Thinking on Paper

Here’s my method of “thinking on paper”.

I use a blank A4 sheet in landscape format and divide it into 4x4 equal cells (by hand - there’s no need for a ruler). On a wall at my desk I have a set of sticky notes, the central one describes a small number of key processes: “orientation”, “make progress”, “make a list of difficulties”, “generate ideas” and perhaps two more. I start with cell 1 and choose a process, normally “orientation”. I fill the cell with ideas, arranging them in a small mind map or in a small tree hierarchy. My handwriting is small, but after 7-10 lines the cell is full. For cell 2 I can continue with more ideas on orientation, or I can zoom into a promising idea from cell 1. I have around two dozen sticky notes with lists - a list of stimuli that help me generate ideas on orientation, a list on making progress, a list of simple questions, a list of diagram types I can use to see my topic in a new way, a list of crucial things to look for, like points of transitions, or points with some special property.
Unexpectedly, I found a lot of advantages in the limited capacity of one cell - if I do something foolish, I can just restart in the next cell, there are no imbalanced asymmetric maps, there is a sense of making progress and, most importantly for me, I am guided to refocus and to do “metacognition” much more frequently than when I use larger units of mapping or writing. For this metacognition aspect, I find it very helpful to make a list of difficulties. And the sticky notes lists provide enough stimuli for a broad range of situations.
It should be obvious that the method is meant for generating thoughts, not for communicating them.
There are lots of possible modifications to this paper-based “integrated thought development environment” - use a different number of cells, instead of a fixed grid use narrow columns and keep the cell height flexible, or adapt the stimuli lists in all sorts of ways. And there are lots of practices for cross referencing cells and sheets, arranging sheets in a zettelkasten or in a dossier, or keeping track of goals and problems.

I tried to compile a list of stimuli for each of the key processes mentioned, in order to make sure that I would find interesting connections between my current topic and at least one of the stimuli. (I had the hazy idea of “concept chemistry” - I would add concepts to a conceptual broth until some kind of reaction happens - which is perhaps alchemy rather than chemistry.)
Most lists are small tables, where I strike out less useful items and add new ones. It’s a work in progress.
Here are some lists. I suspect others won’t find them useful - compiling your own lists is arguably the better way.
Orientation: What are the results, the deliverables? What are possible visions? (This points to a separate list of stimuli for visions.) What is crucial?
List of problems: What seems not right? What is puzzling? Where are the gaps? What is missing? Are there conflicts? Where could a failure occur?
Diagrams: Timelines. Graphs. Pie charts. Flow charts. Decision trees.
Idea generation: I find it easier to react to a stimulus with a lot of aspects, so I often use stimulus concepts like library, market, internet, organs, … Besides that, I have a condensed list of TRIZ principles and of general “modification verbs”, like rearrange, stretch, rotate, in the spirit of the SCAMPER acronym.
Look at the situation - what things are perhaps not obvious? Again, I’m experimenting with evocative questions, like: Is there a fingerprint? Can you identify points of transition? What are extremal points?
What can be measured?
When working on cell after cell, I do not consult these lists at every step, but just go with the flow of thinking. However, I suspect that it is essential that the lists are directly accessible with a glance - previous experiments with lists in a zettelkasten didn’t work nearly as well.


Thank you, Thomas, this is interesting!

I also enjoyed your posts about this:

Especially the example picture was inspiring, maybe you can post again one of these to illustrate your points.

Here’s a first almost-impromptu sample sheet, which I’ve done on the topic of “3x3 sample sheet”. The sheet tries to showcase a number of key features, like the “criticism” topic in cell 1B.

For my own work, I am using a 4x4 layout, a mechanical pencil and tiny tiny handwriting in the proud tradition of the Humboldt brothers and Arno Schmidt;-) For this illustration, a 3x3 layout was more legible.

And here is a very similar digital sample with a 4x4 layout:

1 Like

Here’s another short introduction to the method of combining note-making and thinking tools.
Feedback is very welcome!

@thomasteepe Thanks for sharing! So is this boxed format how you write your notes? Do you find that to be a key feature? Do you also use a digital note system?


So is this boxed format how you write your notes?
Do you find that to be a key feature?

Yes, for me it is. Months after I wrote my last comment, I am still using the method described above, with a combination of a) writing in columns and using flexible box sizes and b) using fixed box sizes. Item a) plays a dominant role in analysis and item b) a dominant role in idea generation.

Do you also use a digital note system?

I’ve done a lot of experiments with different iPad Pro apps, which I found very interesting. But for things I find really relevant, I use paper.
I hope this is not seen as mindlessly defiant - for me, work on paper in the process of developing ideas just has massive advantages over digital work:

  • a change to the screen work in my day job,
  • no distractions,
  • no privacy concerns,
  • a brain - eye - hand interplay I like a lot,
  • and, above all, huge flexibility in combining texts and diagrams and symbols.
    My present view is that my personal shortcomings in developing ideas are better tackled by designing new thinking routines than by finding a superior software that could be used to implement such routines. But that concerns my present motivation - I am still curious about innovative apps, it is just not my focus.

In my initial post from January 2021 I had mentioned a number of thinking routines - what has happened till then?

