Philosophical arguments for the existence of god come in all shapes and forms. In fact, since no empirical evidence has ever been presented to support the existence of any deity, pretty much every argument for the existence of god is a philosophical one. But my focus here is on the type of arguments that are dressed up in philosophized jargon – what I sometimes like to call arguments from blinding by philosophy.
As you set out to debunk these arguments, it may amuse you to notice that even deep and well worded religious arguments still employ the same old simple minded fallacies that all other religious argument employ – and how could they not? For this reason, you will not necessarily need to become a qualified philosopher to uncover what are usually elephant-sized problems hiding behind rhetorical shower curtains.
Beyond their various intricacies, there is one simple problem at the heart of every philosophical argument for the existence of a creator – pure philosophical reasoning cannot support literal and physical – and thus scientific – claims about the origin of the universe. You cannot, in other words, make claims that fall well within the purview of physics, while only supporting those claims with pure philosophical thought. It doesn’t even matter if the philosophical reasoning happens to be correct – let’s just for argument’s sake assume that it is – it’s still not strong and solid enough to support such huge claims. Physical claims require physical evidence, and extraordinary physical claims about the origin of the universe require extraordinary physical evidence – measurable, testable, replicable, falsifiable evidence.
Here’s a classic example of how this often looks; taken from one of the best practitioners of modern religious apology, the evangelical Christian philosopher William Lane Craig – promoting what he calls “Reasonable Faith”:
1. Everything that exists has an explanation of its existence, either in the necessity of its own nature or in an external cause.
2. If the universe has an explanation for its existence, that explanation is God.
3. The universe exists.
4. Therefore, the universe has an explanation of its existence (from 1, 3).
5. Therefore, the explanation of the universe’s existence is God (from 2, 4).
6. Therefore, God exists.
If you are familiar with it, you might notice that this is a Thomas Aquinas type deductive argument for god’s existence. A surprisingly common follow-up tactic that many apologists like Craig tend to employ is to immediately follow such arguments with a challenge to nonbelievers to give a better explanation based on atheism. This already should arouse suspicion, since atheism is a simple lack of belief, not a belief system upon which to build. I can no more give an explanation based on atheism – my lack of belief in deities – than I can give one based on my lack of belief in witchcraft, astrology, or fairies. This tactic actually does more harm than good because it only diverts attention from the fallacies in the argument itself (a composition fallacy followed by a Begging The Question fallacy, in the above case) by introducing yet another fallacy. This is a simple attempt to shift the burden of proof away from the person making the claim – a classic burden of proof fallacy (a type of Argument From ignorance); which implies that a religious claim must be granted the assumption of truth, as long as it has not been disproved or replaced by another claim.
You can easily dispose of this tactic by pointing out that the burden of proof lies with the person making the claim, not with the person simply doubting it. Claims either stand or fall on their own merits, regardless of whether or not a counterargument is presented. The Zeus claim about lightning strikes, for example, was no less incorrect before the real causes of lightning were discovered than it is now – and the same applies to all other god hypotheses as well.
Getting back to the actual argument, notice how it asserts that because many things in the universe have explanations, the totality of the universe must have an explanation as well. This is a classic logical mistake, called a Composition Fallacy – claiming that something is true of the whole from the fact that it is true of a part of the whole (or even of every part). The way Craig put it was “The first premise is constantly confirmed in our experience, which provides atheists who are scientific naturalists with the strongest of motivations to accept it”. By “experience”, Craig is, by definition, talking only about parts of the known universe, and then inferring that these parts indicate how the whole is. Notice also how this premise implies that an explanation is an objective faculty that the universe can somehow bestow upon us, rather than our own opinion in regards to an external thing. The universe doesn’t “have” an explanation for us, it is we who come up with explanations about the universe. When called out on his making a Composition Fallacy, Craig has subsequently denied it by stating that “The causal premise is rooted in the metaphysical intuition that something cannot come into being from nothing”, which then obviously begs the question about where god came from in that case, to say nothing of whatever in the world Craig’s own “metaphysical intuitions” might be. As entertaining as it is to watch a trained philosopher try to wiggle out of one logical fallacy by wiggling himself right into another, and another, there is an important lesson to be learned about the shifty and slippery tactics that many religious apologists try to employ; not the least of which (after all else had failed), is the attempt to claim that your refusal to accept their fallacious explanations somehow demonstrates your inability to comprehend them.
