landing: "We can see the footprints of the astronauts on the moon." This raises two new questions: Is it true?, and How convincing is this as an argument?
You could picture the first half as being a debate all its own. "Can we see the footprints of the Apollo astronauts on the moon?" This part of the statement-as-argument is universal. Once you have proved, disproved, or simply exhausted the evidence on this topic, there's no need to ever do so again (new evidence notwithstanding). This part is the premise, or claim.
A separate, contextual debate can be had over how good, or how effective, the argument is in convincing readers about the veracity of the moon landing. However, this contextual debate only makes sense for proving the landing. It may be a great argument in this case, but, true or not, it would not have the same impact in a debate that "NASA's Lunar Reconnaissance Orbiter wasn't worth the price." It is the use of the claim in the context of another debate that we call an argument.
Accumulate
With just this slight change in the standard model, we provide the opportunity to reuse facts, once established, wherever they are useful. It may seem trivial, but its impact is profound. There is little point in contributing to a debate that will be forgotten. But if the debate is permanent, canonical, then the top experts will want to be there to prove their worth, and fill in the gaps that only they can fill.
Right now, we are exploring other techniques for accumulating and reusing valid information. If you'd like to join in the effort, you are more than welcome! We are currently writing the first draft of our white paper, where you can place comments. You can also join our Slack team the Canonical Debate Lab for live discussions, and for information on how you can help.
