Thursday, June 22, 2017

Evidence vs. Proof

A piece of evidence is an item that, taken by itself, is more likely to exist if the hypothesis is true, than if it is not true. As I see it, there are piece of evidence on both sides of the God question. People sometimes confuse evidence with proof. Proof actually demonstrates that something is true, evidence just, by itself, supports the claim. As I see it, there is evidence for a lot of things that are really false, and for which the preponderance of the evidence supports its denial. 

27 comments:

Joe Hinman said...

"As I see it, there is evidence for a lot of things that are really false, and for which the preponderance of the evidence supports its denial." (VR)

Yes I agree with that. I do not argue to prove the existence of god bit to warrant belief. I my concept warrant supports believe when the preponderance of the evidence supports belief. Preponderance is not just the number of pieces of evidence but quality as well.

A God argument is like a mini thesis. It's not just one single idea but a confluence of several trains of thought,all backed by evidence, converging on this single idea or point.

David Brightly said...

I can agree that car won't start is evidence for battery flat, since p(car won't start|battery flat)=1 and p(car won't start|battery charged)~0 and these probabilities can be backed up from the statistics gathered by roadside assistance companies. But what are we to make of p(consciousness exists|God exists), say, when one or more of the 'events' is metaphysical? If we are to interpret probability as degree of belief or warrant doesn't the whole business become circular---we are explaining 'evidentiality', ie, degree of belief, in terms of degree of belief?

Joe Hinman said...

But what are we to make of p(consciousness exists|God exists), say, when one or more of the 'events' is metaphysical? If we are to interpret probability as degree of belief or warrant doesn't the whole business become circular---we are explaining 'evidentiality', ie, degree of belief, in terms of degree of belief?

you are arguing fro analogy. You are asserting that flat battery is proven fact that can't be questioned but there are other from of argument basec warrant that you also agree to as though they are facts when they are not established by the same kind of certainty.


Secondly no one has rugged mind exists therefore God exists.Few more steps improved if you make an argument from consciousness,I didn't say that.

Joe Hinman said...

The argumet about car battery is the NOT same kind of argument as a God argument. With the car you are appealing to facts that as stipulated as such. T
he Latter is based upon inference from warrant with no attempt at real proof, While the former is not based upon warrant alone but upon assumed factial nature of the assumptions, ie car batteries and how they work,For practical purposes these are things we know.

"Factical" does not mean factual it means the ability to assert a factual nature to an issue. W
e have empirical references for car batters but not for God Yet that does not mean there is no good reason to believe in God it merely means there's no agreed upon factual assumption pertaining to God.

David Brightly said...

In the example of the car failing to start the inference to the battery's being flat is justified by objective probabilities based on the frequencies of different types of faults in vehicles. My question is, What justifies the inference---makes it more or less 'likely'---in the metaphysical example? Indeed, what does 'more or less likely' actually mean in this case?

Joe Hinman said...

I am going to debate Bradley Bowen of the secular out post on the existence of God, This blog post of mine answers his clarification questions interpretation for our debate.

Metacrock's blog

Joe Hinman said...

David Brightly said...
In the example of the car failing to start the inference to the battery's being flat is justified by objective probabilities based on the frequencies of different types of faults in vehicles. My question is, What justifies the inference---makes it more or less 'likely'---in the metaphysical example? Indeed, what does 'more or less likely' actually mean in this case?


I think you misdeed my point about the car thing.You can have empirical evidence of car batteries and you can get new empirical evidence all the time. You can't get empirical evidence of God because God is not given in sense data,you can't get new info about God,so these are fundamentally different kinds of questions.

Ron said...

Just because we can't immediately experience God with sense data, it doesn't follow that we can't have empirical evidence for/against God. We can't immedietly experience a lot of things with or sense data (Big Bang, electrons, Higgs boson, etc.) but we have empirical evidence for them because we can use our senses to detect their *effects*. So if God has effects in the world, we can detect their presence or absence with our senses. This seems uncontroversial to me

David Brightly said...

I agree that the physical question and the metaphysical question are fundamentally different. This difference opens the door to scepticism about evidence in the metaphysical case. Victor defines evidence in terms of likelihood. This is a probabilistic term, and I don't see how to get probability into metaphysics. Quick argument: probability applies only to contingencies. Metaphysics concerns the necessary.

Joe Hinman said...

David Brightly said...
I agree that the physical question and the metaphysical question are fundamentally different. This difference opens the door to scepticism about evidence in the metaphysical case. Victor defines evidence in terms of likelihood. This is a probabilistic term, and I don't see how to get probability into metaphysics. Quick argument: probability applies only to contingencies. Metaphysics concerns the necessary.

evidence for
God does not have to be metaphysical. we have executive and inudcutive and even abdictive arguments.

