Cosmological
Arguments:
Question: "What is the Cosmological argument
for the existence of God?"
Question: "What is a Cosmological argument (for
the existence of God)?"
There are really two issues here.
Typically, cosmological arguments attempt to prove that something exists (a
previously unknown, overlooked something perhaps), and then identify that “something”
with God. But that’s really a two-stepper.
It is one thing to prove, say, that an uncaused cause exists. It is another to prove that this uncaused
cause is God. St. Thomas Aquinas is
unjustly characterized as skipping over this gap when, after concluding his
argument from motion and demonstrating that there must exist an unmoved mover,
he asserts “and this all men call God.” “Wait a minute.” some have said, “Even
if you have established the existence of an Unmoved Mover or an Uncaused Cause,
etc. how do we know that this thing is
God? How do we know it is personal,
conscious, cares about the affairs of the world or humankind? How do we know it is moral or intelligent
etc.?” There’s not only a “gap,” but a
huge one facing a person who tries to move from knowledge of “Necessary Being”
to the “O-God.”
While the gap is legitimate, the criticism of Aquinas is not. He was merely stating a point of historical
fact, a fact Aristotle himself suggests, that all people who consider the
“unmoved mover” argument (and related ones) have concluded that the unmoved
mover is God. As a matter of historical
fact, all men have called this thing “God.” Nevertheless, as William Rowe and
others point out, it is one thing to prove that there is an unmoved mover or an
uncaused cause. It is another to
demonstrate that this thing/ being is God (capital “g” and in the singular) and
still another to demonstrate that this being is a fit object of worship, or the
God referred to in the Bible, etc. Rowe
refers to these two distinct projects as Cosmological Argument Phase 1 and
Cosmological Argument Phase 2 respectively.
But what is a/the Cosmological Argument?
First, the indefinite article is appropriate, and the definite
article is not. There is no “one”
cosmological argument, but rather a family of related arguments under this
umbrella, if you will. Within the family
there are three broad divisions, and, indeed, one subgroup fights with the
other two…sort of. But before talking
about their differences, let’s examine what they have in common.
A cosmological argument begins by observing the world around us
(the cosmos). Thus it is an empirically based argument and not an a priori argument as an ontological
argument is. A cosmological argument
begins with some obvious unassailable observable fact about the world such as
“things exist.” It is then argued that the ultimate cause of/ explanation for
those things’ existence has to be some transcendent thing(s) (Supernatural/ not
of the natural order) and it has or they have some property that only a "God-type"
thing could have. These types of arguments go all the way back to Plato and
have been used by notable philosophers and theologians ever since.
One important difference between the argument types are whether
they are, say, "vertical" or "horizontal." These names
indicate the direction from which the explanation/cause for the feature being
explained/ caused comes. In the vertical form, it is argued that every created/
moved/ caused thing is being caused right now.
The horizontal version, by contrast, claims that the cosmos had to have
been caused to exist at some time in the finite past[1].
However, before going into the details of these further, it will
first be necessary to appreciate a number of different items:
Modal
Terms
Modals—words that express modalities—qualify a statement. For example,
the statement "John is happy" might be qualified by saying that John
is usually happy, in which case the term "usually" is functioning as
a modal.
Necessary/
Contingent and Possible/ Impossible
Here we need to discuss necessary vs. contingent existence. Here “necessary” and “contingent” are acting
as modal terms modifying HOW the thing is existing. William Rowe offers a four-part distinction
which is most helpful:
1.
Necessary Things: Their nonexistence is not possible—they cannot
not be
2.
Contingent Things: Both their existence and their nonexistence is
possible.
3.
Possible Things: Everything
that does exist or could exist.
4.
