Peter Tillers raised an important issue on the Rationale google group:
“I will have to figure out how best to use ‘Rationale’ to assess the
occurrence or non-occurrence of “composite events,” e.g., an event X
consisting of events a, b + c. This is because in law the factual
events in question in adjudication and investigation are invariably
such composite events; e.g., Did the Defendant (a) deliberately (b)
cause (d) the death of (e) another person… Incidentally, almost all
work on charting argument about or analysis
of factual questions in litigation assumes that the occurrence or
character of but a single event is in question. But that assumption is
incorrect. See P. Tillers, “Probability and Uncertainty in Law,”
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=738764″
I replied as follows:
Peter’s note concerns what lawyers call the elements of the crime of murder. Murder is a compound event. One way to specify the elements is, as Peter does, by listing the parts of speech which occur in a succinct sentence describing the compound event, i.e. the defendant
– deliberately
– caused
– the death of
– another person
Each of these however is shorthand for a component or aspect of the compound event. Thus “caused” is shorthand for the proposition
The defendant caused the death of the victim.
Note that this proposition takes for granted another proposition
The victim died.
and, if the charge of murder is to be sustained, another proposition must be true:
The victim was a person.
Since these propositions must all be true together, in Rationale we’d use an “Analysis” mode map:
The disadvantage of such a representation is that it is wordy, especially as compared with the shorthand approach mentioned above. However there are corresponding advantages. In particular it more accurately represents what the prosecution will, in fact, have to establish – i.e., the truth of a series of distinct but interrelated propositions. It also represents them in a more logically appropriate order, i.e, getting the order of logical priority roughly right; there’s no point in proving that the defendant did it deliberately if the victim wasn’t even a person. (It is interesting that the natural order in an English sentence is almost the reverse of the order of logical dependency.)
Nevertheless, we’re interested in the idea that there might be some alternative diagramming format in software such as Rationale which provides a more succinct and perhaps in some ways more practically “useable” way of representing these sorts of top-level legal cases.
Tim van Gelder 
On a tangent – this sort of legal situation often gives rise to mutually exclusive reasons (or objections). For example, I might object that X did not actually die, or I might object that X’s death was not the defendant’s fault. Either of these objections works on its own. The trouble comes when you put the two together: one assumes X’s death, the other denies it. At best the objections are no stronger than either one alone, while at worst they are weaker because of the inconsistency.
I remember this being invisible in Reason!Able. Can Rationale highlight these situations, or deal with them somehow?
Yes, this is an interesting issue. Rationale, like Reason!Able, has no particular way of handling these sorts of conflicts. I’m still undecided as to whether the absence of any such method is a problem.