As for premise #2, if Craig wants to call his own subjective explanation of the universe “god”, then that is his right, but it is our right to point out that a god premise leading to a god conclusion is no more than a circular argument. We can also point out that in the absence of any supporting evidence to establish the soundness of such a god premise in the first place (thereby Begging The Question), the “explanation of the universe” is no more a god than it is a bowl of spaghetti, a giant turtle, or anything else you want to say it is. Anyone can play that silly game.
Another common feature in many philosophical arguments is the emphasis on the alleged superior explanatory powers of the deductive reasoning process. I have actually had counterparts tell me that deductions constitute irrefutable proof – as if a deduction is both the claim and the evidence for the claim, all rolled into one. A deduction, if you are unfamiliar with it, is comprised of premises leading to a conclusion. Deductive logic, as it’s repeated by many religious apologists, is built on the fact that if the premises are true, and the rules of deductive logic are followed, then the conclusion reached is necessarily true. The mistake that many religious apologists make however, is to conflate validity with soundness – or truth. In general, validity is the term we use to describe the structure of the deduction. A structurally valid deduction is one in which the conclusion does indeed necessarily follow from its premises. But this does not mean that structurally valid deductions always have true premises, let alone true conclusions. For a true conclusion, the deduction needs to be both structurally valid and have premises that are all true. The GIGO (Garbage In Garbage Out) principle holds very firm here. Enter even a single false premise, and you will end up with a false conclusion – regardless of the structural validity of the deduction. And the only real way to establish the truth of your premises is with EVIDENCE; which is precisely the burden that many religious counterparts try to ignore – and have you ignore as well. Deductions, without any external sources of verifiable evidence, simply cannot prove themselves by themselves.
Another thing to note about philosophical religious arguments, is how they almost always pertain to general deistic ideas rather than to any specific theistic claims. This is to be expected, since it is a simple extension of the age old ‘God of The Gaps’ dodge (a type of argument from ignorance). Well worded arguments for god’s existence only seem to make use of the very edges of our current scientific understanding – most notably the origin of the universe. Indeed, many exchanges about the origin of the universe barely end up resembling a religious debate, at all. In order to keep the debate on a religious track, try not to fall for slippery generalizations, vague presuppositions or fuzzy language, like “first cause”. Call it out for what it actually is – a god hypothesis. And not just any god – a very specific god; usually called Yahweh or Allah. Use these names whenever you can, and try to get your counterpart to, reluctantly, use them as well. While you may be called out for employing a Red Herring when you do this, keep in mind that there is nothing that’s actually being misrepresented here, and that this will make it harder for your theist counterpart to hide behind vague deistic abstractions. Craig, for example, has on occasion been asked how his “first cause” arguments can eventually lead to something like the supposed divinity of Jesus Christ. Craig has rightly pointed out that this is a different and more advanced subject for a separate debate, but when pressed further, he usually explains that Jesus’ empty tomb, and his revelations to his followers after his death, prove his divinity.
As off topic as all of this might seem, it illuminates an important point. As learned as your counterpart might initially seem, answers like these will reveal that you are actually debating someone whose standards of proof are quite a bit lower than yours. Craig is a prime example of this – considering how as one of the champions of religious apology, holding no less than a Ph.D. in philosophy, he takes the equivalent of a 2000 year old ‘Elvis sighting’ as irrefutable proof of divinity. I point this out for more than just comedic value, since these low standard of proof, and the various cherry picking and question begging fallacies they inevitably produce, will consistently come up during the philosophical debate itself.
Here is another example of a classic Thomas Aquinas type argument, taken from an actual debate I had:
The grounds for the argument for god’s existence are found in three main assertions:
1. That there is a division between the real “actual” and the conceptual “potential” – there is a distinction between what is and what could be.
2. Secondly, that Aristotelian motion can be described in terms of a relationship between the actual and the potential; that a potential state of affairs is sometimes made into an actuality by the influence of something that is itself already actual.
3. Third, that there is an infinite regression implicit in the relationship between the actual and the potential, implying the necessity for a terminus.