Impossible Things: Their existence is not possible—they cannot be
To explain these concepts, William Rowe invites us to think about
a double-sided list. On the one side we
put “Things that exist” and on the other we put “things that do not
exist.”
|
|
|
|
A |
Empire State Building, The Planet Mars Dogs Justin Bieber |
Unicorns Mermaids Living Dinosaurs The Fountain of Youth |
|
B |
Triangles (?) Anselm’s God (?) |
Round Squares Married Bachelors |
But note there is an important difference between the items that
find themselves in Row A and those that find themselves in Row B. The items in Row A can “switch sides.” That is, while the Empire State Building
exists, it could “not exist” and in that case, it would be on the other
side. Indeed, it was on the not exist
side before it was built. But the items
in row B cannot switch sides. That is,
while married bachelors are on the non-exist side, there is no way they could
end up on the exists side. In the former
cases, row A, everything listed on either side is contingent. That is, their existence or non-existence is
contingent. Contingent upon what? Well that varies: history, evolution, laws of
physics, etc. It is an accident of
history that unicorns do not exist. Had
evolution gone differently, they might have.
So their nonexistence is contingent upon history and evolution. But in the case of row B, these items’ existence
or non-existence is NOT contingent.
Rather, it is either necessary (not contingent upon anything) or
impossible (again, not contingent on anything).
Now, to be clear, some philosophers claim that quadrant 1B us empty[2],
that there is no such thing as a necessary thing.
Cosmological Arguments claim that there MUST be (at least) one.
|
1. Things That Exist |
2. Things That Don’t Exist |
|
|
A. Things that can Switch
Sides |
A1 |
A2 |
|
B. Things that Can’t Switch
Sides |
B1 |
B2 |
Necessary/ Contingent and Possible/ Impossible
Philosophers do not all agree on the criteria for asserting that
something is possible or impossible, etc.
Modern philosophers usually have in mind logical possibility, but
that may not be the same thing as “metaphysical” possibility. Nevertheless, the table still works
regardless of how you understand the mechanism for possibility.
As mentioned, some philosophers claim that 1B is empty, that there
is no such thing as a necessary thing.
Recall that Hume had claimed that anything we can conceive of as
existing we can conceive of as not-existing.
Thus, he claims that it is possible for any existing thing to be on the
“other side of the line.” This is why
you cannot prove the existence of anything a
priori according to Hume.
But Cosmological Arguments claim that there MUST be something in
there (1B).
Another term that is will useful to cover before looking at
Cosmological Arguments is “Infinite Regress.”
This is the text of an email that I sent to a student on this
subject and I still like it so…
There is a story told in Stephen
Hawkings’s A Brief History of Time
where a cosmologist's speech is interrupted by a lady who claims that the
universe rests on the back of a giant tortoise.
The scientist gave a superior smile before replying, “What is the
tortoise standing on?” The old lady shoots back:
“You’re very clever, young man,
very clever,” said the old lady, “but turtles all the way down.”[3]
That's the problem of an infinite regress. A turtle cannot support something unless it
too is in turn supported. But that can't
be the whole of the story: supported things in turn supporting other
things. If anything supports anything,
there must be, in addition to supported supports, something different,
something which supports, but does not itself need to be supported. An unsupported supporter.[4]
Put another way, turtles have no ability to support anything on
their own; rather they borrow their ability to support from whatever is
supporting them. They derive their
supporting powers from some other supporter.
Can it be the case that everything that supports is itself
supported by another thing? No because
that would lead to an infinite regress, the logical consequence of which is
"Nothing is supported." But
the claim "Nothing is supported" is obviously false so any position
that that logically entails it must rest on false assumptions (at least
one). This is essentially the engine of
Aquinas' first three “ways.” ( i.e. there must be an unmoved mover, there must
be an uncaused cause, necessary being.)
Infinite
Regress in Ethics:
But the problem of the infinite regress is not just seen in
metaphysics. Aristotle and Aquinas
employ it in their Ethics as well. They
note that some of the things one values only because they are a means to some
other end (instrumentally). Money is a good example. We certainly value money, but we only value
money because it is a means to other things.
Money that we cannot use to get or to do stuff is literally
worthless. Money has no “intrinsic value,”
but only has derivative value.
In addition to the things you value instrumentally, that is, as
means to some other end (say, money), there must be something you value as an
end in itself and not merely as a means.
Were I to ask you “Why are you taking this class?” you might respond,
“To clear a requirement.” And if I asked
you why do you want to do that, you might respond, “So I can complete my FIU
degree.” And why that?... But you see where this is going. There’s got to be and end somewhere. At some point I will ask, “Why do you want
that?” and you will respond “Because I just do.” If this is the case, then the
value of taking this class is derived from something else (i.e. clearing the
requirement). That derives its value
from something else (finishing your degree at FIU) and so on. But this cannot be an infinite series. It must terminate in a thing with underived
value, something which can bestow value, but does not have to get its
value from something else.
