The most baffling mystery of all

[Syntax]

First of all, what you offer is not really a scientific hypothesis but just gambling. If one uses the word “hypothesis” loosely, it can be called one, sort of.

huh? How so?

Now if you wish to establish your hypothesis, one experiment of tossing a coin a few times is not sufficient. What you say about the 50% chance of each coin toss, is of course accurate, but it has nothing to do with verification.

Why does this not have to do with confirmation (verification)? What do you think confirmation even means in a scientific context? I explicitly said in post #100 that,

“Confirmation is raising the probability that the next **unexamined **case will be like the set of all **past examined **cases. And this task confirmation alone cannot do without an implicit assumption about natural laws from which predictions about the likelihood of future events within test scenarious can be deduced.”

So it simply doesn’t matter what the past probability distributions are. Flipping the coin an indefinite number of times never raises the likelihood the next coin will turn up head no matter what the probability distribution function actually is with respect to your past empirical frequencies in your “complicated statistical sampling.” But this is exactly what confirmation must do to make any sense, otherwise it’s just fluff in the wind. See next…

What you need to learn about is statistical sampling, the different distributions stemming from the method of sample taking, chi-square testing, and whole lot of other, not too simple methods.

Here is the proper form of making a hypothesis of the coin toss. You examine the coin, see no obvious fault with it, and therefore hypothesize that it is a “good” coin. You wish to verify that assumption by making “n” coin tosses, and observe the distribution of heads and tails in the sequence. The coin tosses are independent events, in other words, the result of a toss does not influence the result of the next one (the opposite of this assumption is correctly called the Gambler’s Fallacy). We start with the fact that the probability of a head toss is “p” and the probability of a tail toss is “(1-p)”.

The null hypothesis is that “p” = 1/2.

Yes, I am familiar with classical probability here. Thanks…

To establish the veracity of this hypothesis we decide to make “n” number of tosses. If our hypothesis is true, we shall see approximately “n/2” of heads, and “n/2” of tails. How do we -]deduct/-] [deduce] from this result that the coin is really “good”, or not?

No. Scientists don’t “deduce” the hypothesis from the evidence. The hypothesis is INDUCED from the statistical sampling and then subjected to further testing. This is precisely the issue at stake for which I mentioned concerning the problem of Induction in post #100 if you haven’t already read that. Under no circumstance does any number of past case trials raise the likelihood that the next case will conform to the frequencies of past cases without a prior assumption that regularities exist in Nature. So the “veracity” or likelihood of the general hypothesis that “the coin is a fair coin” is never raised over 0.5 with respect to the next unexamined case. You need this additional assumption in order to “raise the probability” from one instance to the next as you increase your number of trials in experiments.

We set up a confidence level. In practice we usually use a 95% confidence, sometimes we might go up to 97.5% confidence. Now we look up a chi-square table, and find what kind of discrepancy we can expect from the “ideal” 1/2 value, and be still 95% sure that the coin was really “good”. This discrepancy is dependent on the number of tosses, obviously.

Sure, this just deals with the converging of initial and posterior probabilities very similar to Bayes Theorem. But neither of them deals with the problem of induction.

The larger the number of experiments is the more precise our verification process turns out to be.

Precision is only a feature of probability densities, not of confirmation which is the raising of the probability that the next trial will be like your past trials.

Have fun. It might take a few years to learn the beuties of probability theory, but it is a good investment.

It’s pretty much a red-herring, but I don’t think you meant that.

 

[Syntax]

This whole debate appears as if we are running in place and I think I have observed why. You are much too selective in which posts you will answer to. Thus, you will always choose a post for which you think you have a good response and for the ones for which you don’t, you simply ignore. Rather than address a question that is difficult to answer and at least confess that there is a possible mistake, you simply evade it.

Believe me, we all know the feeling. It’s obvious which people are just here to jeer at others without actually wanting to engage in any good discussions. I stopped conversing with those who cherry-pick what they want to answer scolding your easier points and ignoring the really poignant questions. It’s no use.

