# Leibnizian Cosmological Argument

#### loko5

• 207 Posts
##### Cosmological argument seems unconvincing
« Reply #15 on: May 20, 2009, 09:51:14 am »

daveknott wrote: I am not convinced that an initial value is required to seed the regression in order to provide substance on which the rule can act. That requirement would only be necessary if the regression began at some point, but since it is infinite, there is no "blank spot" at the beginning requiring input by which subsequent states can be characterized. All of the "slots" are already filled in, courtesy of their predecessors, of which there is no limit. To nonetheless demand that an initial point be inserted to get the process going is to utterly miss the point that the regression has no beginning. A recursively defined series is a coherent entity with observable properties regardless of whether an initial point is provided or not.

Let me give you a mathematical example.  Let's construct an infinite set of numbers such that each member of the set is derived from each previous member of the set under the following rule:

If the previous member is an odd number, add 3 to it; if it is an even number add 5 to it.

Now let me ask you, is the number 0 a member of this set?

I don't know how good you are at math, so I will tell you the answer.  The answer is it is impossible to say.  It should be obvious that not every integer will be a member of the set (since we obvioiusly skip some numbers by adding 3 or 5 instead of 1).  But whether a particular integer will belong to the set or not depends on where you start applying the rules.  As it stands, we have not really defined a set at all.

daveknott wrote: And I can't help but wonder how you find infinite regression incoherent on the grounds that it has no initial point from which to build, but readily accept the existence of an eternal deity, who would seem subject to the same objection. Unless you espouse something about his "transcendence" of time--but if that is your position, I could of course claim the same thing about my Superverse.

Because God is the initial point from which to build.  The same could be said about your Superverse.

1

#### James D Sinclair

• 120 Posts
##### Cosmological argument seems unconvincing
« Reply #16 on: May 20, 2009, 11:56:22 am »
Dave,

Actually I was defending Craig's position and opposing the caricature of Christianity in the quote, although I guess I was too subtle. With regard to the quote, I am pointing out that such a characterization is actually impossible to make for a person who denies the principle of sufficient reason (which is pretty much what drives the Liebnizian argument for the existence of God).

The Liebnizian argument is centered around the 'Principle of Sufficient Reason'. Pruss explains "The PSR states that every fact, or every contingent fact, has an explanation."
A related, but not identical, claim in the Kalam argument is that things which 'begin-to-exist' have a cause.

An argument that is often made is that if there is no PSR (or, similarly, something comes from nothing given no rules) then why don't we see anything and everything popping into being around us? What Koons adds to this is that no probability attaches to the frequency or type of these events. So, if they happen, it is not possible to say we shouldn't see them. Nor can one say it is more likely that an observed fact was produced via an explicable scientific foundation as opposed to fiat. In fact, even the thoughts in your head could come from this process.

Now, some Young Earth Christians (I am not one) used to say in the 19th century that the dinosaur bones were put there by God to give an 'appearance of age'. Thus they were (are) belittled. But the denial of the PSR produces 'appearance of age' on steroids. Your whole current perception of the world around you could be one of these events. Philosophers have a name for this: 'envatment'. It is based on brain-in-a-vat scenarios, such as the movie Matrix. They use it as a tool to discern a serious philosophical outlook from an unserious one.

Now, the interesting thing about Koons' objection is that one cannot refute it by saying 'the reason you are wrong is because the peculiar nature of this process is (fill in the blank)'. For the moment that you do you are attaching a fact to the process and/or asserting a property to 'nothingness', which would be incoherent.

2

#### David Knott

• 24 Posts
##### Cosmological argument seems unconvincing
« Reply #17 on: May 20, 2009, 12:06:55 pm »
Loko5, I am not trying to be deliberately contradictory. To put it in words you yourself were probably thinking, no, I'm not trying to tick you off, or stroking my ego.

My hypothetical uncaused event A, being unconnected with previous circumstances and facts, far from being forced into some form which cannot be imagined otherwise, is actually free to be anything it wants, as nothing in the entire universe (by definition) can be said to determine its character in any way. In this sense, it is random in the fullest, most ideal sense of the word. Not only are its characteristics not dependent upon anything, but its very existence is not dependent upon anything. If we were to vary the circumstances, as you suggest, and hit the "reset" button on our universe, and observe the nature of A a "second time around", to see whether it varied or not with the circumstances, we'd be disappointed: even A's existence is not guaranteed, as even that isn't dependent upon circumstances.

