Hierarchy of Agreement

“Hush, minion.”

Yesterday I wrote a posting about online discussions. One topic was Graham’s “Hierarchy of Disagreement”. It provides a hierarchy of different ways people can disagree and the value of these disagreements.

I wonder whether there is also something like a “Hierarchy of Agreement”. After all, there is a long research history regarding persuasion and different ways to get people to agree. Thinking about the issue I came up with the following hierarchy. Essentially it mirrors Graham’s “Hierarchy of Disagreement” (link to his essay) and it shows “how to agree” (well).

Hierarchy of Agreement, based on the “Hierarchy of Disagreement” by Paul Graham (essay).

It might sound strange to apply different value to different kinds of agreement. Isn’t all agreement good? After all, who cares why people do what you want, as long as they do it?

Personally, I think that “agreeing well” is as important as “disagreeing well”, perhaps even more so. I am very critical of “unreflected praise”, “blind loyalty”, or “style over substance”. It’s one thing to make a point or win an argument. Demagogues can do this quite easily. Sway the public, get them to agree. But it is quite another thing to ensure that people understand the issue and carry on the position on their own. To not only win one battle, but the war.

This issue becomes more clear when looking at the different levels and why you may or may not want people agreeing with you this way. Again I follow Graham’s essay.

AH0: Unspecific Praise (was: name calling)

The lowest level is unspecific praise, something which you can give to anything. Comments like

“Love it.”



are essentially useless. So someone loves it, so what? It’s usually a short-term gut-reaction. This is actually a pet issue for me and one thing I try to convey to students. “I liked it” is nice feedback — but not good feedback.

Just imagine if you only have fellow campaigners on this level. People who like it, but can’t say why. Not convincing. And without the material in front of them, they would probably move on. Not a basis to build anything on.

AH1: Personal Praise (was: ad hominem)

Getting praise as

You have/are/have experienced x, you know what you speak about.

carries little weight. You have to show the effect in the arguments themselves. This level is similar to an “ad hominem”, but positive. I thought about reusing the term, but “ad hominem” is usually used for disagreement.

This form of agreement is fairly common and extensively used in advertisement when it comes to so-called “experts“. Yes, expert statements carry weight, but not because experts make them. It’s the quality of the arguments and evidence that should be convincing, and experts should have better arguments and evidence.

Likewise, people with personal experience might have a better perspective and more knowledge. But they also might not see the wood for the trees. Even worse, personal experience can provide a strong bias. There is a good reason why victims of violent crimes should not be part of the jury to determine guilt and innocence. What works for them and what they want might be different from what is best for you, because they have made experiences you do not share.

If a campaign would rely on this level, then you probably have some figure-heads. These people know the truth and that’s it. Adherents would still not know why they agree, it would rely on blind trust. And without the figure-heads, the movement would lose its direction.

AH2: Praising the Tone (was: Responding to Tone)

Good writing is a joy to read. But just because something is written well does not mean that it is true. Frequently, writing that appears to be deep is without substance. A “postcard philosophy”. Also, one of the easiest ways to get agreement is to make fun of the opponent. If done well, it might even get people on the other side to laugh at one of their own. However, this does not mean that it is also true and it carries no weight. Beautiful writing does not equal true writing, and you can make devastating jokes about anyone.

Thinking in campaign terms, well-written essays can start change or provide additional punch. Ridicule and satire can be very effective. But style is no replacement for substance. Even worse, sometimes irony can work like a pressure relief valve. It can also turn into a “Bread and circuses” situation, where it is more fun to watch the opponents claw at each other than improve the situation. And the people agreeing still don’t know why they agree.

AH3: Agreement with Arguments (was: Contradiction)

In contrast to the previous levels, here the person states to which arguments s/he agrees to. At the very least, it repeats the arguments for others to hear and gives the author feedback on which arguments are effective. It does not add to the discussion, but at least it addresses the actual arguments and evidence.

For a campaign, people agreeing on this level can repeat why they agree. So there is a chance that this position might spread. The question remains whether they are actually able to hold their ground when they are confronted with counterarguments and rebuttals. They might if they have been provided with them as well (“innoculated“), but it’s unlikely that they form their own. Still, if it’s a movement, here the movement might spread.

AH3 also marks a qualitative change. The arguments become important, not the peripheral aspects of the discussion. Using the Elaboration Likelihood Model of persuasion, it is now the central route, not the peripheral route, that is used.

AH4: Add to Arguments (was: Counterargument)

This is the first level where the discussion is moved forward. The writing not only convinces, it stimulates the person to provide additional material. Here, the person who agrees provides an additional argument to the author and anyone who reads the comments. Perhaps it’s something the author did not see, or did not deem important enough to mention because it might be peripheral. In the later case, his/her readers disagree and close this gap. It is a more additive and thus quantitative improvement.

From a campaign perspective, you can reasonably expect that the movement moves forward. Other people might improve the position and carry on.

AH5: Qualitative Improvement (was: Refutation)

No argument is perfect and there likely will be imperfections in the provided arguments. A qualitative improvement addresses these and removes contradictions. It shows ways to deal with borderline cases. It can provide additional evidence for links between the arguments. Even more important, it addresses and refutes criticisms. It improves the position in a perpetual discussion with its opponents.

Here you not only have people who advance the cause, you have intelligent people who invest effort in understanding and improving it.

AH6: Improving the Central Point (was: Refuting the Central Point)

It matters what you address, in disagreement and agreement. Here, the improvement deals with the central point.

From a campaign perspective, you have people who not only advance the cause and invest a lot of effort. You have smart people who do so.

“What It Means”

As with Graham’s “Hierarchy of Disagreement”, the level of agreement is not related to the truth of the position. But it can be used to evaluate positions and followers.

Unless you need a high body count (for demonstrations or retweets), everything below AH3 can be discarded. And even if you “only” need a high body count, you better pray that no-one asks hard questions why they agree. You might have just fielded an army, but destroyed your recruitment efforts.

With AH3, you have at least people who understand the arguments and can repeat them. But if you want a movement that improves, is resilient to disagreement, and continues without figure-heads, you need more. AH4 and higher is desired, and the higher the better.

I think that improvement is only possible if people discuss on the higher levels of the “hierarchy of disagreement” and the “hierarchy of agreement”. In this case, people argue for their positions on the higher levels of agreements and react to criticism on the higher levels of disagreement. They make good arguments and provide evidence and that is why people agree to this position. When faced with disagreement adherents can improve their position. They advance the discussion with new arguments. They might even advance science by requesting new evidence.

Thus, high level agreement and disagreement might advance society.

2 Trackbacks / Pingbacks

  1. Minimum Requirements to React to Comments | ORGANIZING CREATIVITY
  2. Useful discussions | ORGANIZING CREATIVITY

Comments are closed.