  • Diagrams and visualisations play a larger role in my notes. I’m bad at drawing, so some visual things are just in my head with a verbal description on paper.
  • I find it surprisingly useful to outline an imaginary letter to a person about my topic - this is in the tradition of Kleist’s “Gradual construction of thoughts”, or the software development practice of “rubber ducking”.
  • I often try to “evolve a model” - I start with a simple model of my topic and then play around with its elements and their relations, using different stimuli.
  • I’ve dedicated a somewhat larger time budget to idea generation with a “wilder” set of stimuli.
  • I’m still experimenting with options for an interplay between “control tower sheets” and “work in progress sheets” to improve the coherence and straightforwardness of thinking over several sheets and several sessions.
  • Several of these items are linked to my background in mathematics - in both directions:
    How can practices from math problem solving be used in general thinking, and how can paper-based methods be adapted specifically to math problem solving?

Here is a slightly updated version.

What is this about?
I am looking for a method of “thinking on paper” with the following properties:

  • The method should be cheap, stable and available.
  • The method should be open to users of different age with different levels of expertise.
  • The method should be modular and adaptable in various ways, with support for representing ideas as text, diagrams or math terms, and with different sets of thinking tools.
  • The method should integrate writing methods and thinking tools.
  • The method should provide support for common situations in thinking and problem solving - how to understand a problem, how to deal with obstacles, how to generate ideas etc.
  • The method and its user should develop, co-evolve together - starting perhaps with a simplified version for writing practices and thinking tools and from there moving towards more elaborate versions.

What follows is the present state of the method.
It consists of five building blocks that can be combined and integrated in a number of ways. Based on this modular design, we can develop all sorts of variations: A method based on notebooks, a method that only uses mind maps, a method with an emphasis on generating ideas, a method focused on logical deductions etc.

Building Blocks

Building block 1 - the sheet layout:

  • I use A4 sheets (90 GSM) in landscape format and divide a sheet in 4 columns and 3 rows - there’s no need for exactness and I do it without a ruler.
  • Alternatively, I can use just 4 columns and make boxes of different sizes.
  • I fill the boxes in column 1, then in column 2 and so on.
  • The single boxes can be referenced via 1A, 1B etc., with the sheet number where needed: 123.1:2C.

Building block 2 - the box layout:

  • I use each box for a simple small mind map, or for indented text, or for a diagram, or for math calculations.
  • The size limit of the box will come as a surprise, but it has a number of advantages which I will describe in a moment.

Building block 3 - thinking tools:

  • I can use thinking tools as topics for the boxes, to stimulate my thinking about the problem, like: What are key questions? What are my options? How can I split the problem into smaller parts?
  • I find a personal collection of thinking tools very useful - I have a corkboard over my work table and can access dozens of tools and stimuli with one glance.
    (In my view, the thinking tools are the core element. They propel the thinking process forward - the other building blocks provide a stable framework that makes using the tools easy. )

Building block 4 - dossiers, work in progress and control tower sheets

  • I collect the sheets in dossiers. There is one or more summary sheets with numbers S123.1, S123.2 etc., and as many work-in-progress / WIP sheets with numbers W123.1, W123.1.1, W123.2 etc.
  • The summary sheets contain all the relevant insights and ideas from the WIP sheets. In column 4 of a summary sheet there is often some sort of road map or a set of open issues I want to examine in that dossier.

Building block 5 - the zettelkasten

  • I keep the dossiers in a regular A4 index card box.
  • I use card tabs with the dossier topic on it.
  • I take out entire dossiers from the ZK and use placeholders to make re-storing easier.

Main advantages - for me

  • A conventional mind maps works best with a couple of main branches of equal size.
    In problem solving however, I often want to develop only one or two branches over many steps, and this leads to an unsymmetrical, unbalanced mind map.
  • Moving from one box to the next needs a moment of orientation - where do I stand, and what can I do next? This reflection and re-orientation happens much more often in the box layout than in an ordinary mind map.
  • I find it easier to reflect on a previous thought in a new box than in a branch of an ordinary mind map.
  • The combination of writing structured by layout and thinking tools seems very powerful to me.

Thinking Processes
With these layouts, I can use thinking processes like the following. In each of the following examples, I can get support from the thinking tool collection on the corkboard at my writing desk.

  • Process 1: To generate ideas, I use the 3x4 layout and write an idea stimulus in each box. As stimuli, I can use the names of different sciences, the verbs from the SCAMPER acronym (substitute, combine, adapt, maximize & minimize, put to other usess, eliminate, rearrange) or some inventive principle from the TRIZ method, or mere brainstorming. Then, I fill the box with ideas that emerge from connections between my topic and the stimulus in that box.
  • Process 2: To analyse a situation, I describe a relevant situation from my topic in the first box, in the next box I look for problems or I collect questions. I can write down ideas for solutions and answers in the next box etc.
  • Process 3: To evolve a model, I can start with a simple model of the topic I’m dealing with, and then I can collect ideas and design different and better models.

Some remarks

  • Depending on handwriting size, paper size and personal taste, we can experiment with different layouts - 2x2, 3x3 or 4x4 boxes, or 2, 3 or 4 columns with boxes of varied size.
  • Obviously, the summary sheets are not atomic, and a large portion of the material on the WIP sheets is redundant once it has been summarized.
    In this respect, the combination of building blocks 1 to 4 can be viewed as a kind of “pre-substrate” to develop ideas that can later be processed into notes for a more conventional ZK.