As a preliminary candidate for such an entity, I might put forth quantum fluctuations. Taken as a whole, one could say they are collectively "caused" by the laws of physics, but as far as the cause of any individual fluctuation, there doesn't seem to be one (if I've not misinterpreted what I've read on the subject). Each fluctuation is evidently uncaused, but by no means is "inconceivably otherwise". Truly random events in general would fit my description of an entity for which there is no cause, but could be imagined otherwise.

I think (not to beat a dead horse) that your difficulty in seeing where I'm coming from might stem from your thinking about variance of circumstances with respect to some particular event. You imagine an uncaused event A to exist, and then imagine yourself "varying the circumstances"--maybe by going back in time a few seconds and moving a few atoms around, then pressing the play button again. You imagine A to faithfully pop into existence again, despite your having shifted around some atoms. Any such shifting, any number of times, you reason, wouldn't prevent A's existence, and hence, A is "necessarily existing", because we can't prevent it from coming into existence. This is an incorrect picture. In my view, if you went back in time a few seconds, shifted around some atoms, and then pressed the play button, event A would not happen--NOT because you shifted atoms around, but because it is a random event, which simply may or may not happen.

Phew! Sorry I spent so space writing on that one.

I don't know how good you are at math, so I will tell you the answer. The answer is it is impossible to say. It should be obvious that not every integer will be a member of the set (since we obvioiusly skip some numbers by adding 3 or 5 instead of 1). But whether a particular integer will belong to the set or not depends on where you start applying the rules. As it stands, we have not really defined a set at all.

I am familiar with recursively defined formulas--they crop up, for instance, when you solve differential equations by assuming that a solution exists in the form of a power series. I can use this example to show what I'm saying. A general solution to the equation is obtained, and it will contain some number of undetermined, arbitrary coefficients. If no initial values are known, then, as you have argued, there is no way to lock those coefficients down to a specific value--they have to be left as arbitrary, defined only in terms of previous coefficients, which are in turn defined by more previous coefficients, etc. Clearly, although such an entity represents a family of functions, and not a specific function, to say that it is "incoherent" is plainly false. A function whose coefficients are given by a recursively defined formula is certainly coherent, and even has well defined, observable properties and behaviors.

So, for instance, let N(X) be the Xth term of (N1, N2, N3, ...). Let

N(X) = N(X - 1) + 2

Without an initial input, like N(5) = 3, not a single term of this sequence is determined. Does that make it incoherent? Does that make it incomprehensible? I think not, nor do I think infinite regression is necessarily incoherent. But that is what you have tried to argue: that such entities are incoherent.

Now, I know I've written a lot in this post and the last post, and it's unreasonable for me to expect you to take the time out of your day to quibble with everything I say. I'm sure you have other things going on in your life! But as a matter of accounting, I wish to remind you that I made some points that remain unanswered, namely:

1) Isn't "zero" as non-arbitrary a value as infinity for the first cause's properties to take on?

2) Is the first cause, in addition to being infinitely wise and strong, infinitely loud, infinitely massive, and infinitely pissed off?

3

#### David Knott

• 24 Posts
##### Cosmological argument seems unconvincing
« Reply #18 on: May 20, 2009, 12:30:34 pm »
Sinclairjd, thanks for the clarification.

An argument that is often made is that if there is no PSR (or, similarly, something comes from nothing given no rules) then why don't we see anything and everything popping into being around us? What Koons adds to this is that no probability attaches to the frequency or type of these events. So, if they happen, it is not possible to say we shouldn't see them. Nor can one say it is more likely that an observed fact was produced via an explicable scientific foundation as opposed to fiat. In fact, even the thoughts in your head could come from this process.

This is an extremely good point, and it quite astutely notices that if an event is uncaused, then similarly no rules restrict it. This is partially why I find the argument that uncaused entities must be "inconceivably otherwise" so fatuous. The very opposite seems to be true, in that entities that are wholly independent of circumstance would seem to be more free than anything else.

Now, here is my response. In one sense, we actually do have things popping into being around us all the time: the virtual particles driven by quantum fluctuations. But these are not satisfactory, as they are small and well behaved, and obey a probabilistic distribution--so they're not the truly "free" monsters that we might expect. I would therefore say that, it might be the case that pockets (universes) can arise in which the frequency of such events is controlled, much as a universes might be, in general, a sort of filling out of previously undetermined values. While the frequency of such uncaused monsters might be indeterminate in the Superverse, or whatever you wish to call it, it is conceivable that universes could arise in which a frequency is randomly assigned. We might just find ourselves in a fortunate world whose monster frequency is sufficiently close to zero.