Philosophers distinguish between these two important senses of
value with the terms “instrumental value” and “intrinsic value.” A thing has instrumental value if it is
valuable as a means to some other end, that is, for some other reason.[5] A thing has intrinsic value if it is valuable
for itself not merely for a means to some other end. The mere fact that ANYTHING has instrumental
value demonstrates that there is something of intrinsic value. Nothing can have instrumental value (derived
value) unless there is something of intrinsic value (underived value). The only question remaining is what the
nature of that thing (or things) of intrinsic value is. A thing of intrinsic value, and specifically
non-derivative value is the only thing that that can keep the whole system of
"valuing things" from collapsing, or perhaps more accurately, ever
getting started in the first place. And
note further, this is not a “temporally” first priority; this is a logically
first priority. That which “causes”
money to have instrumental value does so simultaneously with money having
instrumental value. It’s not because
something had intrinsic value at some point long ago that makes something have
instrumental value today. Quite the
contrary: money has no intrinsic value; it only has derivative (instrumental)
value and that is only possible if something else, right now, has
non-derivative, intrinsic value from which money derives its value (right now).
Now,
on to the Cosmological Arguments Themselves:
Horizontal
Cosmological Arguments
The horizontal is a little easier to understand since it does not
require much in the way of heavy metaphysical thinking. The basic argument can be put in the form of
a syllogism.[6]
Therefore
Notice that the syllogism is deductively valid. If the premises were true, the conclusion
would HAVE to be true. Once we know that
such a cause must have existed (The Phase 1 Cosmological Argument), we can
deduce certain qualities about that cause, qualities that are of the God-type
(Phase 2 Cosmological Argument). For
instance, such a cause must exist outside space and time, and yet be
responsible for space and time. Similarly,
it must be immaterial.[7]
Someone might object that the cause of the universe might itself
have a cause and so on, and so on and so on.
But such a picture would leave us with no real explanation for the
universe at all. This is because caused
things had to have causes, too, and this cannot go on forever. Such an
explanatory model would result in an “infinite regress” and thus fail to
explain anything at all. The only thing
that can stop the regress would be an uncaused cause … and that all men
call God. J
Why should we think that the second premise of the syllogism is true? Well today, we have some science that seems
to support it. Specifically, the Big
Bang Theory[8]
seems to suggest that the Universe is not infinitely old, but rather only
finitely old (about 14 billion years?).
But philosophers have offered philosophical arguments for this position
prior to the claims of Modern Physics.
The Universe cannot be infinitely old, according to this version of the
cosmological argument (The Kalam
Cosmological Argument) because such a picture would commit you to the existence
of a completed infinite set. All series
are finite (limited) by definition. All series have two endings in actuality—at
the end and at the beginning. But if there were no temporally first cause, the
chain of causes never would have started. Therefore, there is, at the beginning
at least, a first cause—one that had no beginning. This first (uncaused) cause
is God.
There is general agreement that this version of the cosmological
argument finds its origins in the thinking of an Alexandrian commentator and
Christian theologian, John Philoponus (d. 580?) and
later the Islamic Kalam theologians. Kalam Theologians wished to lend
philosophical respectability to the teachings of the Quran. Thus they sought to
demonstrate with philosophical proofs that the world was created, and, like the
Christian Philoponus[9],
wished to demonstrate that the philosopher Aristotle was wrong to claim that
the universe had no beginning in time.
To this end, they borrowed a line of argument from Philoponus
which he presented in his book, Contra Aristotelem. Not well known to us, he was well known
among Islamicists as the source of much of the Kalam argumentation against the Aristotelian conception of the
eternity of the world.
The early arguments go something like this:
Motion cannot be from eternity,
for an infinite temporal regress of motion is impossible, since an infinite
cannot be traversed.
