The argument mapping community generally deems the first occurrence of argument mapping to be in Richard Whately’s Elements of Logic textbook, first published in 1826. (See e.g. Reed, C., Walton, D. & Macagno, F. (2007) “Argument diagramming in logic, law and artificial intelligence”, Knowledge Engineering Review, 22 (1), pp87-109; p.93.)
To me it is implausible that there are no earlier examples, but I cannot point to any, despite having “kept an eye out” for a number of years.
Here, for the record, is that earliest occurrence:
This is the only diagram in Whately’s Elements of Logic which is close to an argument map. Note that it is an argument map schema or template rather than an argument map. Whately’s comment “Many students probably will find it a very clear and convenient mode of exhibiting the logical analysis of a course of argument, to draw it out in the form of a Tree, or Logical Division” suggests that the practice of argument mapping did exist at that time.
Tim van Gelder 
Tim,
This is very interesting but as with most things it depends how tightly or loosely you define it. My take on this issue is somewhat looser seeing the essence of an argument map is a generalized reasoning procedure with a graphic (memory) component. I write memory component because this my sense of what graphic do – serve memory.
Unfortunatley, I don’t have time to fully research this, but you might check out some of the works of Raymundus Lullus (1315) who used lists, graphics and other memory tools, and rules to draw conclusions. His tact is different because he uses combinatorics (often attempting full enumeration).
Its eqully fascinating that Fritz Zwicky (1898) developed somewhat similar methods to sucessfully investigate numerous astronomical phenomnon. Both systems seem to run on generalized priciples of generation and elimination.
While it seems distant now, massive combinatorial arguments may be something of the future. What is also pertinent is the role of memory and the way these systems push inquiry in new directions and appear equally applicable in hard sciences and the humanities.
Circolo
Thanks Circolo. You’re correct of course in the general observation about breadth of definition. The way I think of it, an argument map has to show inferential or evidential relationships between propositions. There have been all sorts of diagrams used to good effect in logic or reasoning (e.g. Venn diagrams) but these don’t have nodes corresponding to whole propositions and links representing relations between propositions.
Years ago I read Gardner’s Logic Machines and Diagrams which has an interesting chapter on Lullus (Lull?). Still have a copy stored away somewhere. From memory, Lullus does not do argument mapping in my sense. Sounds like Zwicky also is doing something different, though I haven’t seen that stuff.
How do you show inferential or evidential relationships without memory?
Tim
You said earlier:
Whately’s comment “Many students probably will find it a very clear and convenient mode of exhibiting the logical analysis of a course of argument, to draw it out in the form of a Tree, or Logical Division” suggests that the practice of argument mapping did exist at that time.
Another alternative, at least equally plausible, is that the students will find it very clear and convenient (note – not “will be familiar with”) because it is similar to something else they were familiar with.
One candidate is the “form of a tree” or “Logical Division” referred to in the quote from Whately in Wigmore’s book. There is a striking resemblance to the diagram shown in Whately’s book and the cladograms used in biological classification. (Wikipedia: Cladistics is the hierarchical classification of species based on evolutionary ancestry).
“Cladograms” have been used for classifying things other than species. In “Classification, evolution, and the nature of biology” by Alec L. Panchen “the earliest recorded “cladogram” from Plato’s Sophist” is reproduced (at page 16) ”classifying the different methods of catching fish and putting it in the context of other human activity. The “Tree of Porphyry” is also shown in Panchen’s book (at page 19).
Given that Whately was a logician and the traditional format of the arguments envisaged and illustrated (syllogisms are referred to in the Hinds book Whately refers to) it seems reasonable that the students would be familiar with these traditional forms of representation.
The parts of Panchen’s book referred to are accessible through Google books. The link is shown below. Alternatively a Google search on “Logical division trees’ should bring it up as the first item.
http://books.google.com/books?id=pMCh0vCuAD0C&pg=PA16&lpg=PA16&dq=logical+division+trees&source=bl&ots=7lY72TQtLi&sig=giPkvQft3RIXvVuRbkc6mBmqb-E&hl=en&ei=EoGwSb2TDor2sAPXodiHAQ&sa=X&oi=book_result&resnum=1&ct=result#PPA19,M1
John, very interesting. What you suggest does seem plausible. However it is quite compatible with their being a practice, however nascent, of argument mapping around the time of Whately’s book appeared. Presumably Whately himself, and his students, and some of those who read his textbook, followed his recommendation at least occasionally. This practice may well, as you suggest, have drawn directly on their familiarity with logical divisions in other contexts.