This is, obviously, something I'm making up as I go along. But that is the standard practice, evidently, in discussing first cause arguments. You only have to remain within the realm of the logically conceivable to have a decent argument on your side!

Now, let me turn the question around on you in a different form. If no such uncaused monsters exist, that is, if there is no true, utter, ideal randomness to be found in reality, which is profoundly disconnected from circumstance, then doesn't that leave us with a fully deterministic world, one where the cogwheels just churn out the initial instructions given by the first cause? Wouldn't that imply, therefore, that it is not only the first cause which must exhibit this property of being "inconceivably otherwise", but all of the cosmos itself in addition, because it stems so mechanically from the creator, with no allowance of real randomness, and hence real variability? Why then, as Loko5 may have put it, do we find so many things in our world "denumerable and measurable"?

4

#### loko5

• 207 Posts
##### Cosmological argument seems unconvincing
« Reply #19 on: May 22, 2009, 05:55:59 pm »

My hypothetical uncaused event A, being unconnected with previous circumstances and facts, far from being forced into some form which cannot be imagined otherwise, is actually free to be anything it wants, as nothing in the entire universe (by definition) can be said to determine its character in any way ... Not only are its characteristics not dependent upon anything, but its very existence is not dependent upon anything.

Sounds an awful lot like God to me.  This part of your response makes it sound like you are agreeing with the Leibnizian cosmological argument.

If we were to vary the circumstances, as you suggest, and hit the "reset" button on our universe, and observe the nature of A a "second time around", to see whether it varied or not with the circumstances, we'd be disappointed: even A's existence is not guaranteed, as even that isn't dependent upon circumstances.

Where would we "reset" from?  If A's existence depends on circumstances, what exactly are the circumstances?  I would say if A's existence depends on something, then, by definition, A is not uncaused.

As a preliminary candidate for such an entity, I might put forth quantum fluctuations. Taken as a whole, one could say they are collectively "caused" by the laws of physics, but as far as the cause of any individual fluctuation, there doesn't seem to be one (if I've not misinterpreted what I've read on the subject). Each fluctuation is evidently uncaused, but by no means is "inconceivably otherwise". Truly random events in general would fit my description of an entity for which there is no cause, but could be imagined otherwise.

You are conflating cause with determinism.  I have tried to avoid any discussion in this thread of what constitutes a cause, because it is a complicated area.  There are many different types of causes that philosophers talk about, and nearly everything we can identify as having a cause actually has multiple causes.  Unraveling them can be like trying to untie the Gordian knot.  But for the purposes of our discussion, I will define a cause as being any type of explanation or reason we can point to for the existence of a thing or event.  Philosophers often talk about "probabilistic causes".  When we say that smoking causes cancer, or drunk driving causes accidents, we obviously don't mean that everyone who smokes a cigarette will necessarily get cancer or that every time a drunk person drives a car it will necessarily result in an auto accident.  This is the same category of cause as quantum fluctuations, in the sense that they are probabilistic.  For a quantum fluctuation, we can point to the presence of the quantum field and to the laws of quantum mechanics as causes of the fluctuation.  But what causes the quantum field to exist, and what causes the laws of physics to be what they are?  Are you proposing some kind of cause, or are you taking these to exist necessarily (without cause)?

I think not, nor do I think infinite regression is necessarily incoherent. But that is what you have tried to argue: that such entities are incoherent.

When I say that an infinite regress is incoherent, I don't mean that infinite mathematical series are incoherent.  Certainly, as you point out, we can write a recursive formula for an infinite series and it has some meaning to us even if the specific values aren't filled in.  But as an explanation for the universe, an infinite regression is incoherent because it does not really provide an explanation at all.  A full explanation would explain not only what events or things occur in the universe (the members of the series), but also where the causes come from (the rules for generating the members of the series).  How can your infinite regress do this?

1) Isn't "zero" as non-arbitrary a value as infinity for the first cause's properties to take on?

"Zero" as a value for a property just means that the object lacks that property.  Thus, to say that an object has zero speed means that it is at rest.  I'm not sure what it means for an object to have a property whose value is infinity.

ss=MsoNormal style="MARGIN: 0in 0in 0pt">2) Is the first cause, in addition to being infinitely wise and strong, infinitely loud, infinitely massive, and infinitely pissed off?

You will often hear theists describe God as "infinitely good" or "infinitely knowledgeable", etc.  These terms can't be regarded mathematically, however, because we have no metric for measuring goodness or knowledge.  To say that God is "infinitely good" just means that goodness is part of God's nature, and that anything that is good comes from God.  To say that god is "infinitely knowledgeable" means that God has knowledge of every fact that exists (whether the number of facts in existence is infinite or finite is an interesting question but off topic, I think).