If these three assertions (premises) are true (which I think they are) then it seems to me as though Aquinas’ First Way (“the unmoved mover” – god – conclusion) stands.
My counterpart then goes on to put this deductive reasoning process in formulaic terms in order to point to the necessity of a prime mover (A).
1. A v ¬A;
2.¬A → B;
3. ¬B;
4. ¬¬A
_________
∴ A
You don’t always have to dig that deeply into religious philosophical arguments in order to find some of their biggest flaws. What we have here is yet another beautiful illustration of an argument that isn’t supported by a single shred of evidence, before reaching an enormous objective conclusion about the origin of the universe. I call it beautiful because there’s something almost poetic about it; you almost have to pinch yourself in order to realize just how silly it is to try to balance such an enormous objective claim on no more than deductive abstractions. Exactly zero empirical evidence – or even a single measurable quantity – is presented before we are told that a supernatural deity must have created the universe.
Even without pointing out the classic logical fallacies in arguments of this sort (Bifurcation and Non Sequitur, in the above case), and even without pointing out that recent scientific discoveries have pretty much disproved religious claims about the origin of the universe, try simply asking your counterpart if their argument includes any measurable, testable or demonstrable quantities. Then remind your counterpart that a claim that cannot be measured, tested or demonstrated, is an unfalsifiable one, and unfalsifiable claims are also unprovable. If this doesn’t do the job, challenge your counterpart to find a single example of an unmeasurable, non testable, non demonstrable claim that is widely accepted as objectively true.
You can pretty much anticipate that when challenged in this way, many religious counterparts will defensively resort to claiming that their arguments are metaphysical, or non scientific, and that they therefore do not conform to the rules of science. This is pretty much when the debate reaches its end, since claiming such a thing is tantamount to forfeiting the debate altogether. A god hypothesis pertains to the creation of all matter and energy, and is therefore most definitely a scientific claim – requiring scientific evidence. You will often find that the only people who argue that no evidence is necessary to support their objective truth claims are the ones who tried and failed to produce any.
The age of widely accepted armchair deductions about the origin and nature of the universe has long since ended, and legitimate deductive arguments these days belong in one of two categories: a framework for empirically measured and verifiable premises and conclusions; or a structured hypothesis that has yet to be proven by empirical verification. Hypotheses of this latter type are not necessarily wrong, they’re simply unproven until the premises and the conclusion are externally and empirically verified.
When confronted with philosophical arguments about the objective origin and nature of the universe, you should always remind yourself that it is the soundness and strength of the EVIDENCE that will determine the soundness and strength of the claim, not the other way around. Calling a claim about the universe non scientific, or metaphysical, or philosophical, is just another way of calling it unproven, or subjective – or imaginary.
Ami Toben 
How nice to be referred to in an article 🙂
I did, however, have a question regarding the article’s contention that Dr. Craig’s argument commits the fallacy of composition. You say that the argument “asserts that because many things in the universe have explanations, the totality of the universe must have an explanation as well.”
As I understand it, the fallacy is structured like so:
1. Component A is a part of whole B;
2. Component A has property C;
Therefore, whole B has property C
In a pertinent example, it might look like:
1. Philip is part of the universe;
2. Philip has an explanation for his existence;
Therefore, the universe has an explanation for its existence
Even if we replaced “Philip” with all components of the universe, the form would generally remain the same. Any particular instance of the fallacy of composition appeals to a distinction between the component and the whole, along with a particular property held by the component (which is then said to be true of the whole).
Are you saying that this is evident in Dr. Craig’s argument as it is written here in those six statements, or are you referring to a specific defence of it that he made elsewhere? (i.e. there’s no obvious inference from component to whole in the argument in the article, so where does he actually say this?)
Philip
“1. Everything that exists has an explanation of its existence, either in the necessity of its own nature or in an external cause.”
This “everything that exists”, to the extent of its known existence, is contained within the known universe, and is therefore a part of the universe.
This “everything that exists” does indeed include components of the universe, but it is also said to include the universe itself.
Consider premises #1, 3, & 4 (re-numbered here, leaving out the God bit for a minute):
~
1. Everything that exists has an explanation of its existence, either in the necessity of its own nature or in an external cause;
2. The universe exists;
Therefore, the universe has an explanation of its existence.