So it’s usually better to stick to a very narrow topic with which you might be well-acquainted than addressing such broad generalities. It’s simply too easy for atheists to slip through the cracks by always taking the default skeptical position, and it doesn’t require a lick of intelligence or critical thinking either, since even a 4 year old child can do it. Just notice how rapid-fire fast a child can ask “why?” as soon as he or she gets on a role, but none of your answers are ever really heeded to spurn further investigation into that one topic because he or she moves on to the next question. Much of the alleged “skepticism” on here, ironically enough, is actually a poor excuse to not actually have to think through your own questions or engage with you, but rather a useful facade to vent the atheist’s insecure ego–and the atheist’s actual intentions subsequently becomes strikingly obvious after having to watch most of your good arguments getting continually ignored. Some are really honest thinkers, however, and you will know them when you encounter them. But they are rare. So try not to let it upset you.

 

[R_Daneel]

huh? How so?

Bucause hypothesis forming and verification is not like a coin toss. Let’s apply your reasoning to different hypothesis: “When we toss a coin, it will land on its edge, not showing either heads or tails.” You keep tossing that coin and it keeps on showing one of its sides, but never lands on its edge. Yet, you maintain that the result of the next toss has a 50% probability of landing on its edge - since the prior tosses have no predicting value. Therefore I offer you an equal bet: we both put down a 100 dollars on the table. If you coin lands on its edge, you win. If the coin lands on one of its sides, I win. Let’s play this game a few thousand times. Are you willing to put your money where your mouth is?

Why does this not have to do with confirmation (verification)? What do you think confirmation even means in a scientific context? I explicitly said in post #100 that,

“Confirmation is raising the probability that the next **unexamined **case will be like the set of all **past examined **cases. And this task confirmation alone cannot do without an implicit assumption about natural laws from which predictions about the likelihood of future events within test scenarious can be deduced.”

DUH! That is the basic assumption, which we use - obviously. We reached this conclusion after observing literally trillions and quadrillions of events which all confirmed it. The proof of the pudding is that it is edible.

Precision is only a feature of probability densities, not of confirmation which is the raising of the probability that the next trial will be like your past trials.

Are you going to play the game I offered above? How much are you going to bet?

 

[warpspeedpetey]

Bucause hypothesis forming and verification is not like a coin toss. Let’s apply your reasoning to different hypothesis: “When we toss a coin, it will land on its edge, not showing either heads or tails.” You keep tossing that coin and it keeps on showing one of its sides, but never lands on its edge. Yet, you maintain that the result of the next toss has a 50% probability of landing on its edge - since the prior tosses have no predicting value. Therefore I offer you an equal bet: we both put down a 100 dollars on the table. If you coin lands on its edge, you win. If the coin lands on one of its sides, I win. Let’s play this game a few thousand times. Are you willing to put your money where your mouth is?

DUH! That is the basic assumption, which we use - obviously. We reached this conclusion after observing literally trillions and quadrillions of events which all confirmed it. The proof of the pudding is that it is edible.

Are you going to play the game I offered above? How much are you going to bet?

as a mathematician, you do understand that there is no such thing as “proof is in the pudding” right?

that no amount of induction, secures any certainty right?

that only one observation, can overturn any number of experiments, even trillions and quadrillions right?

that verification/falsification are logical contradictions? and that no mathematician i have ever known would accept the truth, much less fight for an easily demonstrable, logical contradiction, right?

you keep giving us information any poli sci major gets in a class on public opinion polling. why?

how does that change the bare facts about the problem of induction?

 

[Syntax]

Bucause hypothesis forming and verification is not like a coin toss. Let’s apply your reasoning to different hypothesis: “When we toss a coin, it will land on its edge, not showing either heads or tails.” You keep tossing that coin and it keeps on showing one of its sides, but never lands on its edge. Yet, you maintain that the result of the next toss has a 50% probability of landing on its edge - since the prior tosses have no predicting value. Therefore I offer you an equal bet: we both put down a 100 dollars on the table. If you coin lands on its edge, you win. If the coin lands on one of its sides, I win. Let’s play this game a few thousand times. Are you willing to put your money where your mouth is?

That’s precisely the point I am making. I would NOT give you any personal odds that that hypothesis is true because I, like you, think it is rational to believe *natural regularities *exist. The point is that this natural necessity is never “confirmed” emprically because past observed cases never raise the likelihood that the next coin toss will be like past ones. So any general hypothesis such as “the coin is a fair coin” is an *additional *hypothesis supplementing our past observations in order to explain these past frequencies so as to provide a basis for projecting these regularities to all unobserved cases. So the rational justification for believing natural necessity exists can only lie outside the actual observed probability frequencies. Thus “predictive power” has utility only insofar as we rationally believe there exist regularities in nature, but this additional hypothesis that there are regularities in nature is NOT a product of empirically observed probability frequencies, but is rather used to explain them. Q.E.D

DUH! That is the basic assumption, which we use - obviously. We reached this conclusion after observing literally trillions and quadrillions of events which all confirmed it. The proof of the pudding is that it is edible.