Imagine a staircase the top of which you cannot see, but someone
has descended it and tells you that the staircase is infinitely long and he has
traversed the whole of it. “But that is
impossible.” you would say. However many steps he may have descended, it
must be a finite number of steps between where he started and where he and you
are now. He could never have reached you
otherwise. Well, the thinking is that
the arrow of time is the traveler and the moments of time are the steps. For the arrow of time to meet us today, there
must be only a finite number of moments between now and the beginning of time. So the universe cannot be infinitely old, but
rather had a beginning.
Vertical
Cosmological Arguments
The vertical form is a bit more difficult to understand, but it is
thought by some to be more powerful because not only does it show that God had
to cause the "chain of causes" in the beginning, He must still be
causing things now. I’ll sketch the Cosmological Argument that uses the existence
of contingent things as its starting place.
Again, we begin by noting an obvious empirical fact: “things
exist.” Next, while we often tend to think of existence as a property that
things sort of "have"—that once something is created, existence is
just part of what it is (what Edward Feser terms “Existential Inertia”) —but this
is not the case. Consider my dog. I
might define my dog as a black and brown dog, with a waggily
tail, four legs, two eyes, a low IQ, etc.
But no matter how detailed my definition/ description of my dog is, notice
what is NOT part of this definition: existence.[10]
This definition of my dog would be perfectly cogent even if my dog
did
not exist. Therefore, my dog’s
nature —what he is—does not guarantee that he is, exists (Think
of unicorns. —We know what
they are, but that does not make them exist).[11]
Because it is not part of my dog’s or unicorns’, or President Obama’s nature (essence)
to exist (esse), these things must be made to exist (if
they exist at all) by something else that already exists. Thus my dog is caused to exist by something
else. More strictly, my dog’s essence is
being given its existence. He has derivative
being. He is being “held in being.”
Now this thing that causes my dog to exist either itself is caused
to exist or it is not. If the former, we
begin a regress again. Thus, the only
thing that can terminate the infinite regress of derivative beings would be an
entity with underived being, something whose very essence is to exist. Therefore, something that does not need to
be given existence (a thing which has existence as part of its nature, a
Necessary Being) must exist to give everything else existence.
Now, apply this example to everything in the universe. Does any of
it exist on its own? No. As far as we can tell, nothing we have ever
encountered has necessary existence, that is, is the sort of thing that could
not fail to exist. As Hume puts it,
anything we can imagine as existing, we can imagine as not existing. We have no (empirical) commerce with
“necessary existing things” in the ordinary, mundane world. Nevertheless, Aquinas (et al.) argues that there
must be more to reality than what we witness via our senses. We know, via the cosmological argument, that
there must be a necessary thing.
So, not only did the universe have to have a first cause to get
started (according to horizontal arguments), it needs something to give it
existence right now (vertical arguments). The only thing that would not have to
be given existence is a thing that exists by its very nature. It is
existence. It is the pure act of Being (Actus
purus) as Aquinas puts it. This something would
exist, have no cause, have no beginning, have no end, have no limit, be outside
of time, and be infinite. That something is God! J
Essentially
Ordered and Accidentally Ordered Causal Series
This is what is referred to as an
essentially ordered causal series.
Accidentally
ordered causal series
In an accidentally ordered causal series,
all the causes are of the same kind.
Since all the causes are of “one kind” (essentially the same) the
multiple causes (even infinitely multiplied) should have the order of only one
cause, their multiplication of causes being accidental. For a sculptor, for
instance, no one particular hammer acts essentially after the action of the
previous hammer in the series and thus their order is accidental. An artisan
might go through a series of hammers (five, ten twenty, it really doesn’t
matter since they are all the same in causal efficacy) while sculpting a
statue. Thus the order of the series is
accidental. One hammer’s efficacy has
nothing to do with the efficacy of the succeeding or preceding hammer in the
series. The same can be said of human
generation according to Aquinas:
“Hence, it is not impossible for a man to
be generated by man to infinity.”
In an accidentally ordered series
A->B->C, the causal efficacy of A is not simultaneously required to
explain the causal efficacy of B. Note
Grandpa might have caused Dad who then causes Son, but Grandpa does not need to
be around for Dad to cause Son. Grandpa
may have died long ago (Sad, yea, I know.) before Dad creates Son.[12]
Likewise, the hammer #2 does not have to still be around for hammer #3 to do
its work.