~
As described earlier, the fallacy of composition follows the pattern:
1. Component A is a part of whole B;
2. Component A has property C;
Therefore, whole B has property C
e.g.
1. All of the individual Lego bricks are part of the giant Lego castle
2. All of the individual Lego bricks have a length of 8mm
Therefore, the giant Lego castle has a length of 8mm
Now, the problem with such arguments is that they shift the unit of analysis from the components to the whole. There’s generally a categorical proposition involved (such as statement #2 with the Lego bricks), but the error is in conflating A and B – a Lego castle may be made of Lego bricks, but that doesn’t mean it *is* a Lego brick.
We can accept the categorical proposition “All of the individual Lego bricks have a length of 8mm,” but that won’t necessarily transition over to the giant Lego castle precisely because the giant Lego castle is not included in the category of “individual Lego bricks.”
Dr. Craig’s argument, however, does not (as far as I can tell) rely upon treating the whole as a component of itself, even though they fall under the same categorical proposition. The universe (which is the whole) is not included in the category “components of the universe.” Rather, they both belong to a wider category, namely “things that exist” (even though he never even discusses components vs. whole). Components of the universe exist, as does the universe. From what I can tell, it follows the form:
1. All A’s are B’s
2. X is an A
Therefore, X is a B
1. Everything that exists has an explanation for its existence;
2. The universe exists
Therefore, the universe has an explanation for its existence.
It isn’t “components of the universe have explanations, ergo the universe does”; it’s “everything that exists has an explanation; the universe exists; ergo the universe has an explanation.” Parts and the whole are both considered to be existent things. I don’t really see where an inference from components to the whole is being made.
In summary, the fallacy of composition is:
1. Component A is a part of whole B;
2. Component A has property C;
Therefore, whole B has property C
Whereas Dr. Craig’s argument with regards to explanations is:
1. All A’s are B’s
2. X is an A
Therefore, X is a B
Of course, one can dispute the premises that “Everything that exists has an explanation of its existence” or that “the universe exists.” They may be false, but I don’t think there is a fallacy of composition here – the form isn’t the same, and component/whole considerations are never even mentioned.
//It isn’t “components of the universe have explanations, ergo the universe does”; it’s “everything that exists has an explanation; the universe exists; ergo the universe has an explanation.” Parts and the whole are both considered to be existent things. I don’t really see where an inference from components to the whole is being made.//
One problem you might want to consider is that you, and Craig, are attempting to limit the definition of what the universe is in order to make room, and create a necessity, for a god character. What exactly is a thing that exists outside of our known universe? To the extent that we know a thing exists, and to the extent that it can be shown to exist at all, it must belong to, and be a part of, the known universe. And yes, the boundaries of our known universe keep expanding as we gain more information. It might even turn out that what we call the ‘known universe’ turns out to be an eternal multiverse, but anything new we discover will simply be an expansion of our concept of the universe, not something outside of it.
Carl Sagan put is well when he said “The Cosmos is all that is or was or ever will be.”
You are spinning your wheels.
I don’t immediately see how anything you’ve said in this post is relevant to the question of whether or not Dr. Craig is committing the fallacy of composition in arguing that there is an explanation for the universe through the first, third, and fourth statements in his argument. There’s nothing in this portion of the argument that suggests the universe cannot be eternal, nor is there a distinction between parts of the universe and the whole.
All that is said is that (1) things which exist have explanations of their existence, and (3) that the universe exists. There is no attempt to “limit the definition of what the universe is in order to make room… for a god character” – in fact, there is no definition of the universe offered apart from the statement that it exists.
Here’s a rather simple question… Do you think the below is an accurate representation of Dr. Craig’s sub-argument regarding the contention that the universe must have an explanation?
(Craig’s argument):
“1. Everything that exists has an explanation of its existence, either in the necessity of its own nature or in an external cause.
3. The universe exists.
4. Therefore, the universe has an explanation of its existence (from 1, 3).”
(Philip’s suggested form):
1. Every A is a B
2. X is an A
Therefore, X is a B
//All that is said is that (1) things which exist have explanations of their existence, and (3) that the universe exists. There is no attempt to “limit the definition of what the universe is in order to make room… for a god character”//
Can you think of any “things that exist” which aren’t contained in the known universe, and which are therefore not PARTS of the universe? This is still just an attempt to define the universe, based on the definition of parts it contains.