This “assumption” is the very crux of what I am talking about. And no, we’ve never “confirmed” that laws exist literaly “trillions and quadrillians” of times. You’re just begging the very question at stake here. And I hate to break it to you, verbally repeating your same error trillions and quadrillians of times does not thereby make it more likely your error is going to be true in the next post I imagine you will send my way.

Further, I never heard of a “proof” being demonstrated because it was edible.

Are you going to play the game I offered above? How much are you going to bet?

I would no more make that bet than you would.

 

[warpspeedpetey]

And I hate to break it to you, verbally repeating your same error trillions and quadrillians of times does not thereby make it more likely your error is going to be true in the next post I imagine you will send my way.

(heavy sarcasm)

im sorry but youre wrong here.

according to the “proof is in the pudding” theory. he can repeat the same flawed argument trillions and quadrillions of times, and at some point it becomes verified and therefore true. dont you know anything?

you really, truly dont understand probabilities or science do you?

 

[Syntax]

(heavy sarcasm)

im sorry but youre wrong here.

according to the “proof is in the pudding” theory. he can repeat the same flawed argument trillions and quadrillions of times, and at some point it becomes verified and therefore true. dont you know anything?

you really, truly dont understand probabilities or science do you?

Or, “Here, eat this pudding…See? Your munching the pudding is proof the pudding is likely to be true.”

How can pudding be “true”?

 

[Syntax]

that no amount of induction, secures any certainty right?

It’s even worse than that. Without the non-empirical assumption that Nature is Uniform, we haven’t even confirmed any universal inductive generalization once, since we have to actually test all possible cases the hypothesis mentions in order to get beyond the 0.5 probability that that hypothesis is true.

So induction doesn’t even secure likelihood with respect to universal generalizations about all cases such as “this coin is a fair coin.” In the very first post, I dealt with the hypothesis merely concerning a total of 10 trials, 9 trials all of which came out heads raising the initial probability that the hypothesis is true from 002 to .5. But in order for the number of trials to **actually confirm **the hypothesis that “this coin will come out heads 10 consecutvie times” I have to raise the probability that the *last *and final trial will be above .5. After all, if can’t do this, how can I have said to confirm anything? So confimation of this hypothesis is simply impossible to do until we actually flip the coin the tenth and final time making the probability that this hypothesis is true=1.

Now when we are dealing with experiments that are falsely thought to “confirm” a universal generalization such as “this coin is a fair coin,” which is actually a statement about all cases in which this coin is flipped, the situation is even worse since the finite amount of past observed cases *never even comes close *to raising the probability that the next observed case will be 0.5–hence the actual probability that the next case will resemble past cases is **infinitesimally low **since there is a potentially infinite amount of possibly observable cases!

 

[R_Daneel]

I would no more make that bet than you would.

I sure would take my side of the bet offered. Precisely because I infer from the observation of past experiments that the probability of a coin landing on its edge is not 50% as you have been suggesting, but it is literally zero.

Suppose that the bet is not as obvious as it was stated. I offer you a coin and offer a 50% chance type of game. You examine the coin and it looks like a regular, fair coin. Unbeknownst to you, it is not. One side is heavier than the other. You hypothesize that the next toss has an equal likelyhood of being either heads or tails. If you are willing to put your money where your mouth is, you will be a very poor man in no time at all.

If, however, after a few dozens of trials you observe and realize that the coin is not fair, and refuse to play any further, then you admitted that the experimental data refuted your 50% hypothesis.

The regularity of nature is the result of observations, not the result of some ivory-tower speculation.

 

[R_Daneel]

I would no more make that bet than you would.

Here is another fun game and bet I am offering to you.

We sit down and each of us raises either one or two fingers when a signal is heard. The result is that either 2, or 3 or 4 fingers will be seen. If either 2 or 4 fingers are visible, you win 2 or 4 dollars, respectively. If 3 fingers are shown, I win 3 dollars.

Are you willing to play this game? If yes, why? If no, why not? Give me your analysis.