We can deduce some properties of God through philosophical reasoning.  I think, for example, that God would have to be regarded as eternal, in that he has always existed (there was never a moment in time in which he did not exist).  Since God is the first cause, we can also think of him as being omnipotent in some sense, because he could have created the universe differently than what it is, and there is nothing external to God that would have limited what kind of universe he made (if there were, then this would be causing God to do what he does, thus he would not be the first cause).  To fully describe the Christian concept of God, though, requires divine revelation.

I would say that it is absurd to ascribe physical properties to God (except perhaps metaphorically), since he is not a physical being.

5

#### David Knott

• 24 Posts
##### Cosmological argument seems unconvincing
« Reply #20 on: May 23, 2009, 11:03:24 am »
Sounds an awful lot like God to me. This part of your response makes it sound like you are agreeing with the Leibnizian cosmological argument.

By "free to be anything it wants", I didn't mean "free to change into anything it literally chooses, as an agent with intentionality". There's no intentionality involved--the uncaused entity is simply not restricted in what kind of thing it turns out to be. It could be a deity, or it could be proton. The point I'm making is that, being unconnected with previous circumstances, and being in a sense wholly independent of its surroundings, there is nothing in the world that can "pin down" the uncaused event ahead of time--meaning that its character, form, nature, etc., are all undetermined, and therefore free variables. This is in contradiction to the idea that an uncaused entity has to be "that which cannot be imagined otherwise".

Where would we "reset" from? If A's existence depends on circumstances, what exactly are the circumstances? I would say if A's existence depends on something, then, by definition, A is not uncaused.

We'd reset from some point in time, after having watched A's occurrence. Everything after the first sentence seems to totally misunderstand what I've said; I said the very opposite of what you've understood. I claimed that A was wholly independent of circumstance, prior events, etc. It is, effectively, a random event, which will or will not occur, without any regard to the present situation of the universe. In that sense it is uncaused. But there is nothing in this definition that implies the property of "being incapable of being otherwise", and indeed the opposite is implied, as I've said above, and in my previous post.

My purpose in talking about such an entity is to show that one can conceive of an entity which is uncaused, but which nonetheless lacks the property of being "unimaginably otherwise", which undoes your argument before it can get off the ground.

I agree with you that the notion of causation is a little slippery, and can mean different things in different contexts. That is partially why I'm skeptical of the whole cosmological argument: it more or less presumes to have defined causality with a mathematical certainty, and proceeds as though it were working from axiomatic truths. But there are ambiguities lurking beneath the surface, and they have the potential to unravel the argument. But anyway. You go on to say, after talking a little about causality:

When we say that smoking causes cancer, or drunk driving causes accidents, we obviously don't mean that everyone who smokes a cigarette will necessarily get cancer or that every time a drunk person drives a car it will necessarily result in an auto accident. This is the same category of cause as quantum fluctuations, in the sense that they are probabilistic. For a quantum fluctuation, we can point to the presence of the quantum field and to the laws of quantum mechanics as causes of the fluctuation. But what causes the quantum field to exist, and what causes the laws of physics to be what they are? Are you proposing some kind of cause, or are you taking these to exist necessarily (without cause)?

You are wrong in equating fluctuations with statistically measurable effects like cancer from cigarette smoke. If you had the technology and patience, in principle, you could observe the chain of billiard ball type events that occur between chains of particles, observing how cigarette smoke literally instigates cancer. The cancer, then, really does depend upon a definite, specific prior circumstance (namely a large number of carbon monoxide molecules floating into the lungs). When we say that smoking causes cancer, we are, as you hint, speaking of an aggregate effect, and not claiming that one cigarette causes cancer in every case. We are making a probabilistic statement that can be inferred statistically. But this certainly does not mean that the causal chain itself is in some way fundamentally probabilistic, or random, observable only as an aggregate effect. We make such statements merely because the effect is too small to be coherent or recognizable on the level of atoms and molecules. But nonetheless, there is a billiard ball type phenomenon underlying the cancerous effects of cigarette smoking.

In other words, while the aggregate effect of trillions of carbon monoxide molecules on thousands of human lungs may lend itself to a probabilistic description, the underlying phenomenon is decidedly not probabilistic at its core. This is not so with quantum mechanics. The probabilities that characterize quantum mechanics do NOT spring up from humans considering their aggregate effects; the probabilities exist all the way down, no matter how closely you look. They are fundamental. That is why, in the early 20th century, it was such an utterly difficult pill to swallow: because probability and randomness existed not as convenient human approximations to the true actions of particles, but as the true actions themselves.