Things that exist which are not “parts” of the universe? The universe itself.
The notion of the “fallacy” of composition rests upon there being a distinction between the whole and its parts, which is why I can make a universal proposition such as “All individual Lego blocks have a length of 8mm” and not have it apply to the giant Lego castle (the castle, though it is made of individual Lego blocks, is not itself an individual Lego block).
~
“This is still just an attempt to define the universe, based on the definition of parts it contains.”
Er, I don’t think so. There is no “definition” of the universe provided, other than the statement that it is something which exists. There is no inference being made from constituent parts to the whole (i.e. “components of the universe have explanations, therefore the universe has an explanation”).
“The Cosmos is all that is or was or ever will be.” – Seems to me like that implies the existence of the “Cosmos.” It *is*, it *has been*, and it *will be*.
So when Dr. Craig says “Everything that exists has an explanation of its existence,” the subject is not simply “parts of the universe.” It’s “everything that exists,” which is taken to include even the universe considered as a whole as per the third premise.
That’s assuming I’ve understood him correctly, of course. You didn’t comment on whether or not you think I represented the form of his argument correctly.
Let’s try this again:
“Everything that exists”, to the extent of its existence, is contained in the concept of the known universe/cosmos. You can either talk about parts of the cosmos or the totality of the cosmos. If something isn’t a part of the cosmos or isn’t the totality of the cosmos, it doesn’t exist.
Everything that exists = all the parts of, or the totality of, the cosmos.
When Craig says:
//1. Everything that exists has an explanation of its existence, either in the necessity of its own nature or in an external cause.
2. If the universe has an explanation for its existence, that explanation is God.//
he could either be talking about everything – as in practical/experienced things you can think of – as in parts of the cosmos, or everything – as in the totality of the cosmos. Those are the only two logical options. In the former case (which I think he is promoting) he is making a composition fallacy, in the latter, he is making a circular argument. In either case, he adds a non sequitur in the form of “..that explanation is god”, which is his ironic way of claiming to know something he admits he cannot logically know, since that thing is OUTSIDE of the known cosmos.
//“This is still just an attempt to define the universe, based on the definition of parts it contains.”//
Sorry about that. I should have restated this: “This is still just an attempt to define a certain characteristic of the universe, based on the definition of a characteristics of parts it contains.”
If by the “totality of the Cosmos” you mean the consideration of the universe as a single albeit composite thing (e.g. considering the Lego castle as a single thing as opposed to thinking of its individual parts), then I would say that this is indeed something that Dr. Craig is referring to when he says “everything that exists.”
“he could either be talking about everything – as in practical/experienced things you can think of – as in parts of the cosmos, or everything – as in the totality of the cosmos. Those are the only two logical options.”
I don’t see why those are the only two options. I see a third as being more congruent with both the first and third premises, namely that he is simultaneously speaking about both. Individual parts of the universe such as mountains, planets, ponies, stars etc. all fit the bill for “things that exist,” but the universe considered in its totality also does.
Oh, and I’m sure he would use the term “universe” in a fashion other than what you’re currently thinking when you refer to the Cosmos – but that isn’t the issue. My point is that the actual inference from part to whole isn’t present in the argument.
“The universe has an explanation” is not derived from saying that parts of the universe have an explanation – no such inference made here. Rather, it is derived from the categorical proposition that all things which exist have explanations: “All A’s are B’s.” The universe, considered in its totality as a single “thing,” is an “A.”
Things which are A’s include not only parts of the universe such as ponies, mountains, planets etc., but also the consideration of the universe in its totality as a single thing. I would think that the response to Craig’s argument with regards to explanations is relatively obvious – a denial of the statement that “everything that exists has an explanation of its existence” would suffice. Or perhaps a denial that the “the universe” in its totality actually exists (but that seems slightly less plausible.
Perhaps the previous post should be ignored; this might be clearer, as I didn’t make it explicit in the first that Dr. Craig would most likely disagree with your very first statement.
