How much of foundational epistemology does simply starting with induction solve? All of it?
If you start with induction...and conclude (very quickly) that deduction predicts well inside math, and rarely predicts outside of math...don't you just win out real fast?
You give up entirely on the notion of certainty...and you adhere pretty fast to the "map, not territory" model.
The scientific method falls out real fast as a how to purchase information from reality using induction .
Lean/iterative processes fall out almost as fast as you look for improvement methods.
I'm thinking that this is, while not necessarily shocking or new, the BIG win for epistemology.
Evolution cycles back around and supports the notion of the inductive mind...
Subscribe to:
Post Comments (Atom)
Posts
8 comments:
I casually suggest you think about the acquisition of concepts. Or substitute out 'concepts' and put in your favourite flavour of fundamental unit of thought.
How does one learn to induce?
How does one learn to deduce?
How does one learn that such a thing as 'two' exists to deduce from?
How does one learn to learn? How does one learn to observe?
I've been studying early childhood acquisition of concepts and philosophical opinions of where concepts come from for 20 years. Where do you think I get these ideas. As my first love is learning/teaching math, this is especially relevant.
The first learning done by infants is trial/error induction.
Learning to deduce is done by noting common rules that work. By induction, deduction works.
"Two" doesn't "exist". Two is an abstraction that is learned two different ways separately: count and extent...and then attached to one another. Check out the Montessori exploration of that...golden beads and long rods.
One does not learn to learn...it's built in to the brain.
One also has focus interests built into the brain when born. Faces are high priority...and mom's face especially. Boys have higher priority on objects, girls have higher priority on other faces. Minutes after birth.
The concept of 'two' exists.
So the genes learned to learn, and in turn taught the brain.
My concept of two exists in some sense of the word "exists" in my "mind".
A different concept of two exists in some sense of the word exists in your mind.
A very different concept of two exists in some sense in my 6yo's mind...and a different one existed in Plato's mind (No modern algebra).
I agree that I mean something by "two", and that you mean something by "two", and that on first attempt, we will agree on almost all of our referents.
Is that what you mean by the concept existing?
Yes, that's a better way to put it. A concept of 'two' exists. It has to simply because it takes energy to learn, which means the brain increases in mass. So 'two' has mass.
So it's obvious to you that the gene pool functions as an AI? That taught itself to learn?
I keep being rewarded by these discussions because they better reveal to myself what I'm thinking.
I believe all steps of inductions are performed through deduction. There are three; abstraction, comparison, and extrapolation.
The only thing separate is the actual observations themselves, which are performed entirely through billiard-ball physics, Light hits the eye, eye pokes brain, neurons change shape, hwonk. Later, hwonks can be compared to each other. Hwonk includes the abstraction step, as some post-processing occurs before the recording.
If Ye Olde philosophers realized the steps of induction all form deduction, that would explain why they had issues with the relationship between the two.
Alrenous,
One of my dearest friends is an animal-learning professor, studying insect learning (crickets and honeybees) for the last 10-15 years.
One of my the few people I have met who clearly understands (parts of) education FAR better than I do is an intellectual heir to Maria Montessori. He inherited much of her original materials when her heirs died.
It is, to me, very obvious that inductive learning is pre-deduction, pre-consciousness, which I suppose means that it MUST be genetically coded.
Who learns: Crickets, Honeybees, 15 minute old babies who have ONLY the proto-concepts of good and bad, and are working to, sometime in the next 6 months, induct their way into the distinction between self and not-self.
Abstraction is an unnecessary step in induction.
The mind (as opposed to the visual cortex) perceives objects in the world, which is part of what makes art so hard...one has to half-unlearn the perception of objects...because what your mind sees and what your eyes see don't match.
I was chasing a different direction...but let me instead ask:
What gives you reason to believe that inductions come from deduction? It's fairly obvious that all creatures with brains induct...which means that under any reasonable start point, induction has priority.
A. Why (what purpose is served) are you sticking deduction in the middle?
B. Why (what evidence suggests) do you believe that deduction happens first?
B2. Specifically, doesn't Ockham point out that all available evidence is handled simply, and elegantly by induction first approaches?
Induction does not come from deduction, it is constituted by a series of deductive steps.
B2. Certainly, induction comes very early, but if you break induction down into its physical actions, they are all implicit deductions, except the observation step.
A. If I wanted to build an induction machine, what would I build it to do? Observe, compare, extrapolate. Steps 2 & 3 are [A=A?] and [if_observe(A) then(B)]. Deductive logic.
B1. If you can think of some other mechanism with which to carry out induction, I would likely change my mind. However, it seems significant to me that induction can be carried out through deduction.
C. What I think actually comes first is trial and error. So what constitutes trial and error?
Some success condition.
A method of creating data/[a design] and implementing it.
Data->implementation is also deduction. But in the case of genetics, that's constituted by physics. So physics give you deduction for free, which gives you genetic induction, which gives you neuronal deduction, which gives you induction again.
And now I've stopped seeing a point in making a distinction between the two at all.
You don't need consciousness to deduce. Just need a conclusion from premises.
The reason I include an abstraction step is that the comparison step only compares goal-meaningful properties. If you try to directly compare unabstracted data, you get almost no matches, and in any case the raw data from the eye is not recorded.
Which is, incidentally, why re-creating what the brain sees and running it through the eye again creates a new image instead of a copy, as opposed to doing the same thing with a high-fi camera.
Post a Comment