How to build an argument with the zettelkasten?

fernandolasman

Hi, I'm just beginning to learn about the zettelkasten method and I'm struggling to understand how logical implications are recorded/expressed between the notes. As far as I understand now, the links between the notes do not imply logical relations, so to actually build an argument one has to search in general for related cards.

Can anyone help?

iylock

Maybe I misunderstood? I'm not sure.

I think it would be a good idea to create a structure note for the argument.

How about this idea?

1. Thread format
   You can write like a Reddit thread. If you attach a UID to each post, you can link to it when you refute it.

However, I think this method may be difficult to see at a glance.

2. Card game method
   Claim vs. counterargument. Each side holds a card. That card is the basis(or ground or evidence). Each basis can be made with Zettels and given a UID. Then create a structure note (or relationship note) as follows.

# 202412162029 Battle_Milk is good for your health

pro: Milk is good for your health.
- Ground 1 [[UID]]
- Ground 2 [[UID]]
- Ground 3 [[UID]]

con: Milk is not good for your health.
- Ground 4 [[UID]]
- Ground 5 [[UID]]
- Ground 6 [[UID]]

And each basis fights against each other. If the basis loses strength, it is marked as a lost card.

# 202412162029 Battle_Milk is good for your health

pro: Milk is good for your health.
Cards remaining:
- Ground 1 [[UID]]
- Ground 3 [[UID]]

Cards lost:
- Ground 2 [[UID]] <-- Removed by Counterargument 1 [[UID]]


con: Milk is not good for your health.
Cards remaining:
- Ground 5 [[UID]]
- Ground 6 [[UID]]

Cards lost:
- Ground 4 [[UID]] <-- Removed by Ground 3 [[UID]]

And at the end, there is a one-line summary of the results of the game.

# 202412162029 Battle_Milk is good for your health

Game result: Milk is good for your health. pro win.

...

And other arguments related to that debate can be collected in another structure note.

# 202412162054 Foods that are thought to be healthy if consumed

+ Is milk good for your health? Yes [[202412162029]]
  - So when should I eat it?
  - ...To be added...
+ Will eating tomatoes make you healthy? In progress [[202412162133]]

iylock

I'm not sure if my translator is wrong or if I'm not understanding properly.

If your question is simply about how notes are connected (for example, the contents of this note become the basis for that note, or this note and this note lead to a conclusion based on the contents of that note.),

You just need to write a description of the link after the link [[UID]]. This is called link context. All you have to do is describe what the relationship is between the linked note and this note and why you linked it.

Will

@iylock that is an interesting card game.

I use the context model and make it a habit to include a one-sentence summary, which I call subatomic, with each link to give context to the link.

Another idea would be to prepend an argument's name/filename/title with a symbol or acronym. Consider using foreign keyboard symbols, icons, and dingbats; here are a couple of examples.

BA-Rainy Days Make Me Sad [[202412160450]] (Basic Argument)
C-Stay Indoors on Rainy Days [[202412160451]] (Confims Argument)
CC-Human Skin is Waterproof [[202412160452]] (Counter Argument)

We don't have to limit ourselves to one method or the other. We can mix and match, blend and extend. A third way is to use explicit tags.

Titles can contain info about argument structure.

Counterarguments To Minimal Effective Dose 202401311657

All these tactics can serve other mental models, such as evidence, theory, or any area of study. I'm sure there are other, more sane ideas, but these are the ones I've used.

JasperMcFly

link between notes is an association you make- can be anything.
Can write a few words at link to remind you why you made association.

Examples
continues on 10/1a
subargument 2 on 10/1a
counterpoint on 10/1a
see references on 10/1a
see definition on 10/1a
questions on 10/1a
see related idea on 10/1a
also consider 10/1a
author 2 disagrees, see 10/1a
author 1 agrees, see 10/1a
see another example on 10/1a
changed my opinion since then, see update on 10/1a

tjex

It would also be good to read through "three layers of evidence" if you're looking for some guidance on thinking in a structured way, in addition to actually building an argument technically within your Zettelkasten system.

https://zettelkasten.de/posts/layers-of-evidence/

iylock

I think the translator didn't translate the word 'logical implications' properly.
Can't you express logical implications using structure notes?

# 202412171635 TRUE - if (get 100 points) then (get paid $1 million) 

summary: cond(p, q) = 1 

p implies q.
p: I get 100 points on the test. [[202410302231]] = 0
q: I get paid $1 million. [[202307281314]] = 0 

p = 0; q = 0, therefore: cond(p, q) = 1

or

I get 100 points on the test. [[202410302231]] --> I get paid $1 million. [[202307281314]]
p = 0; q = 0, therefore: cond(p,q) = 1

and make a list.

# 202412171643 Logical implications related to X 

- (get 100 points) X (get paid $1 million) --> T [[202412171635]]
- ...

I think it will be resolved by creating a structure note regarding the logical implications.