“Everything that exists = all the parts of, or the totality of, the cosmos.”
Essentially, the problem is that Dr. Craig wouldn’t agree. In fact, he might actually explicitly state this in the book. I think “the universe” is meant to be taken here as a particular thing, even though it’s a composition. It’s just one more composite object (on a vast scale) considered as being a member of the category “everything that exists” – it isn’t the category itself. “Universe” is not essentially meant to describe “all that exists.” Carl Sagan is not William Lane Craig, so taking definitions from Carl and using them to re-define what Dr. Craig is saying doesn’t really fly.
“The universe,” as well as all of its constituent components, are considered to belong to the category of “things that exist.” That does not necessitate the proposition that they *are* the category itself – I think that’s your definition/assumption (or Sagan’s), not Dr. Craig’s.
Despite any other flaws that the argument might have (such as the allegation of circular reasoning you mentioned, or even just plain question-begging with the second premise), I still don’t see an indication that Dr. Craig is actually making an inference about the entire universe based on statements about its constituent parts. Perhaps he is completely wrong, and the universe does in fact exist by the necessity of its own nature. Or maybe his first premise is just plain false – but that still doesn’t mean he’s committing a fallacy of composition.
Let me somewhat rephrase my previous explanation:
Craig “supports” his first premise by both a composition fallacy AND a question begging fallacy (not either or, as I previously said). This leads me to the conclusion that I might want to edit my original article to account for this double, rather than single, fallacy – which is exactly why I appreciate your comments so much (in case you thought I can’t admit getting something wrong – or slightly wrong).
In case you are wondering how I came to this conclusion, here is Craig himself defending his premise:
//…let me first review three reasons I have given for believing the first premiss of the kalam cosmological argument. First and foremost, the causal premiss is rooted in the metaphysical intuition that something cannot come into being from nothing. To suggest that things could just pop into being uncaused out of nothing is to quit doing serious metaphysics and to resort to magic.//
It’s almost needless to say that Craig’s “metaphysical intuitions” simply won’t cut it here, as a support for such a premise. This is question begging raised to new heights.
//Second, if things really could come into being uncaused out of nothing, then it becomes inexplicable why just anything and everything do not come into existence uncaused from nothing.//
Craig seems to be oblivious to the possibility of a singularity having occurred, or to conditions being different after this singularity, which is part of the reason this supposed “reason” isn’t even worth mentioning or refuting.
//Finally, the first premiss is constantly confirmed in our experience, which provides atheists who are scientific naturalists with the strongest of motivations to accept it.//
To “confirm it from experience” in scientific terms, is to confirm it through observation and/or experimentation – which can only be done to PARTS of the universe – leading to a composition fallacy. For Craig to claim that this last reason “provides atheists who are scientific naturalists with the STRONGEST of motivations to accept it”, is the reason why I gave it so much attention in the first place.
That’s the first premise of the Kalam cosmological argument. The argument you referred to in your article is a different argument, but Kalam is the one Dr. Craig is famous for, so maybe the argument currently dealt with in the article (Leibniz) should be exchanged for the more popular Kalam?
But if we proceed with the treatment of the first premise in Kalam, I still don’t think it commits the fallacy of composition. Setting aside his first two points regarding the premise (intuition and something-from-nothing), it seems to me as though the third point would rely upon simple induction rather than an argument from composition.
An argument from composition relies upon drawing a distinction between the whole and its components, and then saying that the whole must possess the property of its components by virtue of the fact that it is comprised of them. For instance, the classic example (I realise that some cells are actually visible, but this is for the form):
1. Human bodies are made from cells;
2. All cells are invisible to the naked eye;
Therefore, human bodies are invisible to the naked eye
or, in pertinent terms
1. The universe is comprised of things that begin to exist;
2. Things that begin to exist have causes;
Therefore, the universe has a cause
The conclusion(s) don’t follow because the whole isn’t an individual part. A human body is not a cell; there’s no premise saying that the universe is a thing that begins to exist, even though it is comprised of them. The key element here is that even though all of the individual components of a whole are said to belong to a specific category (“All cells are things that are invisible to the naked eye”), there’s nothing in the premises of the argument which would necessitate the conclusion that the whole (the human body) must also belong to that category (invisible to the naked eye).