Joe Hinman said...

what that means is arguments that stem from analogies like the car battery carry no power agaisnt belief because they are comparing apples and oranges. the skeptic asserts that science is the only valid form of knowledge and doubing it metaphysical is a way to say it's not a valid knowledge,that is just an ideologiocalobjection.

Joe Hinman said...

Ron said...
Just because we can't immediately experience God with sense data, it doesn't follow that we can't have empirical evidence for/against God.

of course it does that's what empirical means.


We can't immedietly experience a lot of things with or sense data (Big Bang, electrons, Higgs boson, etc.) but we have empirical evidence for them because we can use our senses to detect their *effects*. So if God has effects in the world, we can detect their presence or absence with our senses. This seems uncontroversial to me

Yes we are really on the same page but expressing it differently. While we can't have direct empirical observation of God we can have empirical knowledge of the co-determinate or God correlate. That's the point of my title of my book The Trace of God, The trace is the God correlate the signature the foot print. trace is Derridian term meaning track or foot pint.

In the case of my God argument that is mystical experience, or Schleiermacher;s feeling of utter depemdeceisa another exmaple.

Ron said...

Ok so we basically agree, just use words differently. So would you say we don't have empirical evidence for electrons, black holes, but we do have traces of them?

Mortal said...

Quick argument: probability applies only to contingencies. Metaphysics concerns the necessary.

Thank you! Thank you! THANK YOU!!!

I've always felt that all these discussions about probability were inappropriate.

David Brightly said...

Let's go back to what Victor says: A piece of evidence is an item that, taken by itself, is more likely to exist if the hypothesis is true, than if it is not true. And let's rephrase the question: If the hypothesis is that of God's existence and the notion of likelihood is not the usual probabilistic one that applies to contingencies, then what exactly is this notion of likelihood that is constitutive of evidence?

Ron said...

Since when does metaphysics only concern necessities? Metaphysics deals with questions concerning space, time, free will, etc. Take free will for example. One popular view of free will is libertarian free will, which says that the choices we make are contingent matters of fact without antecedent causes. Whether or not this is true (I personally think LFW is false) it's a metaphysical hypothesis clearly addressing contingencies.

Furthermore, even necessary truths can have epistemically probabilities applied to them. For example, Goldbach's conjecture is either necessarily true or necessarily false, but we currently do not know if it is true. So mathematicians have varying degrees of confidence about whether it is true or false. Some might say it is near certain, others might says it is 50/50.

Joe Hinman said...

just use words differently. So would you say we don't have empirical evidence for electrons, black holes, but we do have traces of them?

for electrons yes. We might have actual empirical evidence of black holes i'm not sure.

Joe Hinman said...

David Brightly said...
Let's go back to what Victor says: A piece of evidence is an item that, taken by itself, is more likely to exist if the hypothesis is true, than if it is not true. And let's rephrase the question: If the hypothesis is that of God's existence and the notion of likelihood is not the usual probabilistic one that applies to contingencies, then what exactly is this notion of likelihood that is constitutive of evidence?

I don't agree with the statement. I think a piece of evidence is an item that taken by it;self may be meaningless or have no apparent relation to God's existence.

Joe Hinman said...

Ron said...
Since when does metaphysics only concern necessities? Metaphysics deals with questions concerning space, time, free will, etc. Take free will for example. One popular view of free will is libertarian free will, which says that the choices we make are contingent matters of fact without antecedent causes. Whether or not this is true (I personally think LFW is false) it's a metaphysical hypothesis clearly addressing contingencies.

for that matter the whole dichotomy of Necessity/contingency is metaphysical.So contingencies are as much metaphysical as are necessities. If you are Heideggerian even science is metaphysical.

People connected with internet apologetic have gotten into this notion that metaphysics means magic or God or supernatural. That;snot true.

David Brightly said...

An item's being meaningless in itself and having no apparent relation to the hypothesis is consistent with its being evidence for the hypothesis in Victor's sense. Example: that pesky light on the dashboard that we've never bothered with before. It's evidence for an engine management fault.

Metaphysics concerns the necessary because our only access to metaphysical conclusions, like mathematical results, is deduction from givens. No metaphysician would say that there is Form and Prime Matter in some possible world but not in some other, I think.

I'd still like to know more about what Victor means by 'more likely' when the hypothesis is metaphysical.

Ron said...