 

[warpspeedpetey]

I sure would take my side of the bet offered. Precisely because I infer from the observation of past experiments that the probability of a coin landing on its edge is not 50% as you have been suggesting, but it is literally zero.

Suppose that the bet is not as obvious as it was stated. I offer you a coin and offer a 50% chance type of game. You examine the coin and it looks like a regular, fair coin. Unbeknownst to you, it is not. One side is heavier than the other. You hypothesize that the next toss has an equal likelyhood of being either heads or tails. If you are willing to put your money where your mouth is, you will be a very poor man in no time at all.

If, however, after a few dozens of trials you observe and realize that the coin is not fair, and refuse to play any further, then you admitted that the experimental data refuted your 50% hypothesis.

The regularity of nature is the result of observations, not the result of some ivory-tower speculation.

which still suffers from the problem of induction.

consider the sun rise, given past observations we could expect the sun to always rise. yet one day it wont. all those quadrillions of observations, overturned, by just one contradictory observation.

you are simply ignoring a centuries old fault in your arguments, which considering that youre a mathematician, makes it seem as though you are not interested in the truth so much as in protecting a cherished belif.

 

[warpspeedpetey]

Here is another fun game and bet I am offering to you.

We sit down and each of us raises either one or two fingers when a signal is heard. The result is that either 2, or 3 or 4 fingers will be seen. If either 2 or 4 fingers are visible, you win 2 or 4 dollars, respectively. If 3 fingers are shown, I win 3 dollars.

Are you willing to play this game? If yes, why? If no, why not? Give me your analysis.

there is 1 way to make a 2, we both hold up 1 finger

there are 2 ways to make a 3, either can hold up 1 or 2 fingers

there is 1 way to make a 4, we both hold up 2 fingers

then 50% of the throws will be a 3 earning you $3

25% will be a 2 earning me $2

25% will be a 4 earning me $4

so my 50% of winning throws will average me $3

given a random distribution. it looks break even. am i missing something?

 

[Syntax]

I sure would take my side of the bet offered. Precisely because I infer from the observation of past experiments that the probability of a coin landing on its edge is not 50% as you have been suggesting, but it is literally zero.

Again, this just assumes empirical frequencies are any guide to future projections.

Suppose that the bet is not as obvious as it was stated. I offer you a coin and offer a 50% chance type of game. You examine the coin and it looks like a regular, fair coin. Unbeknownst to you, it is not. One side is heavier than the other. You hypothesize that the next toss has an equal likelyhood of being either heads or tails. If you are willing to put your money where your mouth is, you will be a very poor man in no time at all.

You’re missing the entire point because you are assuming induction is going to work for all cases, a very finite amount for which you’ve actually seen this reasoning be successful. Here’s a rather simple illustration to drive the point home:

There is box–you cannot see inside it, but you can put your hand in it. The box contains1000 balls. You randomly take out 100 black balls.

If past probabilities are any guide to the future, what can you say about the contents inside the box?

And by the way: you can repeat this experiment until you are blue in the face, each time randmoly pulling out 100 black balls.

 

[R_Daneel]

there is 1 way to make a 2, we both hold up 1 finger

there are 2 ways to make a 3, either can hold up 1 or 2 fingers

there is 1 way to make a 4, we both hold up 2 fingers

then 50% of the throws will be a 3 earning you $3

25% will be a 2 earning me $2

25% will be a 4 earning me $4

so my 50% of winning throws will average me $3

given a random distribution. it looks break even. am i missing something?

Yes, you do. You are missing the background you need to solve such a simple problem.

 

[Syntax]

I sure would take my side of the bet offered. Precisely because I **infer from the observation **of past experiments that the probability of a coin landing on its edge is not 50% as you have been suggesting, but it is literally zero.

You infer from observation?? Let me get this straight. You only observe a finite amount of cases, and then conjecture that the **fairness of the coin **is supposed to hold for all cases based on these very few frequencies of the past? On what ground? There are a potentially infinite amount of possible observable cases making the probability that the next set of coin tosses will approach whatever range of fairness, say 99.7%, intinitessimally close to zero!

Suppose that the bet is not as obvious as it was stated. I offer you a coin and offer a 50% chance type of game. You examine the coin and it looks like a regular, fair coin. Unbeknownst to you, it is not. One side is heavier than the other. You hypothesize that the next toss has an equal likelyhood of being either heads or tails. If you are willing to put your money where your mouth is, you will be a very poor man in no time at all.