That’s why the conclusion doesn’t follow from the premises regarding parts/whole, making it a fallacious argument from composition. Dr. Craig’s third point regarding the Kalam premise, however, relies upon not upon composition but rather enumerative induction:
1. All observed swans have white feathers;
(likely) All swans have white feathers.
1. All observed things that begin to exist have causes;
(likely) Therefore, anything that begins to exist has a cause.
The universe isn’t even mentioned in the specific defence of the first premise, but this is presumably followed by the Kalam itself:
1. Whatever begin to exist has a cause;
2. The universe began to exist;
Therefore, the universe has a cause
The claim that the universe had a cause is not made in light of the claim that it is composed of things that began to exist and therefore had causes, but rather by virtue of the statement that it *is* a thing which began to exist.
Both “the universe” and “parts of the universe” belong to a wider category, namely “things which begin to exist.” Needless to say there are a number of objections one could raise (e.g. lack of explanation as to why “the universe” must belong to the category, or perhaps insufficient sample size upon which to justify the inductive inference in the first premise). What is conspicuous by its absence, however, is not an appeal to composition, but even a mere recognition of composition.
So essentially, Craig’s third point in defence of the first premise is from an inductive inference, not an appeal to composition.
1. All observed things that begin to exist have causes;
(likely) Therefore, anything that begins to exist has a cause.
A generalised statement is derived from many particular observations. There is no distinction between parts and the whole, hence no fallacy of composition is present in the defence of the first premise. In fact, the notion of “the universe” doesn’t even enter into the defence of the premise.
Nice try.
The problem is that the inductive inference “All observed things that begin to exist have causes; (likely) Therefore, anything that begins to exist has a cause.” talks about observed things WITHIN the universe (as I keep mentioning). And things within the universe are PARTS of the universe. This inductive inference about parts of the universe is then stretched to fit the whole universe itself, and voila – we have a composition fallacy. Now, if you can show this principle applying to any observed things that AREN’T parts of the universe, then that’ll change things (mostly in the form of a paradox, though).
Maybe it’ll help you understand this if I use the very analogy you yourself used:
1. Human bodies are made from cells;
2. All cells are invisible to the naked eye;
Therefore, human bodies are invisible to the naked eye.
What you fail to take into account is that our perspective here is as a cell trying to look at the body – which the cell cannot see beyond. In other words, you (the cell) are trying to apply to the universe (the body) a principal (inductive inference) that has only ever been observed in regards to PARTS of the universe (cells). You are demonstrating exactly how you (and Craig) are committing the fallacy of composition.
I still don’t agree because as I said, the argument makes no statements about the composition of the universe. It doesn’t even consider it. The Kalam argument’s unit of analysis is not “parts of the universe,” but rather “things that begin to exist.” The only members of that set which we are aware of happen to be parts of the universe, yes – but that would simply imply in a rather weak induction, not an appeal to composition.
A fallacious argument from composition asserts that whole X necessarily has property Y because its components do. This is reflected like so:
“Whole X” composed of: (1, 2, 3, 4, 5)
“Universe” composed of: (mountain, pony, Philip, computer, cellphone)
…In which components 1, 2, 3, 4, and 5 (or the listed objects) have the property of “having a cause.” The fallacious argument from composition would assert that the universe (the whole) must have a cause by virtue of the fact that it is composed of things that have causes.
The Kalam argument, however, says nothing about the composition of the universe. It does not say that whole X must have property Y because it is composed of things that have property Y. It has a fundamentally different form:
Members of “Category X”: (1, 2, 3, 4, 5)
Members of “Things that begin to exist”: (mountain, pony, Philip, computer, universe)
In an argument from composition, the inference runs from parts to whole – the objects are the parts, and the universe is the whole. But in an inductive argument in which the unit of analysis is neither strictly “parts of the universe” nor the whole, the universe is instead treated as just one member of the category rather than the category itself. Member #5 is said to have property Y because members 1, 2, 3, and 4 have it, and they belong to the same category. This doesn’t fit the logical pattern of argument that would be a fallacy of composition.
Perhaps it might be illustrated through another example. Is the airbase covered under the Military Asset Protection Act?