David,

What exactly do you mean by metaphysical? If you just mean supernatural, then supernatural hypotheses can definitely make probabilistic predictions. Consider the hypothesis: God created the universe 6,000 years ago. This entails that we should expect to observe certain things when we look at the distance between galaxies, radiometeric dating, fossils, etc. But what we observe about this things (call them X) is very unlikely on the assumption that the universe was created by God 6,000 years ago, and we can only salvage that hypothesis by inventing as hoc excuses. But X is exactly what we'd expect if that hypothesis is false. So P(X|God created 6,000 years ago)<P(X|~God created 6,000 years ago).

Or consider the metaphysical hypothesis: the universe was created by an omnipotent immaterial agent who wanted everything in the world to be purple. There is good evidence that this is false because most things aren't purple. P(most things aren't purple|purple God)<P(most things aren't purple|~purple God).

So we can are certainly apply probabilistic reasoning to hypotheses with supernatural or transcendent hypotheses that go beyond the bounds of what science normally addresses. Is this what you mean by metaphysical

Ron said...

Some metaphysical hypothesis are only knowable in that deductive a priori way (e.g. Platonism). But what about other metaphysical hypotheses that have empirical consequences (free will, young earth creationism). These can be assessed evidentially because they make empirical predictions about contingent matters of fact e.g., the age of the earth

David Brightly said...

Let’s put to one side the interesting question of characterising the metaphysical and concentrate on Victor’s claim about evidence in the metaphysical realm. We seem to be saying that the probability or likelihood in this case comes down to confidence or degree of belief or expectation. Victor’s characterisation of evidence is essentially a comparison of confidences. I worry about this. The idea of ‘evidence’, I would have thought, carries a normative element. Evidence, in some sense, tells us what we ought to believe, or at least it points us in a certain direction. But Victor’s characterisation is in terms of what we are already inclined to believe. Hence my earlier charge (second comment here) of circularity. Note that in the ordinary, practical, non-metaphysical case, such as in my example of the car that won’t start, the probabilities are objective ones not subjective ones, and Victor’s characterisation is properly grounded. My concern is that our reasoning in this case, which we can be confident about, doesn’t carry over into the metaphysical case.

Ron said...

You get a present. You don't know if it's from Tom or Bob. Both of these people are strangers, you don't know anything about either except that Tom hates you and Bob loves you. You open the package and it's a bomb. You survive. Is this evidence that it was from Tom? I think so. A person who hates you is more likely to send you a bomb than someone who loves you. And we can say this is evidence for the package being from Tom, even though we have zero experience with Tom and no objective frequency data on how Bob treats people he likes/hates and no objective frequency data on how Tom treats people he likes/hates. Not you we really know exact numerical objective frequencies on how often people send bombs to people they hate. Instead, this judgment is based on a more general inference. From the knowledge "Bob hates me" we can rationally derive expectations about how likely it is he would behave relative to someone who likes me. And even though this isn't based on objective frequency data, it is still a rational, non arbitrary and entirely objective assessment to say the package is more likely to come from Bob. Because it is an objective fact that sending someone a bomb would be contrary to your goals if you love that person and want them to be well, and it's objectively true that sending someone a bomb would be in service of your goals if you want to destroy them.

The same goes for God. Built into the theism hypothesis is that he loves us. This gives us objective reasons to think he is more likely to have certain goals (I.e., he's more likely to want to promote well being than he is to want to promote suffering) and this in turn gives us objective reasons to think he's more likely to take certain actions, because certain actions are more likely to produce the outcomes he is more likely to have.

David Brightly said...

Ron, can't we assimilate your story to the objective probabilities model? There are two suspects, Bob Nice and Tom Nasty; nice people send nice packages with high probability and nasty packages with low probability; nasty people are the reverse. We might not know the relevant probabilities, but they must exist in so far as they form the basis for our classifying people as nice or nasty in the first place. But note: suppose I were paranoid and had mistakenly convinced myself of the existence of Tom Nasty and his malice. The bomb would likely confirm me in my false belief in Tom even when I was the randomly chosen victim of some terrorist, say.

Ron said...

David,

The same logic can be applied to God then. God is, by hypothesis, a nice person. Nice people behave in certain ways with objectively high probabilities. Presto! We have an objective basis for deriving probabilistic predictions from theism.

Ron said...

If you believed in Nasty Tom, even if he didn't exist, then the bomb might still be evidence of his existence. On Victor's account of evidence, you can have evidence even for false hypotheses. The question is whether P(receiving bomb|Nasty Tom exists)>P(receiving bomb|nasty Tom isn't real). It probably is more likely on the assumption nasty Tom is real, but this, on its own, isn't enough to show Nasty Tom exists, because the prior probability of nasty Tom might still be low, because there are no independent reasons to think Tom exists. You might say the same about God: certain observations are more likely assuming He exists, but the prior probability of theism is low, so he still probably isn't real