If, however, after a few dozens of trials you observe and realize that the coin is not fair, and refuse to play any further, then you admitted that the experimental data refuted your 50% hypothesis.

FYI, “refuted” is a term reserved for logical proofs not empirical testing. You mean “falsify.”

And I hate to say it, but your experiment has only revealed results for only a very limited subset of all possible cases

Suppose your hypothesis is “the coin is fair”; and my hypothesis is that “the coin is biased.” Then we perform a set of 1 million experiments to test our hypotheses. So suppose the data collected from our statistical sampling in these **1 **million experiments reveals to us the probability that the coin is fair comes within the range of 99.7%. So I lose a lot of money.

But how do we know the next set of 3 million experiments will not give you results that the coin is biased in favor on heads on the order of 96.3%? We don’t know this empirically whatsoever. So suppose the next 3 million trials does, in fact, reveal exactly these results. So even if I come out poor the the 1st 1 million trials, I will be 3 times more rich than you after a total of 4 million experiments.

The regularity of nature is the result of observations, not the result of some ivory-tower speculation.

But you cannot establish this empirically at all, simply because you have not even come close to observing more than 50% of possible test cases…since **1millioncases/a potential infinity of cases **approaches infinitesimally close to 0.

Your problem is this:

You’re missing the entire point because you are assuming induction is going to work for all cases based on a very finite amount of cases in which you’ve actually seen this reasoning be successful. Hence you are using induction to justify inductive reasoning. So your arguments are circular at their core.

Now that I’ve answered your questions, in fairness you have to answer mine. Here’s a rather simple illustration to drive the point home:

There is box–you cannot see inside it, but you can put your hand in it. The box contains1000 balls. You randomly take out 100 black balls.

If past probabilities are any guide to the future, what can you say about the rest of the contents inside the box? What’s your answer?

Oh!..And you can repeat this experiment until you are blue in the face, each time randmoly pulling out 100 black balls.

 

[warpspeedpetey]

Yes, you do. You are missing the background you need to solve such a simple problem.

yup, i made a mistake. you will win 2/3rds of the time.

funny though, didnt you claim that you were going to ignore me, that i wasnt worth talking too?

but you just couldnt pass up, commenting on my mistake, could you?

now i know that you read every post i write.

 

[Sarpedon]

My amazement stems from their utmost reluctance to admit: “they believe in God’s benevolence on blind faith, they need no evidence for it”.

The idea of God’s benevolence is rooted in the concept of God as the ultimate being and the fact that we exist. As a self-existent and perfect (not in the moral sense) being, God has no need of a creation. As a perfect and unchanging being, God cannot increase or decrease in perfection. Therefore, He cannot benefit from our existence. God can’t be a scientist studying us, because He cannot increase His already perfect knowledge. He can’t create us just for fun, because he can’t increase His personal satisfaction from us.

Nevertheless, we do exist. Assuming the existence of God has already been established, Catholics posit that God is benevolent because God cannot benefit from His creation. Therefore, a sort of completely selfless love or benevolence is the only reason God would have for creating us with nothing to gain Himself.

 

[R_Daneel]

You infer from observation?? Let me get this straight. You only observe a finite amount of cases, and then conjecture that the **fairness of the coin **is supposed to hold for all cases based on these very few frequencies of the past? On what ground? There are a potentially infinite amount of possible observable cases making the probability that the next set of coin tosses will approach whatever range of fairness, say 99.7%, intinitessimally close to zero!

Suppose your hypothesis is “the coin is fair”; and my hypothesis is that “the coin is biased.” Then we perform a set of 1 million experiments to test our hypotheses. So suppose the data collected from our statistical sampling in these **1 **million experiments reveals to us the probability that the coin is fair comes within the range of 99.7%. So I lose a lot of money.

But how do we know the next set of 3 million experiments will not give you results that the coin is biased in favor on heads on the order of 96.3%? We don’t know this empirically whatsoever. So suppose the next 3 million trials does, in fact, reveal exactly these results. So even if I come out poor the the 1st 1 million trials, I will be 3 times more rich than you after a total of 4 million experiments.

You argue that intuitively and based upon common sense that the first 1 million experiments cannot predict the next 3 million ones. Well, yes they can. There is this subject called mathematical statistics (part of the theory of probabilities) which deals exactly with this problem. It is very counterintuitive, but nevertheless mathematically proven that the absolute size of the sample, and not its relative size compared to the full population is what matters.