To say that the laws of quantum physics are what cause the fluctuations is a tautology. The laws are, themselves, merely a description of what we observe in experiments. If you allow yourself this bit of reasoning, then no event could ever be construed as uncaused, even in principle. You could, for instance, observe God's (apparently uncaused) existence as a product of the "God Law", which merely states that "God must exist". I would then be able to say that God isn't really uncaused--he is caused by the God Law. Clearly such a thing would be ridiculous, for the law would merely be a human abstraction, superficially slipped underneath the most fundamental, real event: God. And so too is your proposition ridiculous, for you are taking your human abstraction and sticking it underneath the quantum fluctuations, imagining that you have found a cause for them, when you have, in fact, merely described them. The fluctuations stand as, if not the genuine article, then a highly convincing analog for the uncaused, but variable entities of which I am always speaking.

When I say that an infinite regress is incoherent, I don't mean that infinite mathematical series are incoherent. Certainly, as you point out, we can write a recursive formula for an infinite series and it has some meaning to us even if the specific values aren't filled in.

You have effectively stepped back from your original position. You said, several posts back, that:

[from several posts ago] An infinite regress of causes would be analogous to an infinite mathematical set in which each member of the set (analogous to each state of existence in the world) is generated by performing an operation (analogous to the cause in an infinite regress of causes) on the previous member, but each operation to be performed is different. In this case, it is impossible to say what the set is, because we have no "starting point" from which to start applying the rules.

You have now conceded the coherence of an undetermined infinite series, but changed your mind to say that, perhaps the analogy between the universe and the series is not so strong after all. But more to the point, after downplaying the strength of the analogy, you repeat that which you stated previously, but failed to demonstrate:

But as an explanation for the universe, an infinite regression is incoherent because it does not really provide an explanation at all. A full explanation would explain not only what events or things occur in the universe (the members of the series), but also where the causes come from (the rules for generating the members of the series). How can your infinite regress do this?

This is exactly the same argument that you made much earlier, when you said that an infinite regress can explain only "objects", but not "mechanisms" (the causal mediums themselves, like gravity). And I showed you exactly how an infinite regress could accomplish this. Consider a series of events ("objects") which cause each other (A1, A2, ...). Let B1 be the cause that mediates between each event (or "mechanism" or "rule for generating numbers of the series"). It is true that an explanation for B1 cannot be found amongst A1, A2, etc. But we can postulate a series of higher level causes ("objects") on the B level: (B1, B2, ...). We will need yet another "mechanism" or "rule" for explaining our B-level events. Call it C1. We can explain C1 in terms of other C-level events (C1, C2, ...).

In other words, we can imagine constantly superseding mechanisms which are responsible for the daughter events. We induce a new infinite regression, not in terms of ever preceding events, but in terms of ever superseding rules, or mechanisms. Your objection, therefore, that the rules/mechanisms/causes cannot be explained in an infinite regression, is simply not true. But I said all of this before. I won't say it again.

We can deduce some properties of God through philosophical reasoning. I think, for example, that God would have to be regarded as eternal, in that he has always existed (there was never a moment in time in which he did not exist). Since God is the first cause, we can also think of him as being omnipotent in some sense, because he could have created the universe differently than what it is, and there is nothing external to God that would have limited what kind of universe he made (if there were, then this would be causing God to do what he does, thus he would not be the first cause).

No. You have it backwards. You are ascribing properties to God, GIVEN that he is the first cause. That God is the first cause remains to be demonstrated! You must first demonstrate what properties the first cause must have, according to your cosmological argument, and then show that those properties coincide with our description of God.

I thought that, in insisting upon this idea that uncaused entities are "unimaginably otherwise", you would conclude that the first cause could not have any arbitrary properties (such as weighing eleven pounds, or being 76.2% powerful), and would therefore imagine each of those properties (power, intelligence) to be infinite, in order to make it fit the "unimaginably otherwise" description, and remove all arbitrariness. But apparently this is not what you plan to do? If you wish to narrow the scope of our argument and allow our posts to be shorter (I certainly do), you could agree to solely take up this challenge: if it is true that causality is universal, and that an infinite regression is an unacceptable explanation, and that uncaused entities are that which "could not be imagined otherwise" (as you know I dispute all of these), then how do you get arrive at the idea that God is the first cause?

I sincerely apologize for writing so much. I know it must be a pain to read through it. I won't hold it against you if you just don't respond to much of it.. there has to be a limit to the length of our posts.