Essentially
Ordered Causal Series
An essentially ordered causal series, by
contrast, is asymmetric, irreflexive, and wholly derivative. The subsequent
members in such series are not only caused by and ontologically dependent on
the preceding member(s), as in a transitive series, they also serve as causes only
insofar as they have been caused by and are effects of all the
preceding member(s). Because these
intermediate causes possess no intrinsic causal powers of their own, but rather
only by deriving them from the preceding cause(s), they need a first and
non-derivative cause to have any causal efficacy at all.
Something has a causal power
derivatively only if that causal power is, in fact, derived from something
else. Now if there were only
intermediate and derivative causes, then there would be no source from which
the causal powers of these intermediate causes could be derived, regardless of
whether there were a finite or an infinite number of these intermediate causes,
and thus no causation at all. A system
of all and only intermediate derivative accidentally ordered causes is
incoherent. If there were no first,
non-derivative cause, the intermediate causes would not actually be
causes. Thus there would be no effects.
But we do observe effects (and intermediate causes) For Aquinas, the hand (A) moves the stick (B)
which moves the stone (C). The stick has
no causal efficacy on its own, but only derives it from the hand. In this series each member depends upon its
predecessor for its very act of causing its successor. The order of the items listed here is
accidental rather than essential stemming from the fact that the causes are all
essentially of the same kind (derivative).
Aquinas likes to use the example of the “hand that moves the stick that
moves the stone.” But while this may be
the arrangement that actually occurs, there is no reason why it could not have
been the hand that moves the stone that then moves the stick. Indeed, it could also have been the stone
that moves the stick that moves the hand.
They all have derivative causality and it is an accident of history
which one derives its causal efficacy from which other one. But this
observation alone demonstrates the existence of an essentially different
cause. A non-derivative which stands in
a non-accidental order to the observed effects.
(e.g. an unmoved mover, uncaused cause, an uncreated creator/ unbegotten
begetter.) And this must exist simultaneously with the observed effects.
In a similar way. think of the way an
ultimate end bestows value on all the steps necessary to the achieving of that
end. I value going to the dentist. But the value of going to the dentist is bestowed
value. It is caused by the value of
having teeth in my old age. The value of
going to the dentist (today) depends simultaneously on the value of having
teeth in my old age… today. That is, the
bestower of value must exist simultaneously with and bestow value
simultaneously on the thing which receives value. In an essentially ordered series, the causal
activity of subsequent members of the series depends essentially on the simultaneous
causal activity of metaphysically prior member(s).
[1] Contemporary
philosopher William Lane Craig prefers this version. https://www.reasonablefaith.org/
[2] Note, Hume says precisely this in his criticism of the Ontological Argument.
[3] Hawking, Stephen A Brief History of Time 1988 p1
[4] I think Margaret Thatcher had something like this in mind
when she famously criticized the prospects of an increasingly
"service" oriented economy when she quipped, "not everybody can
take in everybody's washing—somebody has to make the clothes."
[5] I value going to the dentist every three months because it is a means to keeping my teeth when I get older. I value it so much in fact that I pay money for it. Of course, if there were a cheaper, more efficient alternative means to my end, I would be silly for continuing to value going to the dentist. But as far as I know, there is none. Sigh.
[6] William Lane Craig is a contemporary defender of this type of cosmological argument and this is his preferred way of presenting it.
[7] Craig argues that such a cause must be personal not mechanical, with an intellect, a will and great power. If the cause of the universe were simply an eternal, mechanistic sufficient cause of the universe, the universe too, though caused, would be eternal.
[8] The actual theory, not the T.V. show.
[9] See Richard Sorabji 2003 You Tube Video: https://www.youtube.com/watch?v=MyKXbAcSznk
[10] In fact, remember that this was crucial to our criticism of Anslem’s first formulation of his Ontological Argument.
[11] This is Aquinas’ distinction between an essence and esse. An essence or form may or may not be actualized. But just as any potential can only be made actual by something with actuality already, the bestower of existence must already possess existence already.
[12] Aquinas likes to talk about the causal efficacy of fathers and sons, etc. This can be a little confusing since we know now that fathers pass along the genes of their own father to their sons. This fact confuses Aquinas point. It does not undermine it however. The point remains, but would perhaps be better served with less confusing examples.