Consider poll taking. In the US there are about 300 million people. Poll takers select a sample of a few thousand (the selection method is another nifty problem) and the result will be highly accurate pertaining to the whole population. Is it 100% accurate? Of couse not. Is it accurate enough to make predictions on it? Most definitely.

Consider the random movement of air molecules. Theoretically it is possible that sitting in a room one can die of asphyxiation, because all the air molecules just happen to go to the walls leaving the rest of the room in a prefect vacuum. What are the chances of that? In the whole lifespan of the universe it will not happen.

Your problem is this:

You’re missing the entire point because you are assuming induction is going to work for all cases based on a very finite amount of cases in which you’ve actually seen this reasoning be successful. Hence you are using induction to justify inductive reasoning. So your arguments are circular at their core.

Actually it is not circular, it is an ever ascending spiral. I am sure you know the difference.

Now that I’ve answered your questions, in fairness you have to answer mine. Here’s a rather simple illustration to drive the point home:

There is box–you cannot see inside it, but you can put your hand in it. The box contains1000 balls. You randomly take out 100 black balls.

If past probabilities are any guide to the future, what can you say about the rest of the contents inside the box? What’s your answer?

Oh!..And you can repeat this experiment until you are blue in the face, each time randmoly pulling out 100 black balls.

Certainly I will answer you. It is a legitimate problem. Let’s make sure that we think about the same question. You have a box with 1000 balls it, some black, the other ones not black (say: white). I have to select 100 balls, randomly, and make inferences about the whole box based upon the result. You stipulate that all 100 selected balls are black.

There are two ways to conduct the experiment.

Method A: One is that I take out 1 ball at a time, examine it, and put it to the side. Then I reach in again and take out another one, etc… The balls selected do not go back into the box. This is the same as reaching into the box and selecting 100 balls at the same time and pulling them out. These two methods are identical, the distribution is called hypergeometric distribution.

Method B: The other way to perform the experiment is to reach into the box, pull out a ball, examine it, jot down the result (it was black) and put it back. Shake the box, and repeat the experiment. This time we deal with binomial distribution.

The two physical methods are not identical, there is a very small difference. However, with having a 1000 balls in the box, the difference is negligible, the binomial distribution very closely approximates the hypergeometric distribution. I suggest to use the second method, the binomial distibution, because the other one is very cumbersome mathematically. Do you agree?

 

[R_Daneel]

yup, i made a mistake. you will win 2/3rds of the time.

Still incorrect. Keep on trying.

funny though, didnt you claim that you were going to ignore me, that i wasnt worth talking too?

but you just couldnt pass up, commenting on my mistake, could you?

Actually that is not the reason. I would have commented upon your answer if it were correct - and would have complimented you. The point is that if you make legitimate comments, I see no reason not to answer. But when you say something utterly nonsensical (like the substantiation of the moon landing is on par with the Bible), which has been explained to you over and over again, I will not waste any more time to respond.

now i know that you read every post i write.

Of course I do. In the treads I post, I read every comment, even if I select which ones should I reflect upon. I never use the “ignore” button. There is always a chance that the poster-to-be-ignored does post something worthy to read.

 

[R_Daneel]

The idea of God’s benevolence is rooted in the concept of God as the ultimate being and the fact that we exist. As a self-existent and perfect (not in the moral sense) being, God has no need of a creation. As a perfect and unchanging being, God cannot increase or decrease in perfection. Therefore, He cannot benefit from our existence. God can’t be a scientist studying us, because He cannot increase His already perfect knowledge. He can’t create us just for fun, because he can’t increase His personal satisfaction from us.

Nevertheless, we do exist. Assuming the existence of God has already been established, Catholics posit that God is benevolent because God cannot benefit from His creation. Therefore, a sort of completely selfless love or benevolence is the only reason God would have for creating us with nothing to gain Himself.

What you say is quite logical. There are a few problems, however. First, it is based upon the assumption that existence (in and by itself) is somehow “better” than nonexistence. I don’t see how can you substantiate that. (I just hope that none of the nincompoops will ask if I would prefer not to exist. That nonsense has been uttered too many times before). Second, this alleged benevolence cannot be “exhausted” in the act af creation. God manifestly does not interfere in a benevolent fasion in our behalf.