Looking forward to more good discussion with you,

Dave

6

#### loko5

• 207 Posts
##### Cosmological argument seems unconvincing
« Reply #21 on: June 01, 2009, 03:53:24 pm »
Dave,

Sorry it has taken me so long to respond.  I was on vacation last week and, unfortunately, without an internet connection.  I hope you are willing to continue the conversation.

These posts are getting long, so if it is OK with you I will just respond to one of your points for now:

daveknott wrote:
But as an explanation for the universe, an infinite regression is incoherent because it does not really provide an explanation at all. A full explanation would explain not only what events or things occur in the universe (the members of the series), but also where the causes come from (the rules for generating the members of the series). How can your infinite regress do this?

This is exactly the same argument that you made much earlier, when you said that an infinite regress can explain only "objects", but not "mechanisms" (the causal mediums themselves, like gravity). And I showed you exactly how an infinite regress could accomplish this. Consider a series of events ("objects") which cause each other (A1, A2, ...). Let B1 be the cause that mediates between each event (or "mechanism" or "rule for generating numbers of the series"). It is true that an explanation for B1 cannot be found amongst A1, A2, etc. But we can postulate a series of higher level causes ("objects") on the B level: (B1, B2, ...). We will need yet another "mechanism" or "rule" for explaining our B-level events. Call it C1. We can explain C1 in terms of other C-level events (C1, C2, ...).

In other words, we can imagine constantly superseding mechanisms which are responsible for the daughter events. We induce a new infinite regression, not in terms of ever preceding events, but in terms of ever superseding rules, or mechanisms. Your objection, therefore, that the rules/mechanisms/causes cannot be explained in an infinite regression, is simply not true. But I said all of this before. I won't say it again.

This seems to be the heart of your argument, that we can consider everything to be contingent (in the sense of having a reason for its existence) by explaining it in terms of other contingent things in an infinite regress.  You probably won't be surprised to know that this is an old argument, apparently originating with Hume in the 18th century.  I seem to be at an impasse with you in trying to explain why this argument is faulty.  Let me try one more time.  The problem with this argument is that it can explain individual facts (much as a member of an infinite series can be specified by applying the rules) but it cannot explain the aggregate, or whole, of the body of facts.  It's a chicken-and-egg problem:  we can explain the existence of the egg in that it was laid by a chicken, and the existence of the chicken in that it hatched from an egg.  If we continue this to an infinite regression, we find that we have explained the existence of each chicken and each egg in the series, but we don't have anywhere near a complete explanation of the whole.  Why does the system (chicken + egg) exist at all?  Perhaps a more realistic example would be a cyclic model of the universe, in which the big bang is followed by gravitational collapse into a "big crunch", which then cycles into another big bang, ad infinitum.  Each big bang can be explained in terms of the preceding big crunch, and each big crunch is the result of gravitational collapse of the material from the previous big bang.  Even if such a model were true, it does not fully explain the universe.  What explains the aggregate existence of the system (big bang + big crunch)?

I'd encourage you to read what some philosophers have to say on the matter.  Richard Swinburne discusses this argument in his book, The Existence of God, which you can find linked here.  Richard Gale and Alexander Pruss have also discussed this argument at length.

7

#### David Knott

• 24 Posts
##### Cosmological argument seems unconvincing
« Reply #22 on: June 04, 2009, 10:38:12 am »
Hi again,

No worries about the long delay. I'm delighted to continue the conversation.

Actually, the infinite regress point I would consider one of the least substantial parts of my argument, but I'll take you up on it. You give the example of the chicken and egg problem, but then ask, "Why does the system (chicken + egg) exist at all?". The "system" you refer to is not some separate, distinct entity from the collection of chickens and eggs. It isn't tangible or observable in any sense. It is just an abstraction you have coined to talk about all the chickens and all the eggs. The system, in this case, really is just the sum of the parts; it doesn't require some additional explanation on top of the explanation given for the individual elements.

Look at it this way. Suppose you challenge me to explain the presence of three helicopters in your back yard. I give you a full causal history of the first helicopter, then give you a full causal history of the second, and then the third. So in other words, taking each helicopter individually, I've shown you how each one was manufactured by the army, how the engineers designed it, where they went to college, where they were born, where we got the electricity to power the manufacturing process, and so on. But you then tell me that I haven't explained the system of helicopters. And actually, there are four systems I need to explain: the system of helicopters 1 and 3, the system of helicopters 1 and 2, the system of helicopters 2 and 3, and the system of all helicopters 1 2 and 3. Would you not consider this a silly request? The "system" of helicopters is an abstraction you have come up with, and doesn't require explanation in addition to what I have already provided.