~
The fallacy of composition:
“Airbase” composed of: [planes, trucks, pillboxes, rifles]
All components of airport are covered by the Military Asset Protection Act
Therefore, the airbase is covered by the Military Asset Protection Act
~
vs.
An inductive argument
Members of class/category “Military Property”: [planes, trucks, pillboxes, rifles, airbase]
Planes, trucks, pillboxes, and rifles are covered by the Military Asset Protection Act
(likely) Therefore, the airbase is covered by the Military Asset Protection Act
The conclusion that the airbase is covered by the MAPA (which I just made up?) is not reached by an appeal to the fact that it is *composed* of things which are covered by it, but by the fact that the airbase belongs to a wider category in which the other members are covered.
I realize we probably won’t agree on any of this, so perhaps it would be best to move on. But before that, I’ll just say that although I don’t think Dr. Craig is committing the fallacy of composition, the style of the argument I referred to does engage in a hasty generalization. Consider an argument with identical form:
Members of category “Humans”: [Mary, Kate, Ashley, Sally, Peter]
Mary, Kate, Ashley, and Sally are all capable of bearing children
(likely) Therefore, Peter is capable of bearing children
Though they all belong to the same category, there are potentially important distinctions that have been completely ignored. Not only is there an extremely limited sample size, but substantive differences in sub-categories (male vs. female) prevent the inductive inference from being justified.
//The Kalam argument’s unit of analysis is not “parts of the universe,” but rather “things that begin to exist.” The only members of that set which we are aware of happen to be parts of the universe, yes – but that would simply imply in a rather weak induction, not an appeal to composition.//
It might “imply” that in your mind, but I have to thank you for just conceding the point. “The only members of that set which we are aware of happen to be parts of the universe, yes” There we have it. Now, if you want to claim that this “implies” a weak induction or a strong connection or a pink elephant or a giant spaghetti bowl, that’s your right, but you will not get around the fact that you are inferring something about the universe from what is true about “parts of the universe” – your words.
As I already covered in a previous comment, Craig does a better job than you of trying to wiggle out of the composition fallacy, but only by wiggling right into a question begging fallacy. You seem to be trying to do the same, but not as “well” as Craig.
“The only members of that set which we are aware of happen to be parts of the universe”
Sorry, I was actually incorrect here – scratch that. The argument says that there is another known member of the set, namely the universe itself.
“Members of “Things that begin to exist”: (mountain, pony, Philip, computer, universe)”
I’m not sure how else I can put this. Consider the argument dealing with cells again:
1. The human body is made from cells;
2. All cells are invisible to the naked eye;
Therefore, the human body is invisible to the naked eye
– Human Body composed of: [liver cells, kidney cells, brain cells, muscle cells etc.]
– Liver cells, kidney cells, brain cells, muscle cells are all invisible to the naked eye
– Therefore, the human body is invisible to the naked eye
^ Fallacy of composition.
____________________
vs…
Members of category X: [1, 2, 3, 4]
Members 1, 2, and 3 have property Y
(likely) Therefore, member 4 has property Y
Members of Category “Living Things:” [Liver cells, kidney cells, brain cells, body]
Liver cells, kidney cells, and brain cells are invisible to the naked eye
(likely) Therefore, the body is invisible to the naked eye
^ Not the fallacy of composition. The argument is an inductive inference dealing with a set number of members from a spericifc class and makes a probabalistic inference about one member based on prior observations. No appeal to composition – or even a statement that the body is something with parts – is made.
The parts/whole distinction is simply not considered in the argument at any point. The argument relies upon both “cells” and “body” being constituent members of an identifiable category; it does *not* rely upon the contention that the body is composed of cells.
Do you see *any* premise in this argument which states that the body is composed of cells?
That was pretty funny.
For starters, if the universe is part of the “set” here, then this is nothing more than a pointless circular argument – the universe is a part of a set of things that begin to exists, therefore the universe began to exist. I keep reminding you that the only way to wiggle out of the composition fallacy here is by wiggling into a question begging type of circular argument – the type you keep trying, but not quite succeeding, in making.
And even if you did manage to wiggle your way out of this composition fallacy, you will still have a question begging fallacy followed by a non sequitur. I still don’t understand why you would want to defend something like that.