Jan Aldert Bergstra
Utrecht, March 2011
Recently I have posted a paper [1] titled “Real Islamic Logic” (Hereafter RIL) on the open access repository www.arXiv.org. Rather than to repeat the gist of that work here, which adds nothing to what can be obtained by downloading and reading the mentioned work, I will formulate some critique that I was unable or rather unwilling to include in the paper. Given its speculative nature, many weaknesses can be spotted in [1], but in several cases a weakness can be removed simply by removing the section in which it occurs, thus shortening and also simplifying the work, which may even be considered an improvement. I will call this the drastic removal strategy, because a problem is eliminated together with the entire context in which it appears. This strategy for instance applies to the range of arguments used in the discussion of the existence of God, and to the limited substantiation made of the claim that the Islamic Court Legal Proposition Process (ICLPP) is computationally intractable.
The drastic removal strategy is not applicable if it comes to the portrayal of logic as being important for organized and methodical human reasoning. Although a complex logical setting is suggested, RIL is based on the assumption that a well-worked out and formalized reasoning methodology is helpful for humans deliberating alone or in groups about moral issues. It should be noticed that RIL is supposed to be instrumental for reasoning processes outside formalized courts. Efficiency is important in real time applications (where court referral is practically impossible) and for limiting the case load of courts (which is a necessity if their judgments are to be valuable).
That very assumption is hardly supported by experimental work in cognitive psychology. I mention the PhD thesis of Marian Counihan [2] and the vast bibliography contained in that work as a recent and convincing meta-study of the problems one encounters when logic is supposed to be of an explanatory (or normative) value for human thinking.
Four different perceptions of logic come to mind in the context of RIL;
A) RIL supports those humans who became logical experts through detailed personal studies. It is no more than a tool for those who like it, assessments of its use are private. No one can (will or ought) to be forced Into “logic”. (Question: can a useful tool be effectively designed.)
B) RIL explains how reasoning by citizen contemplating whether or not to have the issue rerouted toward a formalized court. The evidence, gained in a wealth of investigations thoroughly discussed in [2] strongly suggests that this hope is vacuous unless
C) RIL (once developed) will obtain normative status: only by consciously and explicitly reasoning according to patterns confirmed by RIL, citizens can be sure not to make reasoning errors for which the may subsequently be held accountable.
D) RIL is useful for developing an overall picture of appropriate decision making, thus leading to a data base of past judgments which can be used by decision makers via association based search engines.
If RIL ever reaches the maturity sketched in the paper, it should work like the following combined scenario:
- Systematic attempts to support persons in different circumstances with extra-court reasoning lead to a toolkit that may support at least those who had an appropriate training, thus validating scenario (A), and
- support may in part be delivered by means of the tools mentioned in (D).
- Then scenario (B) is ignored because focus is on a methodical training of all citizens intending to apply RIL based reasoning effectively and for its intended purposes.
- The resulting state complies with (C) and RIL, once mature, has obtained a normative status.
This description does not depend on a realization of scenario (B) which has almost been ruled out by past research as reported in [2]. Nevertheless the combined scenario may be considered speculative to such an extent that this renders the RIL idea fruitless.
I am optimistic about the future of applied logic so I don’t think that working towards RIL is necessarily a fruitless endeavor. If, however, one intends to modify the RIL project so as to be made independent of an optimistic view of the future of applied logic, the project can be renamed into RIR (Real Islamic Reasoning), where Reasoning builds on a far larger body of psychological knowledge than logic traditionally does. Statistical methods, extensive data mining, pattern matching, results from neuro-imaging, and applications of social computing, can all be brought together to constitute a package of techniques which may replace the bundle of methods which I have listed in order to specify what RIL might take on board. In this way a hypothetical transformation of the RIL paper into an (equally hypothetical) RIR paper can be undertaken, while leaving most of its embedding in a context concerning the Islamic Legal Process invariant.
[1] Jan Aldert Bergstra, Real Islamic Logic, arXiv:1103.4515v1 [cs.LO] (2011), (http://www.arXiv.org/abs/1103.4515).
[2] Marian Counihan, Looking for logic in all the wrong places: an investigation of language, literacy and logic in reasoning. Ph D Thesis, University of Amsterdam, ILLC Dissertation Series DS-2008-10 (2008)
