# Tree proofs

The task is to set proof-trees in the style of the tree alongside, i.e. tableaux of the sort that appear in e.g. Richard Jeffrey’s classic Formal Logic or the first edition of my own Introduction to Formal Logic.

Despite the very wide use of such tableaux, there has only lately appeared a purpose-designed package which looks excellent and very flexible, allowing you also to add line numbers (if you want) and line-by-line justifications:

However, because of issues with the packages that prooftrees.sty invokes, a document with many trees does become slow to type set.

Older options for setting tableaux often require bits of trickery to get nice-looking trees (and don’t supply the option of line numbers etc.).  I’ll mention them, however, in case they may be enough for your purposes. Two first options are

• The qtree.sty package (Jeff Siskind and Alexis Dimitriadis), written for typesetting linguists’ syntactic trees, can also be adapted to logicians’ purposes.
• pgf and TikZ (Till Tantau 2005-: a general TeX macro package for generating graphics, with a user-friendly syntax layer called TikZ). tikz-qtree provides a macro for drawing trees with TikZ using the easy syntax of qtree.
• For a document with some illustrations of trees set with these, see Proofs in LaTeX (Alex Kocurek, 2017).

Further options include

• The pst-tree package, used for the illustration above. This package is part of the powerful pstricks family of packages. But this can be a little cumbersome to use, e.g. to fine-tune the placing of the diagonal lines.
• Another package originally written for linguists is synttree, which has a pleasing simple syntax in use (Matijs van Zuijlen).
• Then there is forest (Sašo Ži­vanović, 2012–2017). This is another package that  pro­vides PGF/TikZ-based apparatus for draw­ing trees, and looks very flexible. There is a nice worked example in the first answer here which shows how to fine-tune the output.

Clea Rees’s prooftrees package is based on forest. Here’s a fully annotated tree produced by this package:

See my document for more examples.

Updated 16 December 2018

### 5 thoughts on “Tree proofs”

1. Is anybody else able to get prooftrees to compile on the 2016 tex distribution? I was able to get it to work once I updated to the 2017 distribution but unfortunately I’m having difficulties submitting the paper to ArXiv as they’re still running the 2016 distribution. Is there any fix for this? Otherwise, great package!

1. As Peter Smith indicates, this is probably due to changes in prooftrees and/or forest, though it is just possible it could be pgf/tikz related. In general, forest and prooftrees have been updated together or only a few days apart, at most, so an updated TeX Live should (hopefully) include a version of prooftrees compatible with the installed version of forest. However, if you try to combine a newer prooftrees with an older forest, it probably won’t work. Assuming that ArXiv has prooftrees and forest installed, they should be compatible. But you might need to adjust the code for your trees to avoid syntax not supported by older versions. But without knowing the error and without any idea what’s in your code, it is hard to give more specific advice.

2. Note that prooftrees CAN now be used with bussproofs. Either load bussproofs first. Or load prooftrees with option ‘tableaux’: ‘\usepackage[tableaux]{prooftrees}’. Either way, you can then use ‘prooftree’ for bussproofs trees and ‘tableau’ for prooftrees trees. Take care not to load the tableaux package instead which, as far as I can tell, does something completely different. The tableaux package is loaded with ‘\usepackage{tableau}`. However, should you need bussproofs, tableaux and prooftrees in the same document, I *think* it should still all play nicely together, as tableaux does not seem to define a ‘tableau’ environment, but only provides a ‘tableau’ .sty file. It would help a bit if package authors didn’t commandeer not one, but multiple names – if bussproofs provided ‘bussproof’, say! In the beginning, package authors believed names would always be plentiful and life simple and good. By now, of course, life is no longer so simple.