From LOJBAN%CUVMB.BITNET@UBVM.CC.BUFFALO.EDU Sat Feb 10 11:05:06 1996 Received: from VMS.DC.LSOFT.COM (vms.dc.lsoft.com [205.186.43.2]) by locke.ccil.org (8.6.9/8.6.10) with ESMTP id LAA12007 for ; Sat, 10 Feb 1996 11:05:02 -0500 Message-Id: <199602101605.LAA12007@locke.ccil.org> Received: from PEACH.EASE.LSOFT.COM (205.186.43.4) by VMS.DC.LSOFT.COM (LSMTP for OpenVMS v1.0a) with SMTP id 8ACDF042 ; Sat, 10 Feb 1996 10:32:29 -0500 Date: Sat, 10 Feb 1996 10:31:56 -0500 Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: *old response to Steven B #2 X-To: lojban@cuvmb.cc.columbia.edu To: John Cowan Status: OR X-Mozilla-Status: 0001 Content-Length: 6390 >>I live in constant low level pain of several >>varieties that I mostly associate with stress. > >Sorry to hear that. I would think that this would make stressful >situations doubly unpleasant. I retract my suggestion that you >quadruple the size of the dictionary to include a scale specific >definition for every gismu. Of course I also weigh 380 pounds, again mostly from stress eating, which is perhaps a more direct explanation for my pain. (Interestingly, I had the stomach flu a week ago, lost 10 lbs in 2 days due to diarrhea and most of my pain symptoms also disappeared thereafter as well, and have not yet returned. My wife jokingly says that this is because after the way my stomach felt for 2 days, the pain is no longer registered on a noticeable scale. This is probably not the case - I just am not experiencing as much pain. But if it were true, it could be seen as evidence that the subjective pain scales are not trustable over time. I am not sure I could compare my stomach cramps of last week with stomach cramps I had as a kid, or even with muscle cramps I had 6 months ago after a heavy exercise session. And in any event, there is no way that you or I or anyone could come up with a meaningful ordinal scale to compare my stomach cramps when I had the flu with the cramps that Nora or my son Avgust had 2 days later from presumably the same illness. >Yet, meaningful, useful conversation can occur about this experience. >Dismissing all this stuff as "subjective" would be a disservice to many >of my patients, and would leave many of them in more pain than >necessary. I don't think anyone disagrees with this. But expressing pain on a scale predated fuzzy logic as a field of study. The issue is whether fuzzy logic principles are well enough understood that a specific linguistic solution to expressing them should be hard-coded into the language. I am inclined to think that a) the expression of fuzzy logic and other scales linguistically in various languages is NOT well studied, and b) the whole subject is sufficiently controversial that we want to tread carefully. We have certainly built a lot of varieties of scalar expression into the language, and the raw linguistic materials exist to construct new ones as the need arises. I don't want to favor a particular approach in the design prescription, because, among other things, it would require us to teach fuzzy logic, and that particular school of fuzzy logic in the refgrammar and textbook. >>I think that people BELIEVE they have to categorize and cubbyhole in >>order to be "objective". Not being categorical has earned the label >>"slippery slope" for good reason. > >Insisting on a nominal or categorical scale has earned the label >"pedantic" for good reason! > >Many people are falsely taught that Aristotlean two-valued logic is the >"only" correct logic. Perhaps it would be better to label this approach >as Zoroastrianism, as Zoroaster's beliefs, as described in the >Zend-Avesta, seem to have influenced many later belief systems, and >Aristotle did a lot of other things besides introducing the law of the >excluded middle. I believe this misinformed insistence on two-valued >logics does much harm. Lojban does not particularly favor two-valued logic, except perhaps in its logical connectives. The explicit distinction between scalar and predicate negation alone removes a key impediment to fuzzy logic since the scalar je'a/no'e/na'e - applicable to ALL predicates - alone establishes an at-least-tri-valued logic. The use of the 7-valued ju'o scale and the 7-valued la'a scale, and the as yet unexplored cu'o selbri unit provide us with so many resources we really need to see what can be done with what we have before we decide we need more (and don't forget that on pain scales you also have .oi+ro'V+intensifier - a total of 6 7-valued scales of pain to play with,not to mention the other attitudinals - this could be far more important to diagnostics because we explicitly recognize that pain has emotive aspects to it independent of the physical ones - something I think that has become recently accepted, but certainly is not built into any language before Lojban.) >I think that the logic boxes in our heads are fuzzy logic boxes, and >that the tuning begins at birth. We can output two-valued logic, if >that's what is required, but again, the output of a logic does not >specify the algorithm used to calculate that output. Granted that litle >is known about the basis of consciousness, but if brains have discrete >states, why do they seem so analog in function? The answer to this is that, if our brains are so fuzzy, why didn't language evolve to more closely reflect that fuzziness. Languages are GOOD at expressing definiteness and poor at expressing fuzziness of categories. Language evolved with the species, and its nature has not substantially changed since or as a result of Aristotle, so far as I know. The answer seems to me that our brains MIGHT be scalar-analog, which is why scalar negation does show up in language, even though it gets the added baggage of contradictory and metalinguistic negation tossed on top of it. Lojban already separates these. >I think much of the argument results from believing that one's preferred >mode of thinking is the only valid mode of thinking. Just cause Frodo >the Hobbit can't wield a greatsword doesn't mean greatswords are not >useful weapons. (I am a great admirer of Frodo, however. He has other >skills.) I never got that from Peter's comments or anyone else's. I heard specific complaints about some of your examples, and a general skepticism that the problem is well-enough understood to choose a specific approach, and perhaps a fear that building in fuzzy logic too explicitly would remove the possibility of 2-valued logic (which would not happen). I don't think anyone really denied that you got benefit from your pain scales, for example - the question was whether those scales had any objective truth or repeatability across multiple observers. Most scientists tend to be skeptical of any variety of "truth" that is observer-dependent or context-dependent, and all of your examples involved very subjective (if real) observations/"truths". lojbab