Calculation of “Zero Position” = Decentralized Socrates

5 min readSep 3, 2021


Decentralized Socrates (Part 16)

Rhetorical verification is essentially only possible in decentralized organizations. That was the conclusion of my last entry. It is also the answer to the question raised at the beginning of this essay, “What can only decentralized organizations do?”

Rhetorical verification has a byproduct. That is related to the Socratic method of dialogue discussed in this series.

Suppose that rhetorical verification results are assigned to many people and/or sites.

These results would be a collection of data in the form of, for example, “site or person -> {topic 1 x position, topic 2 x position…}”. Ex-President Trump -> {Russia -> +0.7, Mexico -> -0.4…}” and so on. In short, a list with a large number of topics and positions would be allocated to each site and person.

Then there is a kind of map where many positional talks are embedded at the same time. You could force it to be displayed as a two-dimensional or three-dimensional cartography.

Yet, there is something more crucial. It is possible to calculate the corresponding “zero position or center of mass,” which is a narrative with a few positions as possible, by utilizing various subjects with different positions as axes.

As a simple illustration, the vertices of a triangle represent subjects based on various positions, and the center of gravity of the triangle represents a subject not belonging to any position. Once we have the positions corresponding to the triangle's vertices (or the base axis positions calculated from various positions), we can derive the “center of mass = zero position” and who are the subjects close to it.

Of course, the actual calculation is not for triangles in two-dimensional space but m-gons or more complex shapes in n-dimensional space. Also, there are various approaches to calculating the “center of mass.” As pointed out last time, such heterogeneity of calculation methods is desirable because a single calculation method cannot eliminate the inherent bias contained in the procedure.

An entity that does not express any opinion is position zero. However, that makes no sense. In this case, there is no topic at all for the subject to mention.

There must be some subjects whose position on a topic is judged to be neutral or close to neutral by rhetorical verification, even though they have expressed many views.

Since “neutral” is a relative matter to other positional talks, the more subjects engage in diverse positional talks, the easier the system to see the subjects with relatively neutral positions and extreme ones.

This is where the decentralizability of rhetorical verification comes into play. A sufficiently decentralized system, like nature or the market, cannot itself take a particular position. As confirmed last time, that fact was the core of the link between the decentralized organization and rhetorical verification.

So a “neutral” position calculated by such a system could be a “ really neutral” position. At least better than the values calculated by specific large platforms and governments having many stakeholders.

We can call such a system to calculate “zero position = an entity or position without a position” from various position talks “Decentralized Socrates.”

Decentralized Socrates is a system that can calculate “zero position” from diverse position talks. “Not promoting a particular position = no bias” was the first stage of the Socratic method of dialogue, a state of bias = position disarmed by the self-contradiction pointed out by Socrates.

Decentralized Socrates is an automated version of this benefit of the question-and-answer method. It is decentralized and has no agency. Therefore, no subject can be hated. In this sense, it removes one of the issues facing the method of dialogue: the unsustainability of hatred. Naturally, there will be individuals who will hate the “Decentralized Socrates” system itself. A blockchain, though, will be more resistant to attack than a vulnerable individual.

The characteristic of decentralized Socrates that there can be no one to hate in principle is also vital, considering the psychological origins of totalitarianism (or anti-intellectualism). The anti-intellectual attitude is likely to occur when a person feeling “unable to follow the arguments of the liberal elite” often engages in the psychological strategy of “ scorning all discussions” due to hate for a good debating opponent.

The more enthusiastic the debate, the more winners and losers are created, and the fewer winners there are. Hence, the undecentralized Socratic method of dialogue produces, as successfully as it works, a multitude of hatreds. If presumptuous people who know this structure dare to deceive the majority, they will be on the road to totalitarianism. Therefore, debate and enlightenment are, unfortunately, the path to totalitarianism, regardless of the intentions of enlighteners.

If a person or a site does not want to be labeled as “having a certain positional bias” against Decentralized Socrates, they can simply act like it. If the decentralized Socrates is sufficiently decentralized and the algorithm is transparent, the person will be judged to be standing in “zero position = universal position” from various perspectives.

There is an important hypothesis here. In the same way for us to optimize searchability, it is impossible to “pretend to be in the universal position” by exploiting weaknesses in the decentralized Socratic algorithm. Or, if we do such mimicry, it will end up behaving the same as the universal position, a hypothesis I mentioned in Part 13.

Since decentralized Socrates is updatable, this hypothesis is feasible to a great extent. Contrary to security strategies, attackers who take action against decentralized Socrates must devise more troublesome information manipulation (rhetoric), which is more costly and less understood by humans (i.e., less beneficial).

What is key here is the constraint that “in the end, humans must take the meaning of the information.”

The technology for AI to deceive each other is being actively researched. If AIs are to fool other AIs, endlessly complex rhetorical and defensive maneuvers will evolve. There is the constraint, however: the need to write articles understandable by humans. In that case, complex rhetorical maneuvers will be hampered by cognitive limitations and will not be worth the cost.

Anyway, if zero position = universal position can be calculated by decentralized Socrates, what is the significance? Since everything, including good and evil, is just a mere relative judgment of taste, shouldn’t we rather obtain followers driven by hatred based on ridiculous beliefs than strive for a universal position if these beliefs are known to be false?

Even if decentralized Socrates could calculate zero position, it would be meaningless or just a hindrance for such a person.

Can’t we create a more proactive mechanism in which “people’s attempt to stand in a universal position” itself becomes beneficial to their “survival”?

Next time, we will discuss this point. Translated with

Next Entry (coming soon)