8

#### loko5

• 207 Posts
##### Cosmological argument seems unconvincing
« Reply #23 on: June 09, 2009, 12:16:11 pm »

daveknott wrote: Hi again,

No worries about the long delay. I'm delighted to continue the conversation.

Glad you're patient.  I try to keep up with this site when I can but life can keep me pretty busy.

I get the feeling we are talking in circles, so let me step back for a minute and briefly review the Leibnizian cosmological argument, so we can pinpoint exactly where we disagree.  This way we can handle points of disagreement one by one as they come up rather than keep shifting back and forth to different points of disagreement.  Here is a version of the Leibnizian argument as it is often presented:

(1)  Everything that exists is either necessary or contingent.
(2)  Necessary things require no explanation for their existence, because they could not fail to exist.
(3)  Contingent things require an explanation.
(4)  The universe is contingent.
(5)  Therefore, the universe must be explainable in terms of something that is necessary.
(6)  This explanation is God.

Personally, I don't think (6) can stand as a conclusion without additional support, so I prefer to modify the traditional Leibnizian argument to read:

(6')  Therefore naturalism is false, i.e. there must be some necessarily existing First Cause external to the universe that explains why the universe exists.

and leave it to other arguments to show why this First Cause must be God.

Now, rather than respond to your last post, let me go through the argument step by step to find our points of disagreement.

You seem to have some dispute on what is meant by "necessary" and "contingent", so let me define more carefully what I mean by these terms:

To say that something is necessary is to say that it could not have been different and that it could not have failed to exist.  If it could not have failed to exist, then it must exist eternally.  That is, there can never have been a time nor could there ever be a time when it does not exist, because this would mean there is a time when it fails to exist, thus contradicting the definition.  Under this definition, we are forced to accept necessary things as brute facts, i.e. they require no reason for being what they are.

Anything that is not necessary is by definition contingent.  Thus we have what is known in philosophy as a disjunct.  Everything in existence can be defined as either contingent or necessary, nothing that is necessary can be contingent, and nothing that is contingent can be necessary.  This is true even if (as some people have tried to argue) everything that exists is necessary, or everything that exists is contingent.

OK, this gets us through steps (1) and (2) of the argument.  From what we have discussed in this thread up to this point, I don't think you should have any objections to these points, but before moving on to step (3), I would like to know hear back from you.  Do you have any objections to premises (1) and (2)?

9

#### David Knott

• 24 Posts
##### Cosmological argument seems unconvincing
« Reply #24 on: June 09, 2009, 10:45:55 pm »
Your suggestion for the direction of the conversation is fine by me. The posts were getting too long, and I'm happy to clean things up.

I hate to be this stubborn, but my alarm bells begin to ring even at your first premise. This dividing up of things into the two categories of "necessary" and "contingent" strikes me as very peculiar, and I cannot help but worry that the definitions are ad hoc, and that they have been designed to smuggle in God before any reasoning even begins.

I realize this isn't really an objection, but more of an ambiguous concern. You might however alleviate my skepticism by meeting one challenge: give me an example of an entity that is necessary, but isn't God. Doing this will lend some much needed concreteness to what is otherwise a bewildering category. From there I'll have a better idea of how much I agree with your first two premises.

10

#### loko5

• 207 Posts
##### Cosmological argument seems unconvincing
« Reply #25 on: June 10, 2009, 11:02:53 am »
daveknott wrote: I hate to be this stubborn, but my alarm bells begin to ring even at your first premise. This dividing up of things into the two categories of "necessary" and "contingent" strikes me as very peculiar, and I cannot help but worry that the definitions are ad hoc, and that they have been designed to smuggle in God before any reasoning even begins.

daveknott wrote: You might however alleviate my skepticism by meeting one challenge: give me an example of an entity that is necessary, but isn't God. Doing this will lend some much needed concreteness to what is otherwise a bewildering category. From there I'll have a better idea of how much I agree with your first two premises.

Very well.  I would contend that the laws of logic are necessary, particularly the Law of Affirming the Antecedent or modus ponens ("If P then Q. P. Therefore Q.")  If this were not true universally and at all times, we could never construct any kind of deductive argument.

11

#### David Knott

• 24 Posts
##### Cosmological argument seems unconvincing
« Reply #26 on: June 10, 2009, 02:01:42 pm »
Quite right about premises always being ad hoc. It's just that your categories strike me as so peculiar that I should fully understand them before we proceed anywhere. They could be a false dichotomy.

