Dear Viveka,

I had decided to not continue to delve into this subject on-list, but your remarks made me feel like I should clarify the following.

For starters, rhetoric has no place in logic.
Rhetoric makes use of logic, as of emotion.
That doesn't mean that there is a "place of rhetoric in logic".
Actually, it is the other way around.
Believing it is otherwise is a categorization error.
Rhetoric is the use of emotion and logic in an argument with the purpose of persuasion.
Take the persuasion away, and you're left with something that is NOT rhetoric.

Talking about a "place of rhetoric in logic" is tantamount to talking about a "place of carpenters in the forest": carpenters use wood, but that doesn't mean that the forest has anything to do with carpenters. If all the carpenters in the world would die in this exact moment, no single tree in the forest would miss any of them. Similarly, if rhetoric is completely banned and never used again, logic and emotion will still exist. The opposite does not hold: if logic and/or emotion were to disappear, rhetoric would be impossible. Hence, there is no "place in logic" for rhetoric, but there the other way around.
There is an old portuguese saying that goes:
"O que .ANi que o cNz tem a ver com as calNgas?"
It translates to:
"What has the ass to do with the trousers?"
It is like the "comparing apples and oranges" adage.
It means that, despite the fact that you wear your trousers to conceal those body parts, your body is categorically unrelated with any specific garment. You could live a perfectly happy life without trousers.
This is something that is so transparent to me, that I find it amazing that some people still struggle with this kind of problem.

Your comment (as some of Susan's reactions) suffer from the same issues your machine learning software suffers: categorization errors. This is because you are picking up on a word ("rhetoric", or "emotion", or "logic", or "trousers", or whatever) instead of grasping the whole meaning of the whole argument. That is why my first reply to Susan complained about "shallow" interpretations.
Free association exercises are all fine and dandy for the chaise longue or for a lazy afternoon outdoors while having a marguerita with your friends, but I do not find that free association is helpful in a thread in a research discussion list. That is why I kept babbling about "focus" and "objectivity" on previous replies.

Comments suffering from categorization errors can only be valuable in a creative way. If we are trying to find new ways of looking at some subject, jumping from one category to another is one way of creating new ideas and discovering new connections. However, if we are having a "convergent" discussion on one particular subject, comments with categorization errors just generate noise.

As for emotion in decision making, that is old news.
I deeply appreciate my fellow countryman Damasio's contribution (and in an emotional and non-logic way am proud that he is my countryman), and I enjoy very much his work, but my "hero" in this matter is Oliver Sacks (who unfortunately passed away recently).

I should probably stress that I never made any claim that "decisions can be made entirely logically". Unfortunately, misrepresentation of one's words is something common everywhere. Still, I always find it annoying, and I find it particularly disappointing in a research discussion list, where people are supposed to have a bit more focus and objectivity.

I will add that I share both your occupations, as I am a designer and a software developer.
However, I find situated cognition a better starting point than machine learning.
The limitations you refer to are not due to the roots of machine learning in logic, but to the fact that machine learning techniques are fundamentally flawed (as compared to actual human learning), as you will probably concur, in the light of the precepts of situated cognition.


Best regards,

==================================
Carlos Pires

[log in to unmask]
[log in to unmask]
-------------------------------------------------------------
Design & New Media MFA // Communication Design PhD Student @ FBA-UL

Check the project blog:
http://thegolemproject.com

On 16/05/2015, at 01:42, Viveka Weiley wrote:

> I'm enjoying these threads about the place of rhetoric in logic, and of emotion in decision making.
> 
> I'm a computer scientist as well as a designer, so I work across these fields. I've built and deployed decision support systems (note that we don't call them decision-making systems). Right now I'm working on a machine learning system using natural language processing techniques to extract meaning from news articles and find relationships between them.
> 
> These techniques are wonderful and growing more powerful exponentially. But to the extent that they are purely logical they have some fundamental limitations.
> 
> The idea that decisions can be made entirely logically goes against all available evidence: without an emotional basis there is no reason to prefer one decision over another. We know this for sure by observing the behaviour of people with very specific neural impairments to their emotional faculties, but who are otherwise intelligent and aware. 
> 
> Here is perhaps the most widely cited paper on this effect:
> 
> http://tamar.tau.ac.il/Genome2Brain2008/Human%20Brain/Damasio%20Emotions%20and%20Decision%20Making.pdf
> 
> The people in this study became "unable to decide advantageously on matters pertaining to their own lives", despite maintaining intellectual capacity. Without emotion they quite literally had no reason to care. Further: they didn't just start deciding arbitrarily or according to some logical set of rules: instead they became unable to make decisions at all. 
> 
> Logic and emotion are not set against each other. To create meaning, we need both.
> 
> Cheers,
> 
> Viveka Weiley
> Creative Director of R&D, Australian Broadcasting Corporation
> PhD Candidate at the Creativity and Cognition Studios, University of Technology, Sydney.



-----------------------------------------------------------------
PhD-Design mailing list  <[log in to unmask]>
Discussion of PhD studies and related research in Design
Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design
-----------------------------------------------------------------