The Tricky Truth about Tractatus Trees (updated)

The Tricky Truth about Tractatus Trees (updated)

Earlier in the week I put up a website that allows one to click through the tree-like structure of Wittgenstein’s Tractatus Logico-Philosophicus in the Heap of Links (in the right sidebar). I believe the visualization is by Pierre Bellon, a web engineer who has “old-school philosophy” as a hobby.

In response, David Stern (Iowa), sent in this helpful message:

That’s a nice-looking website, but it runs into some problems when you look at the details.  In order to handle the relationship between proposition 3 and 3.001, its layout requires a node called “3.0” in between the two (which does not exist in the book’s numbering system).  And there is no “0” in the TLP numbering system, yet for some reason the site presents “0” as the base of the entire tree.

Stern is currently teaching a graduate seminar on the topic of approaching the Tractatus as a logical tree (!). He continues:

I’ve been giving this a lot of thought: the question of how to handle propositions such as 3.001 (and 3.0001) turns out to be a key issue for a systematic account of how the tree structure works. Luciano Bazzocchi has been publishing some very interesting work on this topic for some time–he makes a very strong case that Wittgenstein used the tree structure to compose the book out of his notebooks, and that it explains why Wittgenstein thought the numbering system, which no-one really understood until very recently, was so important.

So, at Professor Stern’s suggestion, here is Bazzocchi’s depiction of the tree-like structure of the Tractatus.

If you know of others, feel free to share them in the comments.

UPDATE: A further update from Professor Stern:

Actually, there are a surprising number of Tractatus tree/hypertext sites.  Two others, each slightly different, are:

There’s also Kevin Klement’s beautiful side-by-side-by-side edition, in various formats.

(photo by JW)

stay back

Horizons Sustainable Financial Services
Notify of

Newest Most Voted
Inline Feedbacks
View all comments
7 years ago

I feel like the Tractatus needs a planetoid model for visualization. Each major proposition is like a planet, with the sub-propositions as the moons. The 1.000x statements are small moons off the major planet 1, while the 1.x are large moons, possibly with their own moons 1.x1, 1.x2, etc.

7 years ago

Hello, I’m the coder of the Tractatus tree and I would like to react to this article.

First of all, thanks for this feedback, I thought this work would go completely unnoticed forever :D. I’m aware of many flaws on this visualization and it’s far from what I’ve imagined but I sort-of achieved my goal because I wanted to have a visual representation of the tractatus’ tree but I know it’s buggy and less helpful from what I’ve imagined.
However, the proposition “0” issue is annoying but it’s also really helpful to keep the tree less crowded by nodes. For instance if I open node 2. it should show 8 nodes instead of 3.
Also, I’m seeing a similar solution used on Bazzochi’s represention. He used 2.00 instead of 2.0 but the idea is the same: having a parent node to these “2.0x” nodes.

But one thing that need to be fixed is the first ‘0’ node that don’t make any sense, it should be a single point without any number representing the root of the tree.
I will try to make this viz a bit more helpful and I might change back the node’s notation system to the notation used by Wittgenstein.

John Wade
John Wade
4 years ago

A logical ordering of the propositions with Wittgenstein’s original symbols and images is now available here:
It uses dropdown boxes, so that previous nodes can be easily reviewed.

Jean Marie Clarke
Jean Marie Clarke
3 years ago

I am not a trained, academic philosopher, but fascinated by the challenge(s) that the Tractatus represents. For one thing, it might be looked at like a musical composition with themes, developments and variations: I compose myself and know that thinking in terms of sounds (sound patterns) is analogous to thinking in terms of words (ideas…). In both cases, there is composition going on, if a “work” is to result. To address one special point in the discussion, the “O” has its place in the tree map because of Wittgenstein’s own admission that what he did not say/write in the Tractatus was actually more important than what he said. This is a major point, since it means that the Tractatus represents an effort to create a whole (model), that the “Unsaid” exerted as it were a gravitational pull on his thinking throughout.