I contend that your claim that the laws of logic are necessary is untrue, for two reasons:

Firstly, they don't seem to qualify very well as an entity which exists in the normal sense. They are merely abstractions; they have no mass, no size, no spacetime coordinates, nor can they be seen to interact with our world in any sense. They would seem to be on the same playing field as a mathematical law: descriptive but not causal. They have as much reality as the color blue. Do you mean to say that, in the list of things that "exist", we include all things that can be conceived of, whether they exist in the normal sense or not?

Secondly, it seems quite possible to imagine the laws failing to hold. Your premise (2) defines necessary things as things which "could not fail to exist". In other words, it is not sufficient for something to "fail to exist"--it has to be incapable of doing so. How do we know it is impossible for logical laws to fail? Can we not imagine a world of utter chaos, where each moment is causally unconnected to the next, blinking in and out of existence, its character being totally indeterminate?

12

#### loko5

• 207 Posts
##### Cosmological argument seems unconvincing
« Reply #27 on: June 10, 2009, 03:22:05 pm »

daveknott wrote: Firstly, they don't seem to qualify very well as an entity which exists in the normal sense. They are merely abstractions; they have no mass, no size, no spacetime coordinates, nor can they be seen to interact with our world in any sense.

It sounds like you want to say that nothing exists that is not material.  This view, of course, presupposes materialism, which I reject.  I don't really want to get into a discussion of ontology here, let me only say that most people would agree that abstract objects can exist, even if their existence is only in the mind.  In fact, the Leibnizian argument means to show that physical objects cannot exist necessarily.

daveknott wrote: How do we know it is impossible for logical laws to fail? Can we not imagine a world of utter chaos, where each moment is causally unconnected to the next, blinking in and out of existence, its character being totally indeterminate?

OK, let's suppose that there is some pre-existing realm out of which the universe came, and in that realm there are no laws of logic or mathematical laws.  Now, somehow out of this lawless realm the universe appears, complete with order, physical laws, mathematics, and logic.  What could explain these laws popping into existence from some undifferentiated chaos?

13

#### David Knott

• 24 Posts
##### Cosmological argument seems unconvincing
« Reply #28 on: June 11, 2009, 12:30:41 am »
I also do not want to get into a lengthy argument, debating the meaning of existence and materialism. I wasn't trying to argue materialism, but let's just leave this point alone unless we absolutely have to face it.

OK, let's suppose that there is some pre-existing realm out of which the universe came, and in that realm there are no laws of logic or mathematical laws. Now, somehow out of this lawless realm the universe appears, complete with order, physical laws, mathematics, and logic. What could explain these laws popping into existence from some undifferentiated chaos?

Fascinating question, but I don't see how it's relevant. I didn't propose that our universe came from a realm in which undifferentiated chaos ruled (though that is an interesting idea). I merely described such a realm in order to show that logical laws might fail to exist, making them contingent. I might not have much intuition concerning such a realm, and certainly it would be an object foreign to our reason. But this doesn't seem sufficient to fully disqualify it from the arena of possibility. It might be that logic can fail to exist. Or, it might be that logic could be slightly different. In either case, I don't see why logic is a necessary entity. Can you elaborate further on why logic is a necessary entity, or give me another (non God) instance of a necessary thing?

14

#### loko5

• 207 Posts
##### Cosmological argument seems unconvincing
« Reply #29 on: June 11, 2009, 10:11:42 am »
Why don't we just agree to disagree at this point, and go back to the Leibnizian argument.  We can cross these bridges if and when we come to them in going through the argument.

Premise (1) of the argument is of the form:

Everything that exists is either A or not A.

In the case A = "necessary", defined the way I have defined it.  I would contend that premises of this form are always true, as long as we have defined A unambiguously (i.e. the argument would be invalid if we mean A in one sense in the premise then mean A in a different sense later in the argument).  I could say, for example, that everything that exists is either a square circle or not a square circle.  I think you and I would agree that there is no such thing as a square circle, so everything in existence would fall into the non-A category.  I might also point out that it is not necessary at this point that we be able to easily distinguish A from non-A.  For example, I could say that everything that exists was either created in the big bang, or not created in the big bang.  I would anticipate a lot of people disagreeing in this case on whether everything falls into the A category or the non-A category, or some things fall into one category and some things into the other.  This doesn't invalidate the premise however; at worst, it will only cause problems in trying to draw conclusions from the argument.

If we are agreed on this, I think are only two possible objections before we move to step (3).  First, do you agree that the definition I have given for "necessary" is sufficiently unambiguous?  And second, do you agree that necessary things, if they exist, do not require an explanation for their existence?